Hetero-Dimensional 2D Ti₃C₂T_x MXene and 1D Graphene Nanoribbon Hybrids for Machine Learning-Assisted Pressure Sensors

Abstract

Hybridization of low-dimensional components with diverse geometrical dimensions should offer an opportunity for the discovery of synergistic nanocomposite structures. In this regard, how to establish a reliable interfacial interaction is the key requirement for the successful integration of geometrically different components. Here, we present 1D/2D heterodimensional hybrids via dopant induced hybridization of 2D Ti3C2T x MXene with 1D nitrogen-doped graphene nanoribbon. Edge abundant nanoribbon structures allow a high level nitrogen doping (∼6.8 at%), desirable for the strong coordination interaction with Ti3C2T x MXene surface. For piezoresistive pressure sensor application, strong adhesion between the conductive layers and at the conductive layer/elastomer interface significantly diminishes the sensing hysteresis down to 1.33% and enhances the sensing stability up to 10 000 cycles at high pressure (100 kPa). Moreover, large-area pressure sensor array reveals a high potential for smart seat cushion-based posture monitoring application with high accuracy (>95%) by exploiting machine learning algorithm.

Self-Assembly 3D Porous Crumpled MXene Spheres as Efficient Gas and Pressure Sensing Material for Transient All-MXene Sensors

“…However, pressure sensor based on porous crumpled MXene spheres shows the widest linear detection range of 0.14-22.22 kPa (Fig. 4f) compared to the previously reported pressure sensors based on MXene and graphene [14,[38][39][40][41][42][43][44][45]. We tested the effect of load residence times on the sensor of 0.1, 0.5, 1.0, 2.0, and 5.0 s. As shown in Fig.…”

“…Gas and pressure sensors made of the same degradable sensing and electrode materials are ideal for achieving full transiency upon only one external trigger. MXenes (Ti 3 C 2 T x ), as a novel class of two-dimensional nanomaterials with rich surface functional groups, have been identified as the sensing layer and electrode due to their high conductivity, excellent signal-tonoise ratio, and abundant hydroxyl on the surface, which is superior to other metal oxides and two-dimensional (2D) materials [11][12][13][14][15][16][17][18]. Meanwhile, because of their chemical instability, MXenes exhibit controllable transiency in H 2 O 2 and NaOH aqueous solutions [19,20].…”

Artificial Intelligence (AI) and Internet of Medical Things (IoMT) Assisted Biomedical Systems for Intelligent Healthcare

“…The physicochemical properties of Mxene help in enhancing the sensitivity of the sensing devices for biomedical, environmental, and food analytics applications [ 65 ]. Creating hybrid nanocomposite materials by combining a 2D Mxene with 1D nanostructures for enhancing the adhesion stability on the transducer surface for long-term monitoring has also been attempted [ 66 ]. The interfacial integration of a 2D Mxene/1D graphene nanoribbon has been investigated for developing the desired pressure sensor with an improved life cycle.…”

“… Interfacing interconnection of 1D graphene nanoribbons with 2D Mxene for developing pressure sensors trained using a machine learning algorithm. (Reproduced with permission from the American Chemical Society [ 66 ]). …”

Where Nanosensors Meet Machine Learning: Prospects and Challenges in Detecting Disease X

“…This warrants the use of machine learning (ML) algorithms, which are smart programs based on logic and mathematics, to facilitate pattern recognition and multiplex correlative data processing of convoluted output signals . Furthermore, ML algorithms have been instrumental in strengthening predictive analytics for heightened nanosensor detection accuracies. − To date, ML algorithms (including supervised, unsupervised, and reinforcement learning) have been extensively applied in chemical, biomedical, and environmental sensing applications, and beyond. − However, the full potential of incorporating ML in nanosensors has yet to be realized given the large assortment of ML tools available for data preprocessing, feature selection, and feature engineering, and the multitude of ML algorithms. Considering the immense potential of ML-empowered nanosensors, we feel that it is timely to provide an assessment of the recent advances in this area toward enabling the practical detection of the dangerous yet elusive Disease X.…”