Laser-Induced Carbon-Based Smart Flexible Sensor Array for

Sep 14, 2018 - By mapping the data points, the sensor system could detect flavor combinations and provide a reliable prediction of analyte concentrati...
0 downloads 0 Views 3MB Size
Subscriber access provided by University of South Dakota

Biological and Medical Applications of Materials and Interfaces

Laser-induced Carbon base Smart Flexible Sensor Array for Multi Flavors Detection Yongchao Yu, Pooran C. Joshi, Jayne Wu, and Anming Hu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b12626 • Publication Date (Web): 14 Sep 2018 Downloaded from http://pubs.acs.org on September 15, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Laser-induced Carbon base Smart Flexible Sensor Array for Multi Flavors Detection Yongchao Yu1, Pooran C Joshi 2, Jayne Wu3, *Anming Hu1 1

Department of Mechanical, Aerospace and Biomedical Engineering, University of Tennessee, Knoxville, TN37996, USA 2

3

Oak Ridge National Lab, Oak ridge, TN 37831-6061 USA

Department of Electrical Engineering & Computer Science, University of Tennessee, Knoxville, TN37996, USA

*E-mail: [email protected]. Phone: 865-974-5993 KEYWORDS: e-Tongue, Laser Direct Writing, Sensor Array, Multi-Detection, Laser Fabrication, Carbon base Sensor

ABSTRACT: We report a flexible sensor array electronic tongue system that fabricated on a polymer substrate by the laser direct writing process for multi-flavors decoction. Electronic tongue is a sensing system that is applied to detect different elements with the same sensor array. By analyzing responses from different measurement units, it enables a cross sensitivity, namely the ability of the system responding to a range of different analytes in solution without specific functionalization of sensors. In this paper, a six-unit sensing array system was fabricated by a laser direct writing process. Sensing units were introduced on a flexible polyamide surface. A high surface-volume ratio porous carbon structure was created by a laser-induced carbonization process, which provides stable conductive carbon electrodes with high sensitivity. Different surface treatments such as gold plating, reduced-graphene oxide (rGO) coating and polyaniline (PANI) coating, were accomplished for different measurement units. By applying principal

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

component analysis (PCA), this sensing system shows a promising result for many flavors detection. The detection limits for each element are about 0.1mM for NaCl and sugar solutions. And it is able to detect 10-4 times diluted commercial table vinegar solution, which originally contains 5% of acetic acid. The detection limit is theoretically lower than the human threshold of 10mM for NaCl and sugar. Besides, the sensing system shows a high sensitivity and selectivity for mixed elements. By mapping the data points, the sensor system could detect flavor combinations and provide a reliable prediction of analyte concentration ratios. 1 Introduction A range of bio-nano-sensing technologies have been developed for biomedical diagnostics

1-2

, environment monitoring or pollution monitoring

3-4

, and for monitoring

temperature 5, humidity and other physical parameters in daily life 6-9. The general drawback for these sensors is that these sensors are specifically designed for one certain objective. A sensor sensitive to one physical chemical parameter can be totally irresponsive to another parameter. In sharp contrast, sensing by human beings and animals is based on the same sensing system, such as tongues or nose, but can detect a number of objectives. Canonical classification of the senses taken by neurophysiologists embraces five senses. They can be categorized into two groups, physical senses and chemical sense. For physical senses, it includes sight, hearing, touch, which are adapted to sense physical stimuli; taste and smell are considered as chemical sense, which can perceive changes in the chemical composition of the environment

10

. The concept of

electronic nose was published in 1982 by Oersaud and Dodd. An array of chemical sensors was used to simulate receptors and neural analysis to recognize the pattern of stimulations through pattern recognition methods

11

. The first system for liquid analysis based on the multisensory

array was reported by Otto and Thomas in 1985, and the name of “electronic tongue” was

ACS Paragon Plus Environment

2

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

presented

12

. Since then, the development of multi-array sensing system had a rapid increase in

early 2000s 13-16. Meanwhile many sensing techniques have been developed for electronic tongue application, such as optical and enzymatic sensors

17-19

, the electrochemical or voltammetric

based sensor is still one of the most widely use sensing techniques10. In this paper, voltammetric measurement technique was applied for sensing. To date, most of the sensor arrays were fabricated by cleanroom technologies, such as the lithography and chemical etchings, which are very time consuming and expensive techniques20-22. As our knowledge, this is the first electrical tongue sensing system that fabricated by laser direct writing process, which is a fast, low-cost and environment friendly fabrication process 23. Laser ablation of polymer material has been studied since 1980s 24. In 2014, Tour et al. of Rice University reported their research progress on laser-induced porous graphene (LIG) films from commercial polyimide. In their research, CO2 infrared laser was used for ablation of polyimide substrate. Localized high temperatures and high pressures, resulting by lattice vibration due to the energy from laser irradiation, could easily break the C-O, C=O and N-C bonds, which are rearranged to form porous structures

25-27

. Recently a high sensitive bio-sensor

was fabricated by laser direct writing method using a femtosecond laser3. The porous carbon structures with a very high surface-volume ratio were functionalized with an aptamer for specific sensing. For a pristine sensor without aptamer coating, the sensing unit only present very limited selectivity and is considered as “nonselective”. However, in this work, by integrating responses from a sensor array, different measurement units enable cross sensitivity, namely the ability of the system to respond to a wide number of different analytes in solution without functionalization with probe molecules

28

. To enhance the selectivity and sensitivity of sensor

systems, the large response differences between each measurement units corresponding to the

ACS Paragon Plus Environment

3

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

same analyte is considered as an ideal situation. Six different types of sensor units were prepared by different surface treatments on carbon electrodes allows an electronic tongue-style sensing. 2 Sensor Fabrication and Functionalization 2.1 Laser Direct Writing As the electrode material, laser-induced porous graphene electrode was induced by laser direct writing (LDW) process. Figure 1 (a) shows the LDW system for sensor fabrication. A femtosecond laser, with 1030 nm wavelength, was used as the laser source. The laser beam was focused by a 20X 0.4NA objective lenses. The laser power was adjusted by tuning a 2/λ waveplate and a polarizing beam-splitter. An optical isolator was placed after the shutter to prevent laser beam from reflecting into the femtosecond laser source. A program controllable 3D stage was placed under the object lens to control the movement of substrate and create different patterns. For this experiment, carbon electrodes were fabricated under 600 mW laser power with 60mm/min writing speed. Figure 1(b) shows the image of fabricated carbon electrodes on polyimide (PI) substrate, which presents as one measurement unit in this study. The working electrode area is 0.5cm x 0.6 cm with 8 pairs of electrodes in each measurement unit. The line width for each electrode is 100 µm with 100 µm gap between each electrode. Fig. 1 (c) displays a sensor array with six sensing units. And Fig.1 (b) shows the sensor array fabrication process and functionalization process.

ACS Paragon Plus Environment

4

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 1 (a) The set up of laser direct writing system. Images of (b) a fabricated carbon electrode and (c) a carbon sensor array with six units fabricated on polymer surface. (d) The schematic of surface sensing process 2.2 Sensor Functionalization For general functionalization of sensing units, different surface treatments were accomplished on the laser-induced carbon electrode surface. Six different electrodes constitute a sensor array for the measurement, which are carbon electrode (CE), gold plated CE (GCE), rGO drop coated CE (rGO-CE), rGO drop coated gold plated GCE (rGO-GCE), polyaniline deposited CE (PANI-CE) and polyaniline deposited gold plated GCE (PANI-GCE). Figure 2 (a-f) are typical SEM images of the surface structures for each type of electrode, respectively. Gold Plating:

ACS Paragon Plus Environment

5

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

Gold particles were electrically plated onto the carbon surface using a commercial gold plating solution (Carswell) with a 6V DC power for 10 min. A Pt wire was used as the counter electrode. Figure 2(d) shows the surface structure of a gold-plated CE. As shown in the Figure, a thick gold nanoparticle-assembled layer was deposited on the carbon electrode. This process can significantly increase the conductivity of the electrode. rGO Coating Graphene oxide was prepared by a simplified Hummer’s method

29

. In brief, 1.5g

graphite powder was added to a mixture of 200 mL H2SO4/H3PO4 with volume ratio 9:1, then 9.0 g KMnO4 was added slowly. The mixture was stirred at 55 °C for 12 h for completed reaction. The reaction was cooled to room temperature and the ice was added into the solution, then about 4 mL of 30% H2O2 was added to solution. After stirring for 12h, the solution was set for sediment and separation. The supernatant was removed and 250 mL 10 wt% HCl was added and kept stirring for 12 h. Then, the obtained mixture was separated by centrifuging. After the graphene oxide was compounded, the 1M ascorbic acid solution was used for reducing graphene oxide with the GO: ascorbic acid volume ratio of 1000:1. Additionally, the solution was heated to 95°C in an oil bath and held for 6 hours. 20µL rGO solution was dropped to the electrode’s surface and stored at room temperature until completely dry. Figure 2 (b and e) show images of electrodes after rGO coating. It could be observed that a thin layer of rGO covered on both carbon and gold coated electrodes. Polyaniline Coating Polyaniline (PANI) is one of the most commonly used conductive polymer materials. It has similar electrical properties to inorganic materials such as semiconductors and metals with

ACS Paragon Plus Environment

6

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

various applications30-32. An electrolyte solution of 1M H2SO4 + 0.05M aniline was used for the electrochemical deposition of PANI onto carbon electrode 33. A Pt wire was used as the counter electrode. For electrical deposition, the voltage range was set from -0.2V to 0.8V with a 10mv/s scan speed for 20 cycles. Figure 2 (c,f) show the surface structures of PANI coated electrodes. For PANI-CE sample, it is very clear that some polymer structure was introduced on top of carbon layers after PANI deposition and generated net structure. However, for PANI-GCE sample, it is very difficult to find PANI from SEM images directly. For better understating of PANI deposition on GCE, Fourier transform infrared spectroscopy (FTIR) analysis was conducted as follows.

Figure 2 SEM images of (a) CE, (b)rGO-CE, (c)PANI-CE, (d) GCE, (e)rGO-GCE and (f)PANIGCE

FTIR Analysis:

ACS Paragon Plus Environment

7

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

Figure 3 FTIR analysis for (a) CE and PANI-CE, (b) rGO-CE and rGO-GCE, and (C) GCE and PANI-GCE Figure 3 shows the Fourier transform infrared spectroscopy (FTIR) spectra. From Figure 3 (a) there are some additional peaks could be observed from PNAI-CE samples. The main bands of the PANI base located at 1585cm-1 and 1040 cm-1 33-34. These peaks prove that anillin was deposited on the surface of carbon electrode and produced PANI. For rGO deposited samples, Figure 3 (b) gives the FTIR results for both rGO-GCE and rGO-CE. A strong C-C peaks at 1625cm- 1 was found from both samples. Additionally, C-O (at 1052 cm−1) and C-O-C (at 1226 cm−1) appear on the spectra results

35-36

. These peaks are considered as peaks from polyamide

substrates and graphene oxide that not reduced completely. Some organic group peaks are obvious in Figure 3(c). Peaks show on 1495cm-1 and 1040 cm-1 are two typical peaks for PANI. Especially, the peak on 1495cm-1, represent the benzenoid ring 37. The C-N bond (1374cm-1) also could be observed from the Figure34, which is provided by PANI . FTIR results evidence that PANI and rGO were successfully deposited on electrodes. 3 Measurement Setup and Sensing Mechanism. Before testing target substances, the impedance response of e-tongue sensor system was characterized with distilled water. Figure 4 (a) shows the sensing system setup for measurements.

ACS Paragon Plus Environment

8

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

The testing solution was applied to each sensor unit and signals were collected through six individual channels with an electrochemical working station. Figure 4 (b) shows impedance responses with different frequencies for each measuring unit. 20µL distilled water was dropped to sensors and impedance data was collect by biologic SP200 Electrochemical working station. The frequency spanned from 1Hz to 1M Hz. As shown in Figure 4 (b), GCE based sensor shows small impedance response than CE based sensor units, which means the GCE sensing unit has a lower resistance. PANI coated sensor show lower impedance responses for both GCE and CE. Non-coated sensors show the largest impedance response for distilled water. Furthermore, all the sensor units provided different responses for the distilled water, which is essential for the principal component analysis (PCA). Typically, the impedance of the measurement unit is dominated from three parts. For lower frequency range, under 100 Hz, the responses are driven by double layer effect formed at the electrode/electrolyte interface. In the range of 102 and 104 Hz, it appears the effect from the coating material over the electrode. And at higher frequencies ( >100 kHz) the impedance is mainly governed by the geometric capacitance 20, 38. The objective for this work is to detect different solutions, therefore, measurement data for the PCA process were grouped and collected from electrical impedance at 100 Hz, which is the frequency that provides more information about the sample solutions themselves and interface between electrodes and sample solutions. Also, as shown in Fig. Fig. 4 (b), the impedances for all electrodes are clearly separated.

ACS Paragon Plus Environment

9

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

Figure 4 (a) The sensing system setup and (b) AC impedance response for distilled water Salt (NaCl), sugar and vinegar solution were used as three elements in this experiment to represent three different flavors, salty, sweet and sour. It is well known that NaCl exists in the form of Na+ and Cl- ions in water. And the vinegar dissolves in water as H+ and CH3COO-. These ions provide a good ionic strength. In addition, vinegar solution presents as a faintly acid solution. However, sugar, as an organic compound, it exists in water with the simple molecule form. Therefore, the impedance responses for each solution will be correlated to unique properties of each electrolyte. In order to achieve a high sensitivity, the porous structure carbon was used as the electrode material. The porous structure carbon electrode provide large surface-volume ratio, which will enforce the sensitivity improvment3. As well known that the carbon material provides a high chemical suitability, none of three components, salt, sugar and vinegar, that used in this experiment will react with carbon material. Therefore, the porous structure carbon is an ideal material for voltammetric base sensor. By plating Au nanoparticles, it not only drops the internal resistance of the sensor, but also provides enhancements of the local electromagnetic field due to its nanoparticle nature39. This makes the gold-plated sensor more sensitive to changes in

ACS Paragon Plus Environment

10

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

impedances. Based on these two types of electrodes, the carbon electrode and the Au coated electrode, rGO and PANI were further coated on both types of electrodes to further distinguish electrochemical properties for different sensing units. Both rGO and PANI are promising materials for electrochemical and bio-sensing40-45. The rGO was adopted on electrodes to change the surface properties and the internal resistance of sensing units. Furthermore, a high density of edge-plane-like defect sites, which provides a positive site in electron transfer to chemical and biological species, has been observed for chemically derived graphene46. This property may show different responses for Na+, Cl-, H+ and CH3COO- ions. Additionally, as reported, sucrose molecules will interact with the surface group of the graphene and more functional groups will be created

47

. Therefore, impedance responses from rGO coated sensor units will be easily

distinguished for different analytes. For the similar purpose, the PANI was coated on both carbon and gold particle coated carbon electrodes. The PANI has been reported as a suitable material for pH sensing in aqueous medium recently

48-50

. Take advantage of this property, PANI is an ideal coating material to

detect the vinegar solution, which is an acidic solution. The mechanism of pH sensing is basically the function of the ion-selectivity for PANI. This property also shows very promising results for NaCl and sucrose solutions detections38, 51. Additionally, PANI has been reported as sugar-sensitive materials for determination of saccharides52. These features of PANI are attractive to improve the selectivity of the sensing system. As a piece of evidence, all sensing units in the sensor array provide different impedance responses for water, as it is shown in Figure 4(b).

ACS Paragon Plus Environment

11

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

4 Principle Component Analysis Result for Multi Elements Sensing To fully explore the sensitivity of the sensor system, different concentrations of target samples were applied, from 0.1M/L to 1 µM/L for NaCl and sugar, which covered five orders difference of analyte concentration. A commercial table vinegar was purchased from a store, which contains 5% acetic acid, and it was used as the initial concentration of the vinegar solution. Then DI water was added to dilute vinegar solution to different concentrations (0.1 to 10-6 times diluted). Figure 5 shows the PCA results for NaCl, vinegar and sugar with different concentrations. Black dots, red dots and blue dots are representing NaCl, vinegar and sugar respectively. The PCA is a statistical analyses process for identifying the data correlations. The principal components present the maximum variation of the data. By retaining most of the variation, the dimension of the data set can be reduced. The first principal component (PC1) represents the direction at which the data show the largest variation. The second principal component (PC2) is the direction independent to the PC1 and shows the 2nd largest variation. A typical PCA plot can evaluate the differences and similarities between samples and group the data points with a low-dimensional plot

53

. Different surface treatments were accomplished for

different sensing units. Therefore, each sensing unit provides different responses for individual analyte. Also, along the concentration changes, the responses from each sensing unit will changes accordingly. However, due to different material properties, the changing rates and tendencies that collected from different sensing units may vary for different analysts. By applying PCA, the correlations of data set could be found and used for grouping, identifying and determining the tested analytes. As Figure 5 shows, separation and grouping of data sets are observed after PCA process, also liner dependencies are found with changing of sample conservations. The NaCl samples are grouping at the left side. As the conversation of NaCl

ACS Paragon Plus Environment

12

Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

decreases, the data points move to right and rise gradually from 0.1M to 0.1mM. However, as concentration decreases to 10uM and 1uM, a large shift occurs. Furthermore, a clear linear dependency also can be observed from sugar samples. Data points for sugar samples grouping on the right side of the plot. As the concentration decreases, the data set move to bottom side of the graph. Same as NaCl mapping result, data points show a significant drop at 10uM and 1uM. This is considered as the demotion limitation for this sensing system. The vinegar samples show a very clear liner dependence as samples were diluted from 0.1 to 10-4. Data points are moving from the left side to the right side as decrease of concentrations. And the result also shows discrepancies for concertation of 10-5 and 10-6. This PCA graph shows that the sensor array system can determine the different targets, NaCl, vinegar and sugar in this case. And the detection limitation for this sensing system is 0.1mM, which is much lower than human threshold that has been reported as 10 mM for both NaCl and sucrose solutions 54. By grouping data points and by mapping data points on the graph, it can detect component of targets, which means, it could be used to differentiate whether the unknown solution is a NaCl solution, a vinegar or a sugar solution.

ACS Paragon Plus Environment

13

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

Figure 5 Principal component factor scores obtained for target solutions For better understanding of the concentration effects in the mapping position, a series of data fittings were conducted. Data points for PCA were replotted to show the dependency between concentrations and values of PC1 and PC2 for each sample. Figure 6 shows fitting results for different analytes. This provides a good base for verifying concentration values of testing samples. For any concentration data point can be enforced by PCA data process for PC1 and PC2 values, there is a specific concentration to corresponding to this point. Meanwhile excepting PC2 fitting for vinegar and PC1 fitting for NaCl showing relatively low coefficients of determination value (R2), 0.62 and 0.83 respectively, R2 values for other fittings are over 0.9, which represent that over 90% of the data could be explained by fitting models. By combining the PCA result in Figure 5 and linear fitting result, the sensor system could easily determine the

ACS Paragon Plus Environment

14

Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

type of an unknown solution and its approximated concentration. This analysis proves that the sensor system can not only differentiate type of measuring samples but also provide a quantitative value for concentration of samples.

Figure 6 Fitting data points for (a) NaCl, (b) vinegar and (c) sugar The PCA and fitting process show a promising result for using the sensor array system to detect and identify the single component and the concentration of the testing sample. Based on this aspect, mixtures of different components with different concentrations were prepared for measuring impedance responses with the sensor system. Figure 7 shows the PCA plot for impedance responses of mixed solutions. Three concentrations of mixture were prepared with two different analytes: 0.1M A + 0.001M B, 0.01M A+ 0.01M B and 0.001M A + 0.1M B, where A and B stand for any two arbitrary analytes chosen from NaCl, sugar and vinegar. 20µL mixed solution was applied to each sensing unit for measurements. As a result, Figure 7(b) shows the PCA result for vinegar and sugar mixture. The PCA result shows as green data points on Figure 7(a). For the low sugar concentration solutions (0.01M sugar and 0.1M vinegar), the data point for the PCA is close to the 0.1 vinegar range. As increase of the sugar concentration to 0.1M, the data point moved to the right and get close 0.1M sugar regime. And data points distribution is very linear. Same phenomena also could be

ACS Paragon Plus Environment

15

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

observed for the NaCl/vinegar mixture and the NaCl/sugar mixture, as be shown in Figure 7(c) and Figure 7(d). Overall, for all the mixture solutions, the PCA results show that sensing system can detect different element combinations with various concentration ratios. The response of the sensing system can reflect and further be used for deducing element combinations in mixed solutions. This provides a dependable selectivity for the sensing system. Moreover, results show that the sensor has remarkable responses with diluted elements. Additionally, this response shows a linear relationship with the concentration change. This displays that the sensing system possesses a very promising sensitivity. Besides, the current analysis indicates the possibility of such a sensor array system to measure a complex mixture solution. i.e., a multi element detection with a single measurement system.

ACS Paragon Plus Environment

16

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 7 (a) The PCA plot for electrical resistance responses of different mixtures, (b) the enlarged figure for vinegar and sugar mixture, (c) the enlarged figure for NaCl and vinegar mixture and (d) the enlarge figure for NaCl and sugar mixture. 5 Conclusions This paper reported a flexible sensor system for multi-flavor detection. A flexible sensor array was fabricated on PI substrate using a laser direct witting system. The carbon-based sensor unites are initially functionalized by different treatments, to offer various impedance responses for sensing NaCl, sugar and vinegar. Principal component analysis was accomplished for analyzing the experimental data. PCA results show that the sensor array system allows to detect three different flavors, i.e., salty, sweet and sour, and the linear dependence of both PC1 and PC2 could be observed as the concentration changes for each flavor. Additionally, the sensor array was also capable of discriminating mixtures containing two different flavors at different concentrations. Acknowledgments This work is jointly supported by the Initiative for PON/POC Nanobiosensing (IPN)-Organized Research Unit (ORU) at the University of Tennessee Knoxville and an ORNL contract on sensing and characterization development. This work was also performed in part at Oak Ridge National Laboratory, operated by UT-Battelle for the U.S. Department of Energy under contract no. DE-AC05-00OR22725. 5. References 1. Yu, Y.; Al Mamun, K. A.; Shanta, A. S.; Islam, S. K.; McFarlane, N., Vertically Aligned Carbon Nanofibers as a Cell Impedance Sensor. IEEE Trans. Nanotechnol. 2016, 15 (6), 856861.

ACS Paragon Plus Environment

17

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

2. Al Mamun, K. A.; Tulip, F. S.; MacArthur, K.; McFarlane, N.; Islam, S. K.; Hensley, D., Vertically Aligned Carbon Nanofiber based Biosensor Platform for Glucose Sensor. nt. J. High Speed Electron. Syst. 2014, 23 (01n02), 1450006. 3. Cheng, C.; Wang, S.; Wu, J.; Yu, Y.; Li, R.; Eda, S.; Chen, J.; Feng, G.; Lawrie, B.; Hu, A., Bisphenol a Sensors on Polyimide Fabricated by Laser Direct Writing for Onsite River Water Monitoring at Attomolar Concentration. ACS Appl. Mater. Interfaces 2016, 8 (28), 17784-17792. 4. Malzahn, K.; Windmiller, J. R.; Valdés-Ramírez, G.; Schöning, M. J.; Wang, J., Wearable Electrochemical Sensors for In Situ Analysis in Marine Environments. Analyst 2011, 136 (14), 2912-2917. 5. Bakker, A.; Huijsing, J. H., Micropower CMOS Temperature Sensor with Digital Output. IEEE J. of Solid-State Circuits 1996, 31 (7), 933-937. 6. Kuang, Q.; Lao, C.; Wang, Z. L.; Xie, Z.; Zheng, L., High-sensitivity Humidity Sensor Based on a Single SnO2 Nanowire. JACS. 2007, 129 (19), 6070-6071. 7. Sikarwar, S.; Yadav, B., Opto-electronic Humidity Sensor: A Review. Sens. Actuators. A Phys. 2015, 233, 54-70. 8. Mukhopadhyay, S. C., Wearable Sensors for Human Activity Monitoring: A Review. IEEE Sens J. 2015, 15 (3), 1321-1330. 9. Gao, W.; Emaminejad, S.; Nyein, H. Y. Y.; Challa, S.; Chen, K.; Peck, A.; Fahad, H. M.; Ota, H.; Shiraki, H.; Kiriya, D., Fully Integrated Wearable Sensor Arrays for Multiplexed in situ Perspiration Analysis. Nature 2016, 529 (7587), 509. 10. Ciosek, P.; Wroblewski, W., Sensor Arrays for Liquid Sensing–Electronic Yongue Systems. Analyst 2007, 132 (10), 963-978. 11. Persaud, K.; Dodd, G., Analysis of Discrimination Mechanisms in the Mammalian Olfactory System Using a Model Nose. Nature 1982, 299 (5881), 352. 12. Otto, M.; Thomas, J., Model Studies on Multiple Channel Analysis of Free Magnesium, Calcium, Sodium, and Potassium at Physiological Concentration Levels with Ion-Selective Electrodes. Anal. Chem. 1985, 57 (13), 2647-2651. 13. Jurs, P. C.; Bakken, G.; McClelland, H., Computational Methods for the Analysis of Chemical Sensor Array Data from Volatile Analytes. Chem. Rev. 2000, 100 (7), 2649-2678. 14. Krantz-Rülcker, C.; Stenberg, M.; Winquist, F.; Lundström, I., Electronic Tongues for Environmental Monitoring based on Sensor Arrays and Pattern Recognition: A Review. Anal. Chim. Act 2001, 426 (2), 217-226. 15. Palacios, M. A.; Nishiyabu, R.; Marquez, M.; Anzenbacher, P., Supramolecular Chemistry Approach to the Design of A High-resolution Sensor Array for Multianion Detection in Water. J. Am. Chem. Soc. 2007, 129 (24), 7538-7544. 16. Palacios, M. A.; Wang, Z.; Montes, V. A.; Zyryanov, G. V.; Anzenbacher Jr, P., Rational Design of a Minimal Size Sensor Array for Metal Ion Detection. J. Am. Chem. Soc. 2008, 130 (31), 10307-10314. 17. Premanode, B.; Toumazou, C., A Novel, Low Power Biosensor for Real Time Monitoring of Creatinine and Urea in Peritoneal Dialysis. Sens Actuators B Chem. 2007, 120 (2), 732-735. 18. Lvova, L.; Pudi, R.; Galloni, P.; Lippolis, V.; Di Natale, C.; Lundström, I.; Paolesse, R., Multi-transduction sensing films for Electronic Tongue applications. Sens Actuators B Chem. 2015, 207, 1076-1086.

ACS Paragon Plus Environment

18

Page 19 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

19. Gutés, A.; Ibanez, A.; Del Valle, M.; Cespedes, F., Automated SIA e‐Tongue Employing a Voltammetric Biosensor Array for the Simultaneous Determination of Glucose and Ascorbic Acid. Electroanalysis. 2006, 18 (1), 82-88. 20. Facure, M. H.; Mercante, L. A.; Mattoso, L. H.; Correa, D. S., Detection of Trace Levels of Organophosphate Pesticides Using an Electronic Tongue based on Graphene Hybrid Nanocomposites. Talanta 2017, 167, 59-66. 21. Barié, N.; Bücking, M.; Stahl, U.; Rapp, M., Detection of Coffee Flavour Ageing by Solid-phase Microextraction/Surface Acoustic Wave Sensor Array Yechnique (SPME/SAW). Food Chem. 2015, 176, 212-218. 22. Zhao, Y.; Yang, Q.; Chang, Y.; Qu, H.; Pang, W.; Zhang, H.; Duan, X., Detection and Discrimination of Volatile Organic Compounds Using a Single Multi-Resonance Mode Piezotransduced Silicon Bulk Acoustic Wave Resonator (PSBAR) as Virtual Sensor Array. Sens Actuators B Chem. 2018, 254, 1191-1199. 23. Yu, Y.; Wang, S.; Ma, D.; Joshi, P.; Hu, A., Recent Progress on Laser Manufacturing of Microsize Energy Devices on Flexible Substrates. JOM 2018,70(9),1816-1822 . 24. Srinivasan, R.; Leigh, W., Ablative Photodecomposition: Action of Far-ultraviolet (193 nm) Laser Radiation on Poly (ethylene terephthalate) Films. J. Am. Chem. Soc. 1982, 104 (24), 6784-6785. 25. Lin, J.; Peng, Z.; Liu, Y.; Ruiz-Zepeda, F.; Ye, R.; Samuel, E. L.; Yacaman, M. J.; Yakobson, B. I.; Tour, J. M., Laser-induced Porous Graphene Films from Commercial Polymers. Nat. Commun. 2014, 5, 5714. 26. Wang, S.; Yu, Y.; Li, R.; Feng, G.; Wu, Z.; Compagnini, G.; Gulino, A.; Feng, Z.; Hu, A., High-performance Stacked in-plane Supercapacitors and Supercapacitor Array Fabricated by Femtosecond Laser 3D Direct Writing on Polyimide Sheets. Electrochim. Act 2017, 241, 153161. 27. Hu, A.; Rybachuk, M.; Lu, Q.-B.; Duley, W., Direct Synthesis of Sp-bonded Carbon Chains on Graphite Surface by Femtosecond Laser Irradiation. Appl Phys Lett 2007, 91 (13), 131906. 28. Riul Jr, A.; Dantas, C. A.; Miyazaki, C. M.; Oliveira Jr, O. N., Recent Advances in Electronic Tongues. Analyst 2010, 135 (10), 2481-2495. 29. Chen, J.; Yao, B.; Li, C.; Shi, G., An Improved Hummers Method for Eco-Friendly Synthesis of Graphene Oxide. Carbon 2013, 64, 225-229. 30. Huang, J.; Virji, S.; Weiller, B. H.; Kaner, R. B., Polyaniline Nanofibers: Facile Synthesis and Chemical Sensors. J. Am. Chem. Soc. 2003, 125 (2), 314-315. 31. Fowler, J. D.; Virji, S.; Kaner, R. B.; Weiller, B. H., Hydrogen Detection by Polyaniline Nanofibers on Gold and Platinum Electrodes. J. Phys. Chem. C 2009, 113 (16), 6444-6449. 32. Baker, C. O.; Huang, X.; Nelson, W.; Kaner, R. B., Polyaniline Nanofibers: Broadening Applications for Conducting Polymers. Chem Soc Rev 2017, 46 (5), 1510-1525. 33. Gupta, V.; Miura, N., Polyaniline/Single-Wall Carbon Nanotube (PANI/SWCNT) Composites for High Performance Supercapacitors. Electrochim. Acta 2006, 52 (4), 1721-1726. 34. Kang, E.; Neoh, K.; Tan, K., Polyaniline: A polymer with Many Interesting Intrinsic Redox States. Prog Polym Sci 1998, 23 (2), 277-324. 35. Lu, J.; Li, Y.; Li, S.; Jiang, S. P., Self-assembled Platinum Nanoparticles on Sulfonic Acid-grafted Graphene as Effective Electrocatalysts for Methanol Oxidation in Direct Methanol Fuel Cells. Sci. Rep. 2016, 6, 21530.

ACS Paragon Plus Environment

19

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

36. Li, R.-Z.; Peng, R.; Kihm, K.; Bai, S.; Bridges, D.; Tumuluri, U.; Wu, Z.; Zhang, T.; Compagnini, G.; Feng, Z., High-rate In-plane Micro-supercapacitors Scribed Onto Photo Paper sing in situ Femtolaser-Reduced Graphene Oxide/Au nanoparticle microelectrodes. Energy Environ. Sci. 2016, 9 (4), 1458-1467. 37. Tang, J.; Jing, X.; Wang, B.; Wang, F., Infrared Spectra of Soluble Polyaniline. Synth. Met. 1988, 24 (3), 231-238. 38. Riul Jr, A.; Soto, A. G.; Mello, S.; Bone, S.; Taylor, D.; Mattoso, L., An Electronic Tongue using Polypyrrole and Polyaniline. Synth. Met. 2003, 132 (2), 109-116. 39. Matsui, J.; Takayose, M.; Akamatsu, K.; Nawafune, H.; Tamaki, K.; Sugimoto, N., Molecularly Imprinted Nanocomposites for Highly Sensitive SPR Detection of A Non-aqueous Atrazine Sample. Analyst 2009, 134 (1), 80-86. 40. Yu, X.; Zhang, W.; Zhang, P.; Su, Z., Fabrication Technologies and Sensing Applications of Graphene-based Composite Films: advances and challenges. Biosens. Bioelectron. 2017, 89, 72-84. 41. Vilian, A. E.; Mani, V.; Chen, S.-M.; Dinesh, B.; Huang, S.-T., The Immobilization of Glucose Oxidase at Manganese Dioxide Particles-decorated Reduced Graphene Oxide Sheets for the Fabrication of a Glucose Biosensor. Ind Eng Chem Res 2014, 53 (40), 15582-15589. 42. Wang, Q.; Wang, Q.; Li, M.; Szunerits, S.; Boukherroub, R., Preparation of Reduced Graphene Oxide/Cu Nanoparticle Composites Through Electrophoretic Deposition: Application for Nonenzymatic Glucose Sensing. RSC Adv. 2015, 5 (21), 15861-15869. 43. Parlak, O.; Tiwari, A.; Turner, A. P.; Tiwari, A., Template-directed Hierarchical Selfassembly of Graphene based Hybrid Structure for Electrochemical Biosensing. Biosens. Bioelectron. 2013, 49, 53-62. 44. Choi, B. G.; Im, J.; Kim, H. S.; Park, H., Flow-injection Amperometric Glucose Biosensors based on Graphene/Nafion Hybrid Electrodes. Electrochim. Acta 2011, 56 (27), 9721-9726. 45. Ruecha, N.; Rangkupan, R.; Rodthongkum, N.; Chailapakul, O., Novel Paper-based Cholesterol Biosensor using Graphene/Polyvinylpyrrolidone/Polyaniline Nanocomposite. Biosens. Bioelectron. 2014, 52, 13-19. 46. Kuila, T.; Bose, S.; Khanra, P.; Mishra, A. K.; Kim, N. H.; Lee, J. H., Recent Advances in Graphene-based Biosensors. Biosens. Bioelectron. 2011, 26 (12), 4637-4648. 47. Wu, Y.; Wen, Z.; Feng, H.; Li, J., Sucrose‐Assisted Loading of LiFePO4 Nanoparticles on Graphene for High‐Performance Lithium‐Ion Battery Cathodes. Chemistry 2013, 19 (18), 5631-5636. 48. Yoon, J. H.; Hong, S. B.; Yun, S.-O.; Lee, S. J.; Lee, T. J.; Lee, K. G.; Choi, B. G., High Performance Flexible pH Sensor based on Polyaniline Nanopillar Array Electrode. J Colloid Interface Sci 2017, 490, 53-58. 49. Pandey, P. C.; Singh, G., Tetraphenylborate Doped Polyaniline based Novel pH Sensor and Solid-State Urea Biosensor. Talanta 2001, 55 (4), 773-782. 50. Karyakin, A. A.; Vuki, M.; Lukachova, L. V.; Karyakina, E. E.; Orlov, A. V.; Karpachova, G. P.; Wang, J., Processible Polyaniline as an Advanced Potentiometric pH Transducer. Application to Biosensors. Anal. Chem. 1999, 71 (13), 2534-2540. 51. Riul Jr, A.; Malmegrim, R.; Fonseca, F.; Mattoso, L., An Artificial Taste Sensor based on Conducting Polymers. Biosens. Bioelectron. 2003, 18 (11), 1365-1369. 52. Bossi, A.; Piletsky, S. A.; Piletska, E. V.; Righetti, P. G.; Turner, A. P., An Assay for Ascorbic Acid based on Polyaniline-coated Microplates. Anal. Chem. 2000, 72 (18), 4296-4300.

ACS Paragon Plus Environment

20

Page 21 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

53. 304. 54.

Ringnér, M., What is Principal Component analysis? Nat. Biotechnol. 2008, 26 (3), 303Pfaffman, C., Handbook of Physiology. American Physiological Society: 1959; Vol. 1.

For Table of Contents

ACS Paragon Plus Environment

21

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1 (a) The set up of laser direct writing system. Images of (b) a fabricated carbon electrode and (c) a carbon sensor array with six units fabricated on polymer surface. (d) The schematic of surface sensing process 297x177mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 22 of 29

Page 23 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 2. SEM images of (a) CE, (b)rGO-CE, (c)PANI-CE, (d) GCE, (e)rGO-GCE and (f)PANI-GCE 165x73mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3 FTIR analysis for (a) CE and PANI-CE, (b) rGO-CE and rGO-GCE, and (C) GCE and PANI-GCE 332x97mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 24 of 29

Page 25 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 4 (a) The sensing system setup and (b) AC impedance response for distilled water 305x114mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5 Principal component factor scores obtained for target solutions 227x175mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 26 of 29

Page 27 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 6 Fitting data points for (a) NaCl, (b) vinegar and (c) sugar 493x128mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7 (a) The PCA plot for electrical resistance responses of different mixtures, (b) the enlarged figure for vinegar and sugar mixture, (c) the enlarged figure for NaCl and vinegar mixture and (d) the enlarge figure for NaCl and sugar mixture. 350x242mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 28 of 29

Page 29 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

For Table of Contents Only 339x169mm (300 x 300 DPI)

ACS Paragon Plus Environment