Home

Search

Collections

Journals

About

Contact us

My IOPscience

Rapid synthesis of flexible conductive polymer nanocomposite films

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2015 Nanotechnology 26 125601 (http://iopscience.iop.org/0957-4484/26/12/125601) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 169.230.243.252 This content was downloaded on 12/04/2015 at 14:19

Please note that terms and conditions apply.

Nanotechnology Nanotechnology 26 (2015) 125601 (6pp)

doi:10.1088/0957-4484/26/12/125601

Rapid synthesis of flexible conductive polymer nanocomposite films C O Blattmann1, G A Sotiriou1,2 and S E Pratsinis1 1

Particle Technology Laboratory, Institute of Process Engineering, Department of Mechanical and Process Engineering, ETH Zürich, Sonneggstrasse 3, 8092 Zürich, Switzerland 2 Department of Environmental Health, School of Public Health, Harvard University, 665 Huntington Ave., Boston, MA 02115, USA E-mail: [email protected] Received 11 December 2014, revised 21 January 2015 Accepted for publication 9 February 2015 Published 4 March 2015 Abstract

Polymer nanocomposite films with nanoparticle-specific properties are sought out in novel functional materials and miniaturized devices for electronic and biomedical applications. Sensors, capacitors, actuators, displays, circuit boards, solar cells, electromagnetic shields and medical electrodes rely on flexible, electrically conductive layers or films. Scalable synthesis of such nanocomposite films, however, remains a challenge. Here, flame aerosol deposition of metallic nanosliver onto bare or polymer-coated glass substrates followed by polymer spincoating on them leads to rapid synthesis of flexible, free-standing, electrically conductive nanocomposite films. Their electrical conductivity is determined during their preparation and depends on substrate composition and nanosilver deposition duration. Accordingly, thin (99%, Sigma-Aldrich) dissolved in a 1:1 mixture of 2-ethylhexanoic acid (>99%, Sigma-Aldrich) and acetonitrile (>99.9%, Sigma-Aldrich) was utilized. Microscope glass slides (micro slides plain, 25 × 75 × 1 mm, Corning) were cut, cleaned with ethanol and dried with Kim Wipes before being used as substrate. Substrates and particle films were coated with PMMA with thickness L by spin-coating (WS-650MZ-23NPP/LITE, Laurell Technologies Incorporated) a PMMA (Mw ∼ 120 000 g mol−1, Aldrich) in anisole (>99%, SigmaAldrich) solution (2–10 wt%) at 1000–2000 rpm and subsequently drying on a hot plate (VWR VHP-C7) set to 180 °C for 2–5 min. For nanocomposite film synthesis, spin-coating settings of the PMMA overlayer were identical to that used for the base layer. Exception is L = 0 μm, where the PMMA overlayer spin-coating settings correspond to those for L = 0.16 μm. Silver nanoparticles were deposited onto the substrate by clamping it to a water-cooled fixture [31]. The deposition area is centred facing the flame at 15 cm height above the burner. Deposition duration (t) was precisely timed by use of a shield in front of the substrate prior to deposition. The substrate temperature was measured with a Fluke Thermal Imager Ti110 directed at the substrate surface, while exposing it to an Ag-free precursor solution at otherwise identical conditions. Particles were collected from a glass fibre filter (GF 6 257, Hahnemühle FineArt GmbH) downstream of the glass substrate (figure 1(a)) by vacuum pump (Busch Mink MM 1104 BV) in order to characterize by x-ray diffraction (XRD, Bruker axs D8 advance) and BET nitrogen adsorption (Micromeritics TriStar) at 77 K. Prior to BET analysis, the particles were placed in a vacuum oven at room temperature (RT) and 20 mbar for >8 h and thereafter flushed with nitrogen gas (PanGas 5.0) at RT for 1 h (Micromeritics Flow Prep 060). This procedure was chosen to prevent Ag particles from sintering at increased temperatures [32]. Particles were visualized by transmission electron microscopy (TEM, Tecnai F20, FEI, FEG, 200 kV). Particle films and Ag–PMMA nanocomposite films were characterized by XRD and scanning electron microscopy (SEM, Hitachi S-4800). Image evaluations (e.g. size, dimensions) were conducted in ImageJ (v1.42q). Nancomposites were additionally analysed by four point resistance measurements (Signatone 302 resistivity probe stand (tungsten tips, 1.02 mm spacing, 0.13 mm tip radius, 85 g pressure) with Keithley 2400) to determine their sheet resistance. Ag–PMMA nanocomposite films are released readily from the glass substrate in a pure water bath without the employment of a sacrificial layer. The low adhesion of PMMA on glass is attributed to the high relative humidity [33] inducing the PMMA to expand. This is particularly beneficial for medical applications since the contact with toxic chemicals (e.g. 1-methyl-2-pyrrolidinone (NMP)-based solvents for sacrificial layer removal) is avoided. Cyclic bending was conducted by clipping a released polymer nanocomposite film onto two sides of a

Figure 1. (a) The FSP set-up (I) for synthesis and deposition of silver nanoparticles which are subsequently embedded by spin-coating (II) to form polymer nanocomposite films. (b) The Ag crystal size of nanoparticle films on bare (L = 0 μm, green squares) and PMMAcoated (L = 0.16 μm, red triangles) glass as a function of deposition duration t. The crystal size increases linearly at 31 nm min−1 independent of PMMA coating thickness L.

and/or equipment (e.g. clean room facilities, high vacuum). That way homogeneous Ag nanoparticle layers are made with controlled structure, cohesion and adhesion onto selected substrates. In fact such flame aerosol deposition processes are quite practical as they are routinely used in making refractory films (e.g. thermal spraying by plasma-torch deposition [25]) and most notably for manufacturing optical fibres [26]. Here we show, for the first time to our knowledge, how such technology can be used for inexpensive synthesis of flexible, free-standing and conductive nanocomposite films by spincoating a polymer onto these silver nanoparticle layers and releasing them from the substrate. The compliant behaviour of such flame-made nanoparticle aggregates/agglomerates (i.e. reversible twisting, folding–unfolding [27]) is expected to render flexible/stretchable nanocomposites without fracturing or buckling of the filler as observed with carbon nanotubes [7], silver nanowires [28] and metal films [29].

2. Experimental Silver nanoparticles were made by flame-spray pyrolysis (FSP) [30]. Here 3 ml min−1 of combustible liquid precursor is injected through a capillary and dispersed by 6 L min−1 O2 (>99.5%, PanGas) and ignited by a concentric methane pilot flame (1.5 L min−1 CH4 (>99.5% PanGas), 3.2 L min−1 O2). 2

C O Blattmann et al

Nanotechnology 26 (2015) 125601

programmable motor (Lambda Vit-Fit). The Ag–PMMA nanocomposite film was mounted on a transparent plastic foil (Freshstar food wrap, Alu-Vertriebsstelle Kreuzlingen) for stability. The electrical resistance during repeated bending (42 mHz) was recorded with a multi-metre (Tektronix DMM4050, 105 measurements min−1).

t = 30

t = 60

L=0

3. Results and discussion Metallic Ag nanoparticles are generated rapidly by FSP and directly deposited onto a back-cooled, bare or polymer-coated (e.g. PMMA) glass substrate just above that flame [34] (figure 1(a)). The substrate front temperature, TF, reaches steady-state at 135, 138 and 146 ± 5 °C depending on its polymer coating thickness (L = 0, 0.16, 0.7 μm, respectively) for t > 30 s at 15 cm above the burner. Figure 1(b) shows the crystal size, dXRD, of such directly deposited metallic Ag (for detailed XRD patterns figure S1 in SI) onto bare (L = 0 μm, green squares) and PMMA-coated substrates (L = 0.16 μm, red triangles) as a function of nanoparticle deposition duration t. The crystal size increases with t at about 31 nm min−1, independent of L. Silver nanoparticle growth on the substrate occurs by continuous sintering or coalescence of freshly depositing nanoparticles with the ones already there. Silver nanoparticles bypassing the deposition substrate are collected on a filter at ∼60 cm above the burner at 80–90 °C. Their dXRD is 18 nm (SI, figure S2(a)), as the lower filter temperature prevents particle growth, in contrast to the higher TF at the deposition substrate. TEM images of nanoparticles collected at the filter reveal high degree of aggregation (SI, figure S2(b)) consistent with their low specific surface area (4.8 m2 g−1). Figure 2 shows top view (a), (c), (e), (g) and crosssection (b), (d), (f), (h) images of Ag nanoparticle films deposited for t = 30 (a)–(d) and 60 s (e)–(h) on bare (L = 0 μm, (a), (b), (e), (f)) and PMMA-coated (L = 0.16 μm, (c), (d), (g), (h)) substrates. The top view images show that particle size increases with increasing t consistent with crystal growth (figure 1(b)). On bare glass (L = 0 μm), mostly separate and rather columnar nanoparticles grow having an average diameter of 52 and 75 nm for t = 30 (a) and 60 s (e), respectively. These sizes are larger than the corresponding Ag crystal sizes of 40 and 59 nm indicating polycrystallinity that is attributed to slower crystal growth than coalescence or sintering of Ag nanoparticles. In contrast, such nanoparticles deposited on PMMAcoated substrates (figures 2(c) and (g) and SI figure S3) exhibit significant sinter-necking resulting in denser and interconnected particle networks with increasing t and L that can be beneficial for electron transport [35]. This is confirmed by electrical resistance measurements across the film width (1 cm). Nanosilver films on bare glass exhibit a resistance above 200 MΩ i.e. non-conductive) while on PMMA-coated glass (L = 0.16 μm) their resistance drops to less than 10 Ω (for t > 120 s, SI figure S3(g)) as will be shown in figure 3. Cross-section images (figures 2(b),(d),(f),(h)) show that a silver nanoparticle monolayer is always formed regardless of

L = 0.16

Figure 2. Top view (a),(c),(e),(g) and side view (b),(d),(f),(h) SEM

images of flame-made Ag nanoparticles deposited for 30 (a)–(d) and 60 s (e)–(h) onto bare (L = 0 μm, (a),(b),(e),(f) and PMMA-coated (L = 0.16 μm, (c),(d),(g),(h)) glass substrates. Increasing t increases the Ag particle size. In the presence of PMMA, the sinter-neck formation is greatly enhanced (white circle in (c)). The poor adhesion of Ag nanoparticles can be inferred from loose ones adhering to the fractured glass surface (white circles in (b),(f)). The scale bars for all top and side view images are identical at 0.5 and 1 μm, respectively. An extended overview for other t and L is shown in the SI (figure S3).

substrate surface composition and deposition duration. Its thickness increases with time at about 90 nm min−1. Nanocomposite Ag–PMMA films are formed by spincoating polymer on the above Ag nanoparticle films [34]. Figure 3(a) shows the sheet resistance of such nanocomposites as a function of t for various substrate PMMA thicknesses L. Composites made from nanoparticles deposited on bare glass (L = 0 μm, green squares) exhibit no conductivity (>109 Ω sq−1) at all deposition durations. The presence, however, of a thin PMMA layer on the substrate during deposition strongly influences nanocomposite conductivity. By increasing the layer thickness L, the nanocomposites become more conductive (lower sheet resistance) at identical duration of Ag deposition or loading. For example, at t = 60 s the nanocomposites become conductive with increasing the initial L on the glass substrate (0.06: purple diamonds, 0.16: red triangles, 0.7: blue circles and 1.4 μm: brown inverse triangles). In fact, nanocomposites made with L = 1.4 μm are highly conductive with sheet resistance of only 1 Ω sq−1. Nevertheless, all nanocomposites made on PMMA-coated glass reach similarly low sheet resistance after sufficient deposition duration (e.g. t = 120 s for L > 0.16 μm). The leveling-off of the sheet resistance at ∼1 Ω sq−1 is in agreement with percolation theory [16], and represents the 3

C O Blattmann et al

Nanotechnology 26 (2015) 125601

Figure 3. (a) The Ag–PMMA nanocomposite film sheet resistance as a function of deposition duration t onto glass substrates with PMMA layer thickness L = 0 (green squares), 0.06 (purple diamonds), 0.16 (red triangles), 0.7 (blue circles) and 1.4 μm (brown inverse triangles). The sheet resistance decreases with increasing L. For bare substrates (L = 0 μm), the sheet resistance remains constant above 1 GΩ sq−1 due to insufficient Ag nanoparticle sintering and network formation. For L = 1.4 μm, a low sheet resistance (∼1 Ω sq−1) is already obtained for t = 60 s. SEM cross-section images of Ag– PMMA nanocomposites (t = 180 s) are shown for L = 0.06 (b), 0.16 (c) and 0.7 μm (d). The scale bar is identical in all images.

Figure 4. A nanocomposite Ag–PMMA film (L = 0.16 μm, t = 180 s)

is released in water (a) and mounted onto a moveable stage connected to an electric circuit (b). The passing current through the bent nanocomposite can be visualized by the illuminated green LEDs. Electrical conductivity and flexibility of nanocomposite Ag– PMMA nanocomposite films is demonstrated by measuring the nearly constant resistance (2.5 Ω) as a function of bending cycle (c) for concave (red line and image (d)) and convex (blue line and image (e)) bending motions (film bending radius: 1.6 mm).

formation of an inter-connected network (e.g. electrical path). Sinter-necking enables the creation of such a network on PMMA-coated substrates (figures 2(c),(d),(g),(h)). In contrast, nanocomposites made on bare glass (L = 0 μm, figure 3, green squares) do not reveal a measureable decay in sheet resistance (t ⩽ 180 s), due to poorly connected [36] silver nanoparticles (figures 2(a) and (e) and SI figures 3(a)–(d)). Cross-section images of nanocomposite Ag–PMMA films (t = 180 s) still on the glass substrate are shown also in figure 3 for L = 0.06 (b), 0.16 (c) and 0.7 μm (d) with a spincoated PMMA overlayer having the corresponding thickness. Thus embedded or sandwiched Ag nanoparticles remain homogeneously distributed within that PMMA matrix without any redistribution upon spin-coating (SI, figure S4). The thickness of the nanocomposite can be selected by the initial

PMMA and/or spin-coated PMMA overlayer thickness L. So the nanocomposite film made with L = 0.06 μm (figure 3(b)) is the thinnest one (∼0.24 μm) here and exhibits the highest surface roughness because the PMMA overlayer barely covers the Ag nanoparticles. The surface roughness, however, can be reduced easily by increasing the overlayer thickness. More specifically, the surface of nanocomposites made with L = 0.16 or 0.7 μm (figure 3(d)) are quite smooth without increasing sheet resistance (figure 3(a)). Their high conductivity is determined from the combined sheet resistance and film thickness. For example, a 425 nm thick nanocomposite (L = 0.16 μm, t = 180 s, figure 3(c)) has a conductivity of 5 × 104 S cm−1, which is equivalent to bulk metal. 4

C O Blattmann et al

Nanotechnology 26 (2015) 125601

Figure 4 demonstrates the combined electrical and mechanical functionality of an Ag–PMMA nanocomposite film (L = 0.16 μm, t = 180 s) after being released from the substrate. Figure 4(a) depicts such a film floating on water. This film is rather fragile due to its ∼425 nm thickness (figure 3(c) for cross-section) but can be removed easily from water by mounting it onto a thin (∼20 μm) transparent foil (barely visible in figure 4(b)). This mounted film is anchored to an electrical circuit onto a two-sided moveable stage. Figure 4(b) shows the connected circuit passing through two green LEDs (background) that are lit upon turning it on. The electric current through the nanocomposite film is demonstrated by the constantly lit LEDs during repeated bending for multiple cycles (for animation, see SI). Such bending (>900 cycles [37]) of the nanocomposite is conducted down to a bent film radius of 1.6 mm while monitoring its electric resistance (figure 4(c)). This is quite low (∼2.5 Ω, without accounting for contact resistance losses) and in agreement with its sheet resistance (figure 3(a), red triangle at t = 180 s). First the film was repetitively bent in a concave manner (figure 4(c), red line) for >900 cycles, corresponding to the image sequence shown in figure 4(d). This was followed by convex bending (blue line, figure 4(c), >900 cycles) depicted in figure 4(e). The resistance increased only by a fraction of an Ohm throughout the whole test, whereas the greatest change (

Rapid synthesis of flexible conductive polymer nanocomposite films.

Polymer nanocomposite films with nanoparticle-specific properties are sought out in novel functional materials and miniaturized devices for electronic...
1MB Sizes 0 Downloads 10 Views