Citation: Zhang, R.; Hao, T.; Hu, S.;
Wang, K.; Ren, S.; Tian, Z.; Jia, Y.
Electrolyte-Gated Graphene Field
Effect Transistor-Based Ca
2+
Detection Aided by Machine
Learning. Sensors 2023, 23, 353.
https://doi.org/10.3390/s23010353
Academic Editors: M. Jamal Deen,
Subhas Mukhopadhyay, Yangquan
Chen, Simone Morais, Nunzio
Cennamo and Junseop Lee
Received: 30 November 2022
Revised: 21 December 2022
Accepted: 27 December 2022
Published: 29 December 2022
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Electrolyte-Gated Graphene Field Effect Transistor-Based Ca
2+
Detection Aided by Machine Learning
Rong Zhang
1,2
, Tiantian Hao
1
, Shihui Hu
1
, Kaiyang Wang
1
, Shuhui Ren
1
, Ziwei Tian
1
and Yunfang Jia
1,
*
1
College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, China
2
School of Integrated Circuit Science and Engineering, Tianjin University of Technology, Tianjin 300384, China
* Correspondence: jiayf@nankai.edu.cn
Abstract:
Flexible electrolyte-gated graphene field effect transistors (Eg-GFETs) are widely developed
as sensors because of fast response, versatility and low-cost. However, their sensitivities and respond-
ing ranges are often altered by different gate voltages. These bias-voltage-induced uncertainties
are an obstacle in the development of Eg-GFETs. To shield from this risk, a machine-learning-
algorithm-based LgGFETs’ data analyzing method is studied in this work by using Ca
2+
detection as
a proof-of-concept. For the as-prepared Eg-GFET-Ca
2+
sensors, their transfer and output features are
first measured. Then, eight regression models are trained with the use of different machine learning
algorithms, including linear regression, support vector machine, decision tree and random forest,
etc. Then, the optimized model is obtained with the random-forest-method-treated transfer curves.
Finally, the proposed method is applied to determine Ca
2+
concentration in a calibration-free way,
and it is found that the relation between the estimated and real Ca
2+
concentrations is close-to y = x.
Accordingly, we think the proposed method may not only provide an accurate result but also simplify
the traditional calibration step in using Eg-GFET sensors.
Keywords:
electrolyte-gated graphene field effect transistor; Ca
2+
detection; machine learning;
regression model; calibration-free; flexible
1. Introduction
Calcium is an essential inorganic element in the human body, and it plays an important
role in physiological activities, such as skeletal development [
1
], regulation of normal
cell functions [
2
], gene transcription [
3
] and so on. Inadequate or excessive intake of
calcium is associated with increased risks of osteoporosis [
4
], urinary stone disease [
5
],
cardiovascular disease [
6
], colorectal cancer [
7
] and hypertension [
8
]. Therefore, it is
important to determine the concentration of Ca
2+
in water which is an important source of
calcium intake. Many techniques have been employed to detect calcium ions, including
atomic absorption spectrometry [
9
], fluorescence detection [
10
] and inductively coupled
plasma optical emission spectrometry [
11
]. However, these methods require expensive
instruments and extensive sample preparation [10].
An electronic method based on electrochemistry is preferable to the currently used instant-
assay conventional methods due to its operational simplicity, cost savings and suitability for
real-time detection [
12
]. As a kind of electrochemical sensor, an electrolyte-gated graphene
field effect transistor (Eg-GFET) uses graphene as the channel material. A unique ambipolar
electric field effect along with high carrier mobility enable Eg-GFET to respond to target
molecules very quickly and sensitively [
13
,
14
]. Traditional bioassay always relies on calibration
which is the relation between output electrical signal and target ions’ concentration obtained
by using single variable approaches under some fixed working condition, i.e., the voltages
of gate (V
g
) and drain source (V
ds
) for Eg-GFET. These variables could be the Dirac point
voltage (V
CNP
) shift on transfer curves with the target concentration changing at a constant
V
ds
[
15
,
16
] or the conductance change ratio of the graphene channel at a constant V
g
[
17
] and
the source-drain current (I
ds
) change ratio at constant V
g
and V
ds
[18].
Sensors 2023, 23, 353. https://doi.org/10.3390/s23010353 https://www.mdpi.com/journal/sensors
