Synthesis of Oxidation-Resistant Cupronickel Nanowires for Transparent Conducting Nanowire Networks

Transcription

1 Supporting Information Synthesis of Oxidation-Resistant Cupronickel Nanowires for Transparent Conducting Nanowire Networks Aaron R. Rathmell, Minh Nguyen, Miaofang Chi, and Benjamin J. Wiley * Department of Chemistry, Duke University, 124 Science Drive, Box Durham, NC (USA) benjamin.wiley@duke.edu Microscopy Group, Oak Ridge National Laboratory, 1 Bethel Valley Road, Building 4515, MS 6064 Oak Ridge, TN 37831, (USA) KEYWORDS: nickel, copper, nanowires, transparent conductor

2 Materials and Methods: Materials: 35 wt % hydrazine in water (309400), polyvinylpyrrolidone MW = 10,000 (PVP10), ethyl acetate (270989), pentyl acetate (109584), and toluene (244511) were purchased from Sigma-Aldrich. Nitrocellulose was purchased from Scientific Polymer (712), and ethanol and isopropanol were purchased from VWR. Polyethylene Terephthalate was purchased from McMaster-Carr and ITO on PET was purchased from VisonTek Systems. Caution should be used when dealing with these chemicals as some are corrosive, flammable, or dangerous if ingested or put in contact with skin. Coating copper nanowires with nickel: The copper nanowires, donated from NanoForge, were synthesized in a manner similar to that previously reported. 1 The copper nanowires were stored at a copper nanowire concentration of 1.4 mg ml -1 in an aqueous solution containing polyvinylpyrrolidone (PVP, 1 wt %) and diethylhyroxylamine (1 wt %). Cupronickel nanowires were synthesized by adding the copper nanowire stock solution (0.732 ml) to a 20 ml scintillation vial containing a solution (1.32 ml) of PVP (2 wt %) in ethylene glycol, a given amount of a solution containing Ni(NO 3 ) 2 6H 2 O (0.1M) in water (157, 78.7, 39.3, or 15.7 µl for nanowires containing 54, 34, 20, 10% Ni, respectively), and hydrazine (132 µl, 35 wt%). This mixture was vortexed for 15 seconds and heated at 120 C for 10 minutes without any stirring. During the heating step, the dispersed copper nanowires became aggregated, floated to the top of the solution, and changed from a copper color to a dark copper or black color (depending on Ni concentration) due to Ni reducing onto the surface of the copper nanowires. After heating for 10 minutes the solution was removed with a pipette, and the nanowires were dispersed in a solution of PVP (1 wt %) and hydrazine (3 wt %). This solution was then centrifuged at 2000 rpm for 5 minutes, and the supernate was removed. The wires were then dispersed in a fresh wash solution (containing 3 wt % hydrazine and 1 wt % PVP) by vortexing for 30 seconds. These steps of centrifugation and redispersion were repeated once more with the same PVP/hydrazine solution, and then repeated two additional times with a solution that contained only hydrazine (3 wt %). Preparation of Transparent Electrodes: Transparent electrodes were made in a manner similar to that which was previously reported. 1 The cupronickel solution was put into a 1.5 ml centrifuge tube after the initial washing steps. This suspension was centrifuged at ~2000 rpm for 1 minute. The cupronickel nanowires were washed 3 times using a solution of hydrazine (3 wt

3 %) containing no PVP to ensure PVP was removed. After the PVP was removed, the cupronickel nanowires were washed with ethanol to remove the majority of the water and then washed once more with the ink formulation. The ink formulation was made by dissolving nitrocellulose (0.06 g) in acetone (2.94 g) and then adding ethanol (3 g), ethyl acetate (0.5 g), pentyl acetate (1 g), isopropanol (1 g), and toluene (1.7 g). After the cupronickel nanowires (~3 mg) were washed with the ink formulation, the ink formulation (0.3 ml) was added to the cupronickel nanowires, and this suspension was vortexed. If significant amounts of aggregates were present, the ink was briefly sonicated (up to 5 seconds) and centrifuged at a low speed (~500 rpm) so that a well-dispersed cupronickel nanowire ink could be pippetted off the solution. To prepare a transparent nanowire electrode, glass microscope slides were placed onto a clipboard to hold them down while the nanowire ink (25 µl) was pippetted in a line at the top of the slide. A Meyer rod (Gardco #13, 33.3 µm wet film thickness) was then quickly (< 1 second) pulled down over the nanowire ink by hand, spreading it across the glass into a thin, uniform film. Different densities of nanowires on the surface of the substrate were obtained by varying the concentration of the nanowires in the ink. The film was dry after approximately 60 seconds. To remove the film former and other organic material from the nanowire network, the films were cleaned in a plasma cleaner (Harrick Plasma PDC- 001) for 15 minutes in an atmosphere of 95% nitrogen and 5% hydrogen at a pressure of mtorr. The nanowire films were then heated at 175 C in a tube furnace for 30 minutes under a constant flow of hydrogen (600 ml min -1 ) to anneal the wires together and decrease the sheet resistance to below 200 Ω sq -1. The transmittance and sheet resistance of each nanowire film was measured using a UV/vis spectrophotometer (Cary 6000i) and a four-point probe (Signatone SP TBS). Each data point in Figure 2 is the average of 5 measurements. Each data point also shows error bars that indicate one standard deviation. SEM, TEM, and EDS Preparation: To prepare the samples for SEM (FEI XL30 SEM- FEG), a small chip of a silicon (Si) wafer (5 mm X 5 mm) was cut for each sample and placed on a piece of double sided tape in a Petri dish. Clean nanowires were dispersed in the hydrazine (3 wt %) solution with vortexing and sonication before 5 µl of the suspension was placed on a Si chip. The Petri dish was then covered with parafilm and nitrogen gas was gently blown into it to

4 dry the sample, creating a balloon out of the parafilm. After drying overnight, the nanowires were rinsed with a gentle flow of water (~150 ml min -1 ) for seconds and dried again. For TEM, a copper grid was used to hold the nanowires instead of a Si chip. The grid was placed on top of a Whatman filter, and 3 µl of the nanowire solution was pipetted onto the grid. The solution was absorbed into the filter paper underneath the grid, leaving the majority of the nanowires on the grid. The sample was then allowed to completely dry under a flow of nitrogen gas. The same sample preparation was done for the EDS samples except a nickel grid was used in place of a copper grid. Also, as soon as the sample was placed on the grid, the grid was placed in the TEM and a vacuum was drawn to completely remove the solution before it was placed in the column. The nanowires were analyzed using a FEI XL30 scanning electron microscope (SEM), a FEI Tecnai G² Twin transmission electron microscope (TEM), and a JEOL 2200FS Aberration- Corrected scanning transmission electron microscope (STEM) with an energy dispersive x-ray spectrometer (EDS). The diameters and lengths of the wires were determined by comparing the pixel diameter/length of the wires with the pixel length of the scale bar. Concentration of cupronickel nanowires: To measure the concentration of the cupronickel nanowires, a set volume of the solution (usually 1 ml) was dissolved in concentrated nitric acid (1 ml). The dissolved nickel and copper was then diluted to a set volume (usually 10 ml). An atomic absorption spectrometer (AAS, Perkin Elmer 3100) was then used to measure the concentration. Number Density Calculation: This section describes how we made films of copper and cupronickel nanowires with a desired number density. First we found the number of nanowires in a given volume of solution (N S ) by equation SI-1, where C Cu is the concentration of copper in the N s = C CuV T ρ Cu V Cu (SI-1) solution (as measured by AAS), V T is the total volume of the solution, ρ Cu is the bulk density of copper, and V Cu is the average volume of the copper nanowires. Next, we filtered out a given amount of solution to achieve a desired nanowire number density on the filter. To match the number densities of the cupronickel nanowires to those of the copper nanowires, we converted the volume of nanowire solution necessary to obtain the same number density (N) of nanowires on the filter with equation SI-2, where V CuNi, is the average volume of one cupronickel nanowire,

5 V T = N(V CuNi V Cu )ρ Ni C Ni (SI-2) ρ Ni is the bulk density of nickel, and C Ni is the concentration of nickel in the solution (as measured by AAS). Set volumes of the copper and cupronickel nanowire solutions were diluted in 200 ml of water and filtered onto 0.6 µm Isopore membrane filters (Millipore 0.6 µm, DTTP04700). As soon as the solutions were through the membrane, the filtrate was immediately put into contact, by hand, with polydimethylsiloxane (PDMS). The membrane filter was gently pressed against the PDMS to ensure transfer of the nanowires before the membrane was peeled away, leaving behind the nanowire film. The transmittance of the films was measured with a UV/vis spectrophotometer (Cary 6000i). The baseline on the UV/vis spectrophotometer was corrected using a piece of PDMS, free of any nanowires. Testing the Flexibility of Nanowire Films: To test the flexibility of the copper-nickel nanowire films, we attached one end of the film to a tabletop and the other end to a ruler which was on top of a spring. The set-up was designed so that the initial radius of curvature was 10 mm (no force applied to the ruler) and after a downward force was applied to the ruler the final radius of curvature was 2.5 mm. The ruler was then allowed to spring back to the starting position and the entire cycle was counted as 1. The sheet resistance was measured every 200 cycles. REFERENCES 1. Rathmell, A. R.; Wiley, B. J. Adv. Mater. 2011, 23, (41),

6 Figure SI-1: Plot of sheet resistance vs. the inverse of the effective nanowire thickness. The slope gives the resistivity of the nanowire networks. Error bars show one standard deviation for five measurements.

7 SI-2: Plot of ΔR/R 0 vs. time for copper nanowire films.

8 SI-3: Plot of ΔR/R 0 vs. time for silver nanowire films.

9 SI-4: Plot of ΔR/R 0 vs. time for 4:1 Cu:Ni nanowire films.

10 SI-5: Plot of ΔR/R 0 vs. time for 1:1 Cu:Ni nanowire films.

11 SI-6: Plot of sheet resistance vs. number of bends for copper nanowires, 4:1 Cu:Ni nanowires, 1:1 Cu:Ni nanowires, and ITO films on PET. Error bars show one standard deviation for five measurements.