www.acsnano.org
Laser-Induced Graphene by Multiple Lasing: Toward Electronics on Cloth, Paper, and Food Yieu Chyan,†,# Ruquan Ye,†,# Yilun Li,† Swatantra Pratap Singh,⊥ Christopher J. Arnusch,*,⊥ and James M. Tour*,†,‡,§ †
Department of Chemistry, ‡Smalley-Curl Institute and The NanoCarbon Center, and §Department of Materials Science and NanoEngineering, Rice University, 6100 Main Street, Houston, Texas 77005, United States ⊥ Department of Desalination and Water Treatment, Zuckerberg Institute for Water Research, The Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede-Boqer Campus, Beersheba 84990, Israel S Supporting Information *
ABSTRACT: A simple and facile method for obtaining patterned graphene under ambient conditions on the surface of diverse materials ranging from renewable precursors such as food, cloth, paper, and cardboard to high-performance polymers like Kevlar or even on natural coal would be highly desirable. Here, we report a method of using multiple pulsedlaser scribing to convert a wide range of substrates into laser-induced graphene (LIG). With the increased versatility of the multiple lase process, highly conductive patterns can be achieved on the surface of a diverse number of substrates in ambient atmosphere. The use of a defocus method results in multiple lases in a single pass of the laser, further simplifying the procedure. This method can be implemented without increasing processing times when compared with laser induction of graphene on polyimide (Kapton) substrates as previously reported. In fact, any carbon precursor that can be converted into amorphous carbon can be converted into graphene using this multiple lase method. This may be a generally applicable technique for forming graphene on diverse substrates in applications such as flexible or even biodegradable and edible electronics. KEYWORDS: porous graphene, laser induction, flexible graphene, multiple lasing, edible electronics
P
we disclose the necessary parameters to form LIG, without the need for inert atmosphere, on a wide variety of carbon precursors including polymers, natural materials, food, and even nonpolymeric materials such as activated carbon or anthracite coal by means of multiple lasing. Multiple lasing allows enhancement of the electrical properties by improving the quality of the LIG obtained. For example, poly(ether imide) (PEI) was previously found to produce LIG that was substantially less conductive than that formed from PI.10 We show here that with multiple lasing, however, highly conductive PEI-based LIG could be obtained
orous 3D graphene-based nanomaterials demonstrate promise for a wide variety of applications due to their unique physical and chemical properties. A straightforward method of synthesizing laser-induced graphene (LIG) from polyimide (PI, Kapton) has been previously reported and applied toward energy storage devices such as supercapcitors,1 electrocatalysts for water splitting,2 piezoelectric strain gauges,3 antibiofouling,4 and for electrochemical biosensors,5 photodetectors,6 and sensing and producing sound.7 The laser scriber is of the type that is routinely used in machine shops, so these units are readily accessed. A limitation of this approach has been the reliance on polyimide as the polymer precursor for the formation of LIG. We recently disclosed a method to make LIG on wood using an inert atmosphere chamber8 as well as for polysulfones using a modified lasing strategy.9 However, here, © XXXX American Chemical Society
Received: December 1, 2017 Accepted: February 8, 2018
A
DOI: 10.1021/acsnano.7b08539 ACS Nano XXXX, XXX, XXX−XXX
Article
Cite This: ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano using both multiple pass and defocus methods. Moreover, multiple lasing allows for the formation of LIG on naturally occurring substrates such as cloth, paper, potato skins, coconut shells, and cork. These materials are inexpensive, abundant, and biodegradable, unlike many of the polymer precursors previously found to afford LIG. The ability to form LIG on these substrates would potentially allow applications such as flexible micro-supercapacitors on renewable materials or even edible electronics, after a thorough toxicity study.
RESULTS AND DISCUSSION Lignin-containing carbon precursors were previously reported to be converted into LIG by lasing under controlled atmospheres such as an inert or reducing gas.8 It was determined that wood with higher lignin content yielded better LIG, suggesting that the cross-linked aromatic structure was important to the formation of graphene. Typically, oven-dried wood consists of approximately 18−30% lignin.11 Other natural materials such as cork, coconut shells, and potato skins have even higher lignin content compared with wood (∼25, 30, and 36% lignin, respectively).12−14 Due to the higher lignin content, these materials were found to be directly converted into LIG upon multiple lasing with a pulsed laser even when processed under ambient atmosphere as compared with wood which burns or ablates under similar conditions. This substantially simplifies the production procedure for LIG since a controlled atmosphere reaction vessel equipped with a laser-transparent ZnSe window is not required. The requirement for a ZnSe window imposes a practical limitation on the size of substrates that can be converted into LIG. Figure 1a shows a coconut surface converted into LIG in the shape of the letter “R”. The surface of a coconut is converted into a 3D porous graphene structure by irradiation with a 10.6 μm CO2 laser under ambient atmosphere (see Experimental Section). As shown, the lignin-containing precursor material can be easily patterned with LIG by computer-controlled laser rastering over the surface. Areas exposed to the laser light are converted by a photothermal process to graphene, whereas areas not exposed to the laser remain unchanged.10 The presence of graphene is evidenced by the Raman spectrum depicted in Figure 1b. The 2D band along with the G band clearly indicates graphene material, whereas the D band is suggestive of bent graphene sheets or other defects sites.15,16 Figure 1c shows the transmission electron microscopy (TEM) image of a coconut-derived LIG flake. Higher-resolution TEM images show that the flake consists of few-layer graphene that reveals clear graphene fringes with the characteristic 0.34 nm dspacing (Figure 1d). The coconut shell was exposed to a 75 W laser with power setting ranging from 5 to 10% power. It was found that a single exposure of the coconut shell using 5% power resulted in the formation of amorphous carbon (Figure S1). Repeated lasing of the same portion of the substrate (up to 5×) resulted the graphene observed in Figure 1. Merely using a higher laser power setting often does not afford more LIG but rather leads to ablation or thermal damage of the surrounding substrate. Similar results were obtained for cork and potato skins (Figure 1e,f) as determined by Raman microscopy (Figure S2). For lasing on coconuts or potatoes, it is important to locate a reasonably flat area to minimize focus variations for laser systems that are unable to automatically maintain a set distance from a nonuniform surface. As a proof of concept, a micro-supercapacitor was fabricated on a coconut by the multiple lase method and then applying
Figure 1. LIG on diverse substrates. (a) Picture of LIG patterned into an “R” on a coconut (2 cm tall). (b) Raman spectrum of coconut-derived LIG lased two times at 10% speed and 5% power. (c) Low-resolution TEM of coconut LIG after 5 lases. The scale bar is 50 nm. (d) High-resolution TEM coconut LIG (10% speed, 5% power, 5×) showing the characteristic 0.34 nm d-spacing of graphene. The scale bar is 5 nm. (e) LIG on cork in the shape of an owl (height 30 mm). (f) Potato scribed with the laser to form LIG in the “R” pattern (2 cm tall).
fabrication methods that we described previously.1 Twice-lased coconut was compared with single-lased PI with the same overall fluence, and the same dimensions and the coconutderived material were found to have a higher areal capacitance (Figure 2). This suggests that in some cases the mechanism of LIG formation involves the conversion of a carbon precursor first to amorphous carbon followed by a conversion to graphene upon subsequent lasing. As shown in Figure S3, amorphous carbon absorbs infrared light strongly in a range from ∼700 to 1200 cm−1. Amorphous carbon can be considered a heterogeneous material consisting of sp2 carbon clusters that are embedded within a sp3 carbon matrix.17 CO2 lasers output a band centered at 10.6 μm but ranges between 927 and 951 cm−1.18 These frequencies are absorbed by the C−C bonds present in the precursor substrate materials that are not present in LIG (Figure S3). As such, it is likely that the substrate is first photothermally converted to amorphous carbon. Subsequent exposures of the amorphous carbon then effect the transformation of the amorphous carbon to graphene. This selective absorption of infrared light by amorphous carbon is one reason why mere thermal treatments or irradiation with other wavelengths upon carbon did not yield LIG. For example, B
DOI: 10.1021/acsnano.7b08539 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 2. Microsupercapacitor on coconut. (a) As proof of concept, a LIG-based supercapacitor was fabricated on the surface of a coconut (scale bar is 1 cm). (b) Cyclic voltammogram at 10 mV/s scan rate shows a higher areal capacitance for the twice-lased coconut microsupercapacitor compared with a single-lased polyimide despite the same total fluence of laser exposure.
lasing of polyimide film with an ultraviolet laser (275−363 nm) was previously reported to only result in amorphous or glassy carbon despite repeated lasing (up to 35 times).19,20 Clearly, the wavelength of the laser irradiation matters for obtaining graphene by multiple lasing. More recently, polyimide was ablated with a 308 nm XeCl excimer laser, and the carbon material was characterized after 200−800 pulses but no graphene-based 2D Raman peaks were detected.21 By contrast, only 3−5 passes of a rastered CO2 laser yields LIG from a wide variety of substrates. Hence, the wavelength of the laser irradiation as well as the number of exposures is important to the formation of LIG. To confirm this hypothesis, activated carbon was irradiated with multiple exposures to a pulsed 75 W 10.6 μm laser at 5% power setting at a 15 cm/s scan rate. Both TEM and Raman characterization show that amorphous carbon powder can be converted to LIG by multiple lasing (Figure 3). Raman shows the characteristic G, D, and 2D peaks but also a D+D′ peak at ∼2800 cm−1 characteristic of disorder, which may be the result
of vacancies, interstitial atoms, or substitutional atoms.15 This result demonstrates that the multiple lasing process for making graphene is generally applicable to any material that can first be converted into a layer of amorphous carbon. Thus, multiple lasing can be used to directly obtain LIG from many substrates that can first be thermally, photochemically, or chemically carbonized. Moreover, inexpensive carbon sources such as activated carbon and coal can now be used in the preparation of graphene, which might have implications in commercial applications for LIG from very inexpensive carbon sources. Two methods were used to obtain multiple lases of a substrate. The first and obvious method involves multiple passes of the rastered laser beam. At 5% power of the 75 W laser, the spot size of the laser is ∼175 μm in diameter. Given that the samples were lased using the 1000 dots per inch (DPI, a setting on commercial laser systems; 1 in. = 2.54 cm) raster density, multiple exposures will occur naturally with the overlap of the laser spots. For a 175 μm diameter spot size, each location at which the substrate is exposed has ∼37 overlapping laser spots. Multiple passes of the laser would result in an additional 37 lases per pass. The fact that a range of focus settings will yield LIG means that a substrate that is sufficiently thin can be converted entirely into LIG. A second method for obtaining additional exposures was developed involving increasing the spot size of the laser while keeping the density of the dots consistent. This was achieved by defocusing the laser to take advantage of the fact that the shape of the focused laser beam is conical. By altering the z-axis distance from the focal plane, different spot sizes can be obtained (Figure 4). Lowering the substrate by ∼1.0 mm relative to the laser focal point results in the increase of the spot size from 175 to 300 μm in diameter. This results in 3 times more lases in any given location of the substrate being lased since the area of each spot increases, but the density of laser spots remains constant. Figure 4b shows the effective number of at focus lases as a function of the z-axis defocus. Images of the spots at various defocus levels can be found in the Supporting Information (Figure S4). The advantage of this method is an increase in processing speed since each spot can be lased many additional times in one pass of the laser, or a combination of defocus and multiple laser passes can be used to lase a material the desired number of times. This technique was applied to ULTEM PEI, which had previously been found to perform much more poorly than PI as an LIG precursor substrate.10 The PEI was lased at various defocus levels ranging from 0 (at focal plane) to ∼2 mm
Figure 3. Activated carbon and activated carbon LIG. (a) TEM of activated carbon lased one time at 5% speed 5% power. Scale bar is 10 μm. (b) Activated carbon lased five times at 5% speed and 5% power, showing the presence of few layer graphene. Scale bar is 10 μm. (c) Raman spectra of activated carbon and activated carbonderived LIG lased once and lased five times at 5% speed and 5% power with ∼0.25 mm defocus. C
DOI: 10.1021/acsnano.7b08539 ACS Nano XXXX, XXX, XXX−XXX
Article
ACS Nano
Figure 5. PEI-derived LIG (PEI LIG) sheet resistance and Raman spectra as a function of defocus. (a) Raman spectra of PEI LIG at various defocus values ranging from 0.0 mm to 1.78 mm (10% speed, 5% power). (b) Summary of ID/IG and I2D/IG of PEI LIG and (c) fwhm of the G peak for various amounts of defocus. (d) Sheet resistance, measured using a four-point probe, of PEI LIG at various levels of defocus.
Figure 4. Spot size vs defocus. (a) Diagram of defocusing on the substrate to increase the laser spot size. At low DPI (
