Search

Dispersion of nanofiller in the polymer matrix is a vital factor that influences the properties of the fabricated nanocomposites at the laboratory level. Characterization techniques like TEM and FESEM, having a small sample size, tend to miss out on the big picture for the analysis of mixing on a larger scale. At the industrial level, conducting such testing with varying experimental conditions is not viable in terms of both cost and time. Through this study, we propose a simple method to examine the extent of dispersion using a simple camera and employing deep learning (DL) models. For this purpose, an analogous study has been performed to study the sensitivity of the processing techniques, to better understand the findings in our previous articles. A two component-colored solution (oil–water) was utilized as a proxy for the nanofiller-polymer matrix system. Different processing methods were employed namely ultrasonication, homogenization, sequential ultrasonication and homogenization and simultaneous ultrasonication and homogenization. The variation in processing technique significantly affects the dispersion which is attributed to the different mixing mechanisms (turbulent, diffusive, and convective) incurred in these processing techniques. Inferences are withdrawn by detecting patterns in a large sample size which highlights that DL models provide us with a holistic viewpoint of real-time observations. It also ameliorates human interpretation by unraveling obscure information which can go unnoticed by human eyes. [ABSTRACT FROM AUTHOR]

This research article proposes a deep learning framework that addresses two major hurdles in adopting deep learning techniques for solving physics-based problems. One is the requirement of a large data set for training the deep learning (DL) model and another is the consistency of a DL model with the physics of a phenomenon. The framework is generic that can be applied to model a phenomenon in physics if its behavior is known. A semi-supervised physics guided neural network (SPGNN) has been developed based on our framework to demonstrate the concept. SPGNN models the I–V characteristics of gallium nitride based high electron mobility transistors (GaN HEMTs). A two-stage method has been proposed to train a DL model. In the first stage, the DL model is trained via an unsupervised learning method using the analytical physics-based model of a field-effect transistor (FET) as a loss function of the DL model that incorporates the physics of the FET in the DL model. Later, the DL model is fine-tuned with a small set of experimental data in the second stage. Performance of SPGNN has been assessed on various sizes of the data set with 100, 500, 1000, 1500, and 2000 samples. SPGNN significantly reduces the training data requirement by more than 80% and provides better performance than a traditionally trained neural network (TTNN), even for the unseen test data set. SPGNN predicts 32.4% of the unseen test data with less than 1% of error and only 0.4% of the unseen test data with more than 10% of error. [ABSTRACT FROM AUTHOR]

This work reports on the development of a multi-output deep learning (DL) model for simultaneous prediction of the figure of merits (I o n , G m , and V t h ) of a gallium nitride (GaN) based high electron mobility transistors (HEMTs) for various epitaxial structures. To generate an initial data set, 2160 GaN HEMTs have also been simulated by an experimentally validated simulation methodology. A generative adversarial network (GAN) has been also introduced in semiconductor device modeling to augment the training data set. The generated data set by GAN is found to be in good agreement with the initial data set with a Frechet Inception Distance score of 0.151. The final data set has seven dimensions, i.e., aluminum gallium nitride (AlGaN) thickness (t AlGaN ), aluminum content in AlGaN, doping in AlGaN, type of doping in AlGaN, I o n , G m , and V t h , where the first four are inputs and the last three are the outputs of the DL model. The DL model is developed with the possibility of reducing unnecessary use of technology computer-aided design simulations for similar types of problems as such simulations require huge computational resources, expertise, and development time to obtain output. Mean squared error and R-squared values for the predicted I o n , G m , and V t h are 59.69, 4.28, and 0.09, and 0.99, 0.99, and 0.97, respectively. [ABSTRACT FROM AUTHOR]

This work reports rapid, label-free and specific detection of the HER2 antigen using a gallium nitride (GaN) high electron mobility transistor (HEMT). Thiol-based chemistry has been utilized to immobilize the corresponding HER2 antibody in the sensing area of the sensor. The formation of a gold–sulfur complex has been confirmed through Raman spectroscopy, giving a peak at around a wavelength of 260 cm−1. Fourier transform infrared spectroscopy and atomic force microscopy (AFM) also reveal the functionalization of thiol and free carboxylic groups. On-chip enzyme-linked immunosorbent assay has been utilized to confirm immobilization of antibody receptors on the sensing area surface, followed by current–voltage measurement. Morphology of the sensing area using AFM and electrical characterization of the sensor have been recorded before and after each functionalization process step. The sensor shows detection of the HER2 antigen in a broad range of 0.7 pg ml−1 to 10 μg ml−1i.e., (5 × 10−15 to 6 × 10−8 M). A long-time study and reusability aspect of the sensor have also been investigated that show good viability of the sensor. For the first time, a three-binding-site model based on the Langmuir isotherm has been developed for HER2 detection using GaN-HEMTs with three dissociation constants, i.e., 7 × 10−10, 8.8 × 10−11, and 7.2 × 10−9 M, respectively. [ABSTRACT FROM AUTHOR]

We have investigated the characteristics of pH and salinity sensor derived from the gated AlGaN/GaN high-electron mobility transistor (HEMT) structures in phosphate buffer saline (PBS) and aqueous salt solutions (NaCl + DI). In deionized (DI) water, the HEMT device showed good drain I – V characteristics, which is very close to the output characteristics of the typical HEMT structures subjected to the air. We observed a significant change in the output drain characteristics curves concerning to the variation in the pH values of PBS solutions, signifying the subsequent potential variation at the AlGaN surface. The output drain current recorded at Vds = +1 V was linearly decremented with the pH value. A high sensitivity of 4.32~μA/mm-pH was obtained. These GaN HEMT structures demonstrated a quick response to the pH changes. It was also investigated that the devices were susceptible toward the aqueous salt solution (NaCl + DI). The percentage change in drain current linearly decreased with decreasing NaCl molar concentration in DI water. We have reported on the change in current with the smaller range of molar concentration of NaCl present in water. Evaluating the sensitivity and response time, we obtained a high sensitivity of 6.48 mA/mm-molar and a response time of 250–350 ms at Vds = +1 V. We have also reported on the change in current with the molar concentration of NaCl present in PBS with a high sensitivity of 2.02 mA/mm-molar at Vds = +5 V. These outcomes show that the AlGaN/GaN HEMTs are exceptionally promising as a high-sensitivity pH sensor and salinity sensor for biological experiments. [ABSTRACT FROM AUTHOR]