Direct Synthesis of Colorful Single-Walled Carbon Nanotube Thin Films

In floating catalyst chemical vapor deposition (FC-CVD), tuning chirality distribution and obtaining narrow chirality distribution of single-walled carbon nanotubes (SWCNTs) is challenging. Herein, by introducing various amount of CO2 in FC-CVD using CO as a carbon source, we have succeeded in directly synthesizing SWCNT films with tunable chirality distribution as well as tunable colors. In particular, with 0.25 and 0.37 volume percent of CO2, the SWCNT films display green and brown colors, respectively. We ascribed various colors to suitable diameter and narrow chirality distribution of SWCNTs. Additionally, by optimizing reactor temperature, we achieved much narrower (n,m) distribution clustered around (11,9) with extremely narrow diameter range (>98% between 1.2 and 1.5 nm). We propose that CO2 may affect CO disproportionation and nucleation modes of SWCNTs, resulting in SWCNTs’ various diameter ranges. Our work could provide a new route for high-yield and direct synthesis of SWCNTs with narrow chirality distribution and offer potential applications in electronics, such as touch sensors or transistors.

Typically, a CO flow of 50 ccm (Chemical Carbon Monoxide, 99 vol %, AGA) was passed through the cartridge where the ferrocene was stored. Then the ferrocene-containing gas was injected through a water-cooled injector probe which is maintained at constant temperature of 24 °C. Another pure CO flow of 250 ccm were introduced from the main inlet, and 100 ccm from bypass inlet to avoid the turbulence. The ferrocene vapor immediately decomposed into iron vapor after coming out from the injector probe to the tube furnace (Entech, Sweden), followed by the nucleation to iron nanoparticles in the quartz tube with a maximum temperature of 850 °C or 880 °C, subsequently the iron nanoparticles will catalyze the growth of SWCNTs inside the reactor with a laminar flow. CO 2 was used to tune growth of SWCNTs, in this work, we used various additional CO 2 flow rate (controlled by mass flow rate, Aalborg System, USA) of 0.0, 1.0, 1.5, 2.0 and 5.0 ccm, corresponding to volumetric fraction of 0, 0.25, 0.37, 0.50, and 1.23 vol%, respectively. Finally the SWCNTs network can be collected downstream of reactor using a membrane filter with the help of sucking vacuum. SWCNT films can be directly transferred on quartz slide without any purification for absorption measurement.

S2
Photos of SWCNT thin films were taken under white light, the camera used here is Nikon, model D810, F-stop is f/3.3, exposure time is 1/1250 sec, ISO is 250 and focal length is 105 mm.

Absorption measurement
The absorption spectra of SWCNT films were conducted by Perkin-Elmer Lambda 950 UV-vis-NIR spectrometer with wavelength ranging from 175 to 3300 nm. In detail, the SWCNT film were transferred from filter on quartz substrate, and another empty quartz was used as the reference to avoid the influence of substrate contribution. The wavelength we used ranges from 250 to 2600 nm, and the beam was tuned to a suitable size so that it can fit the sample size.

Raman spectroscopy
The Raman spectroscopy was conducted by Horiba LabRAM HR 800 with excitation wavelength of 488, 514 and 633 nm.

Transmission electron microscopy (TEM) measurement
To collect samples for TEM measurement, the TEM grid was put on the filter to collect samples from reactor for 20 seconds. Then a JEOL-2200FS double aberration-corrected microscope was operated under 80 kV acceleration voltage for the electron diffraction, and the imaging mode was done under 200 kV acceleration voltage.

Sheet resistance measurement
The pristine SWCNT thin films with different transmittance were obtained by harvesting samples for different time. Then, the pristine thin films were transferred from membrane filter to quartz substrate by pressing. After that, we measured the transmittance at 550 nm of thin films.
On top of that, the sheet resistance was analyzed by HP 3485A multimeter 4-point probe system.
Later on, the pristine films were doped by AuCl 3 (0.016 mol/L), then the sheet resistance of doped thin films was measured again. Figure S1. A schematic of FC-CVD reactor for synthesis of SWCNTs. Figure S2. (Scanning) TEM and energy dispersive spectroscopy (EDS) analysis were employed to disclose the morphology and metal content of our samples. Typical low-magnification TEM images (a-c) and corresponding scanning TEM dark-field images (d-f) for SWCNTs with CO 2 of 0, 0.25 and 0.50 vol%, respectively. It is apparent that all three SWCNT samples appear to be rather clean and contain small amount of metal nanoparticles with similar size distributions. (g-i) EDS from the selected area of d-f, respectively, it shows that the selected area of our sample mainly contains C and Fe elements (Cu peaks are from the TEM grid), and the element composition are similar among these three samples. Figure S3. Radial breathing modes of Raman spectra with 488, 514 and 633 nm laser for SWCNTs synthesized at various CO 2 conecntration. According to the relationship between SWCNT diameter (d t ) and raman shift (ω): d t = 217.8/ω -15.7, we can clearly see that with the increase of CO 2 concentration, the SWCNT diameter increases as well. Figure S4. The magnified absorption spectra in visible range which has been normalized. All our samples appear to have similar transmittance.    Figure S10. Absorption spectroscopy of SWCNT films with 1.23 vol% CO 2 , the quenching of S11, S22 and M11 peaks implies the poor quality of the SWCNTs in the thin film. Figure S11. Statistics of (n,m) abundance for SWCNTs with 0.25 vol% of CO 2 at 880 °C. Figure S12. Sheet resistance vs transmittance (at 550 nm) of SWCNT thin film synthesized with different CO 2 concentration at 850 °C, (a) the pristine thin films and (b) the thin films with AuCl 3 doped. The doped thin films with 0, 0.25, 0.37 and 0.50 vol% CO 2 display a sheet resistance of 165, 186, 140 and 86 Ω/sq., respectively, with 90% transmittance at 550 nm. The high conductivity for SWCNT with 0.50 vol% CO 2 may result from its relatively large tube diameter and high ratio of metallic tubes. Figure S13. Catalyst diameter distribution in SWCNTs with CO 2 of 0 vol% (a) and 0.50 vol% (b) measured from TEM, and the corresponding schematic of growth mode: perpendicular (c) and near-tangential mode (d), respectively.