Data on liquid gated CNT network FETs on flexible substrates

This article presents the raw and analyzed data from a set of experiments performed to study the role of junctions on the electrostatic gating of carbon nanotube (CNT) network field effect transistor (FET) aptasensors. It consists of the raw data used for the calculation of junction and bundle densities and describes the calculation of metallic content of the bundles. In addition, the data set consists of the electrical measurement data in a liquid gated environment for 119 different devices with four different CNT densities and summarizes their electrical properties. The data presented in this article are related to research article titled “Metallic-semiconducting junctions create sensing hot-spots in carbon nanotube FET aptasensors near percolation” (doi:10.1016/j.bios.2018.09.021) [1].


Subject area
Physics More specific subject area CNT FETs, Electrical properties Type of data Table, figure How data was acquired Atomic force microscope (Nanosurf, NaioAFM), Agilent 4156C parameter analyser and Rucker and Kolls probestation.

Data format
Raw and analyzed data Experimental factors CNT film morphology and electrical properties of CNT network FETs Experimental features The CNT FETs were fabricated by using simple solution processing methods and photolithography. The CNT film morphology was studied using AFM and the electrical characterizations were measured using Agilent 4156C parameter analyser and a Rucker and Kolls probestation.

Value of the data
A simple method to estimate the metallic content of CNT bundles is provided The electrical data on 119 working devices with their electrical properties is included Data on transfer curves, and gate leakage of liquid gated network CNT FETs with encapsulated electrodes is presented.

AFM characterization of the CNT network films
CNT network morphology was characterized for different CNT deposition times using AFM images. The relevant data was extracted from 5 mm Â 5 mm AFM images for three samples per deposition time.

CNT bundle diameter distribution
Bundle heights were found by analysing the AFM scan data using a linear cross section taken perpendicular to each bundle. The height of the bundle was determined by subtracting the surrounding substrate height for that cross section from the bundle height. This value was then determined to be the bundle diameter. Fig. 1 shows the distribution of the number of counts of bundles at a particular diameter across our device films. Specifications for the CNT buckypaper used to create the films are given by the manufacturer (NanoIntegris) state that the CNT diameters are in the range of 1.2 nm to 1.7 nm with the composition of 1% metallic tubes. Modelling a bundle as a cylinder and assuming 2D packing of tubes we then estimate the number of tubes in each bundle, n, assuming an average CNT diameter of 1.5 nm [2]. Table 1 lists the n values estimated for each bundle diameter observed.
We characterize a bundle as metallic if it contains at least one metallic CNT. Since the CNT buckypaper source contained 1% metallic tubes the probability that a given bundle composed of n tubes can be characterized as a metallic bundle, is p metallic n ð Þ ¼ 1 À 0:99 n From this, the percentage of metallic tubes P metallic in a sample can be determined by where N n is the number of bundles containing n tubes taken from the observed distribution of bundle diameters for our devices shown in Fig. 1. P metallic was calculated for each sample and was averaged

CNT bundle length distribution
The CNT bundle lengths measured by AFM are shown in Fig. 2 and do not significantly vary with the CNT deposition time. The average bundle lengths are found to be 2.00 ( 70.42) mm, 2.09 (7 0.63) mm, 2.16 (70.24) mm, and 2.17 ( 70.51) mm for 10, 20, 40, and 80 min deposited films respectively with the uncertainty resulting from the standard deviation of bundle lengths as shown in Fig. 3. The overall average bundle length was calculated to be 2.14 7 0.45 mm by averaging the bundle length calculated per deposition time.  [3,4]. Therefore, in our AFM images we identified X and Y type junctions and excluded Y junctions from our count of junction density since only X type junctions contribute significantly to the overall network resistance.

CNT bundle-bundle junction and bundle density
The junction and total bundle density values reported in the main text in Fig. 3 are taken from averaging across multiple chips of the same deposition time. The raw data along with the mean and standard deviation per CNT deposition time is shown in Fig. 4.  Fig. 6. The gate current is less than 0.1 nA for all devices. The CNT FETs exhibit ambipolarity which is consistent with previously reported network CNT FETs measured in dry top gate and back gate geometry [5,6].

Electrical characterization of liquid gated CNT FETs and aptasensors
The on-currents in Fig. 5b show an increase of two decades as the suspension times during device fabrication were increased from 10 min to 80 min. The increase in current is attributed to the increased number of conduction paths between source and drain electrodes as the network density increases [7][8][9][10][11]. For the CNT FETs closest to the percolation threshold (deposition times of 10 min) we see larger variation in the on and off currents as expected due to the proximity of those networks to the percolation threshold. As CNT deposition time increases we see an increase in the off current. This is because as the bundle density is increased the network nears the metallic percolation threshold, increasing the likelihood of continuous metallic segments in the network [10,11].
The on-off ratios shown in Fig. 5c are 3 orders of magnitude higher for devices closer to percolation. More than 30% of the devices with CNT deposition times of 10 min have an on-off ratio of over 10 5 , and all are over 10 3 . The threshold voltages of the devices are shown in Fig. 5d. We observe a positive shift in V th and increases in I ds with increasing CNT deposition time, as expected for p-type CNT FETs [12][13][14].

I-V curves of pristine CNT FETs
The I-V curves for pristine unfunctionalized devices show a non-linear response for devices fabricated with a CNT deposition time of 10 min and 20 min as shown in Fig. 7, while those fabricated at 40 min and 80 min CNT deposition times show a linear response. The non-linearity in the 10 min and 20 min devices indicates that the conduction is dominated by potential barriers at m-s junctions within the CNT bundle network.

Experimental design, materials, and methods
The detail experimental methods for the device fabrication, AFM measurements and electrical measurements are described in detail in the manuscript [1].