Nihar Ranjan Mohapatra
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Nihar Ranjan Mohapatra has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
- Mandar S. Bhoir
Department of Electrical Engineering, IIT Gandhinagar, Gandhinagar 382355, India
8 shared publications - Mohit Dineshkumar Ganeriwala
Department of Electronics and Computer Science Universidad de Granada Avenida de la Fuente Nueva S/N 18071 Granada Spain
8 shared publications - Kumari Neeraj Kaushal
Department of Electrical Engineering, IIT Gandhinagar, Gandhinagar, India
7 shared publications - Om Maheshwari
Indian Institute of Technology Gandhinagar
5 shared publications - Udayan Ganguly
Indian Institute of Technology Bombay
5 shared publications - Aishwarya Singh
Department of Electrical Engineering, Indian Institute of Technology Gandhinagar, Gandhinagar, Gujarat, India
4 shared publications
Latest Publications
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Semiconductor Manufacturing
- Vol. 39  (3) , 410-416
Unified Machine Learning Framework for Process–Device–Circuit Variability Quantification of Inner Spacer Etch in GAA FETs
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Semiconductor Manufacturing
- Vol. 39  (3) , 410-416
Unified Machine Learning Framework for Process–Device–Circuit Variability Quantification of Inner Spacer Etch in GAA FETs
- 1 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Semiconductor Manufacturing
- Vol. 39  (2) , 266-273
Sample-Efficient Modeling and Optimization of High-Dimensional Semiconductor Processes Using Output-Augmented Gaussian Process Regression and Dynamic Bayesian Optimization
Semiconductor process modeling in early stages of process development is challenging due to limited availability of the data, high dimensionality of inputs and outputs, and chamber-to-chamber (C2C) variations. This paper presents a unified ‘OGE’ framework combining Output Augmented Modeling (OAM) to Gaussian Process Regression (GPR), with Ensemble Feature Removal (EFR) to address these issues. EFR robustly eliminates weak predictors across multiple validation trials, while OAM recursively enhances model accuracy by incorporating high-performing outputs as auxiliary inputs. The resulting multi-input single-output models are stitched into a multi-output structure and optimized using Dynamic Bayesian Optimization to meet target specifications. Applied to four semiconductor process datasets with up to 100 input features and fewer than 85 samples, the proposed approach consistently outperforms conventional techniques, demonstrating improved generalization, interpretability, and robustness to C2C variations. The framework provides actionable insights for process tuning and is well-suited for data-driven optimization in early-stage semiconductor manufacturing.
- 1 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Semiconductor Manufacturing
- Vol. 39  (2) , 266-273
Sample-Efficient Modeling and Optimization of High-Dimensional Semiconductor Processes Using Output-Augmented Gaussian Process Regression and Dynamic Bayesian Optimization
Semiconductor process modeling in early stages of process development is challenging due to limited availability of the data, high dimensionality of inputs and outputs, and chamber-to-chamber (C2C) variations. This paper presents a unified ‘OGE’ framework combining Output Augmented Modeling (OAM) to Gaussian Process Regression (GPR), with Ensemble Feature Removal (EFR) to address these issues. EFR robustly eliminates weak predictors across multiple validation trials, while OAM recursively enhances model accuracy by incorporating high-performing outputs as auxiliary inputs. The resulting multi-input single-output models are stitched into a multi-output structure and optimized using Dynamic Bayesian Optimization to meet target specifications. Applied to four semiconductor process datasets with up to 100 input features and fewer than 85 samples, the proposed approach consistently outperforms conventional techniques, demonstrating improved generalization, interpretability, and robustness to C2C variations. The framework provides actionable insights for process tuning and is well-suited for data-driven optimization in early-stage semiconductor manufacturing.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of the Electron Devices Society
- p. 1
Stress-Modulated Defect Engineering in Monolayer WS2: A First-Principles Study
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of the Electron Devices Society
- p. 1
Open AccessStress-Modulated Defect Engineering in Monolayer WS2: A First-Principles Study
- 14 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45  (3) , 1410-1417
Inverse Design of High Power and High Voltage LDMOS Transistors Using Deep Learning-Based Sample-Efficient Surrogate Model
Device design using Machine Learning has been used in the semiconductor industry over the past ten years. However, the generation of the training data set for precise predictions using this technique remains burdensome. Addressing this, in this work, we propose eight sample-efficient techniques to train the deep neural network (DNN) based surrogate models that emulate technology computer-aided design (TCAD). We showcase their efficacy by predicting off-state breakdown voltage $(BV_{\mathrm { DS,off}})$ and specific on-resistance $(R_{\mathrm { sp}})$ of a laterally diffused metal oxide semiconductor field-effect transistor (LDMOSFET). Our findings highlight the potential for ~38% reduction in training dataset size while maintaining a strong predictive baseline accuracy. Specifically, the diverse representative-query by committee (DR-QBC) technique works best yielding ~6.5% Euclidean norm of prediction error (ENPE). We also demonstrate an inverse design framework by leveraging the same surrogate model with differential evolution (DE) and Bayesian optimizer (BO). It mimics the role of a device design engineer by optimizing the values of structural parameters of the LDMOS transistors such that the desired $BV_{\mathrm { DS,off}}$ is attained while minimizing $R_{\mathrm { sp}}$ .
- 14 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45  (3) , 1410-1417
Inverse Design of High Power and High Voltage LDMOS Transistors Using Deep Learning-Based Sample-Efficient Surrogate Model
Device design using Machine Learning has been used in the semiconductor industry over the past ten years. However, the generation of the training data set for precise predictions using this technique remains burdensome. Addressing this, in this work, we propose eight sample-efficient techniques to train the deep neural network (DNN) based surrogate models that emulate technology computer-aided design (TCAD). We showcase their efficacy by predicting off-state breakdown voltage $(BV_{\mathrm { DS,off}})$ and specific on-resistance $(R_{\mathrm { sp}})$ of a laterally diffused metal oxide semiconductor field-effect transistor (LDMOSFET). Our findings highlight the potential for ~38% reduction in training dataset size while maintaining a strong predictive baseline accuracy. Specifically, the diverse representative-query by committee (DR-QBC) technique works best yielding ~6.5% Euclidean norm of prediction error (ENPE). We also demonstrate an inverse design framework by leveraging the same surrogate model with differential evolution (DE) and Bayesian optimizer (BO). It mimics the role of a device design engineer by optimizing the values of structural parameters of the LDMOS transistors such that the desired $BV_{\mathrm { DS,off}}$ is attained while minimizing $R_{\mathrm { sp}}$ .
- 18 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 72  (9) , 4621-4628
Comprehensive Study of Parasitic Capacitance in CFETs: An Analytical Perspective
This work presents a comprehensive study on parasitic capacitance and its corresponding analytical model for complementary field-effect transistor (CFET) devices. The model accounts for various capacitance components, including parallel plate, perpendicular and coplanar plate fringing, junction, separator, and offset capacitances between the gate and source/drain. Individual parasitic capacitance components are isolated using TCAD simulations by adjusting the geometrical and material properties of the device. The fringing capacitance components are modeled using the elliptical integral method, and the model effectively captures the significant contribution of separator capacitance (~20%) to the total parasitic capacitance. With only one fitting parameter, the model demonstrates high accuracy across different device structures. A comparative analysis with lateral nanosheet field-effect transistor (NsFET) devices highlights the impact of the stacked nFET-on-pFET architecture on parasitic capacitance overheads.
- 18 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 72  (9) , 4621-4628
Comprehensive Study of Parasitic Capacitance in CFETs: An Analytical Perspective
This work presents a comprehensive study on parasitic capacitance and its corresponding analytical model for complementary field-effect transistor (CFET) devices. The model accounts for various capacitance components, including parallel plate, perpendicular and coplanar plate fringing, junction, separator, and offset capacitances between the gate and source/drain. Individual parasitic capacitance components are isolated using TCAD simulations by adjusting the geometrical and material properties of the device. The fringing capacitance components are modeled using the elliptical integral method, and the model effectively captures the significant contribution of separator capacitance (~20%) to the total parasitic capacitance. With only one fitting parameter, the model demonstrates high accuracy across different device structures. A comparative analysis with lateral nanosheet field-effect transistor (NsFET) devices highlights the impact of the stacked nFET-on-pFET architecture on parasitic capacitance overheads.
- Open Access
- 10 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of the Electron Devices Society
- Vol. PP, 1
A Scalable Physics-Based Compact Model for Terminal Charge, Intrinsic Capacitance and Drain Current in Nanosheet Field Effect Transistors
This study introduces a physics-based, SPICE-compatible model for Nanosheet Field-Effect Transistors (NsFETs) that offers explicit expressions for the drain current, terminal charges, and intrinsic capacitances applicable to both p-type and n-type devices. The carrier transport is modeled using the drift-diffusion formalism, while the terminal charges are calculated using the Ward-Dutton linear charge partition scheme, ensuring charge conservation. Employing a bottom-up approach, the model effectively captures quantum mechanical confinement-induced effects with minimal reliance on empirical parameters, thus preserving the simplicity characteristic of traditional bulk MOSFET models. Short channel effects are modeled in a self-consistent way. This model has been extensively validated against both experimental data and simulations across varying device dimensions and bias conditions, demonstrating exceptional scalability across all device dimensions. The proposed model has also been implemented in Verilog-A and integrated in a commercial SPICE simulator to simulate NsFETs based circuits, underscoring the model’s practical applicability in contemporary semiconductor design.
- 10 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of the Electron Devices Society
- Vol. PP, 1
Open AccessA Scalable Physics-Based Compact Model for Terminal Charge, Intrinsic Capacitance and Drain Current in Nanosheet Field Effect Transistors
This study introduces a physics-based, SPICE-compatible model for Nanosheet Field-Effect Transistors (NsFETs) that offers explicit expressions for the drain current, terminal charges, and intrinsic capacitances applicable to both p-type and n-type devices. The carrier transport is modeled using the drift-diffusion formalism, while the terminal charges are calculated using the Ward-Dutton linear charge partition scheme, ensuring charge conservation. Employing a bottom-up approach, the model effectively captures quantum mechanical confinement-induced effects with minimal reliance on empirical parameters, thus preserving the simplicity characteristic of traditional bulk MOSFET models. Short channel effects are modeled in a self-consistent way. This model has been extensively validated against both experimental data and simulations across varying device dimensions and bias conditions, demonstrating exceptional scalability across all device dimensions. The proposed model has also been implemented in Verilog-A and integrated in a commercial SPICE simulator to simulate NsFETs based circuits, underscoring the model’s practical applicability in contemporary semiconductor design.
- Open Access
- 13 December 2024
- Published by American Chemical Society (ACS) in ACS Omega
- Vol. 9  (51) , 50634-50641
Exploring p-Type Contact for Monolayer WS2 FETs Using Halogen Doping and Intermediate Layers
This work presents a density functional theory (DFT) study of substitutional and adsorption-based halogen (I or F) doping of WS2-based transistors to enhance their contact properties. Substitutional doping of the WS2 monolayer with halogens results in n-type behavior, while halogen adsorption on the surface of the WS2 monolayer induces p-type behavior. This is attributed to differing directions of charge flow, as supported by the Mulliken analysis. However, due to Fermi-level pinning (FLP) at the WS2-metal interface, the p-type behavior resulting from halogen adsorption is not very prominent. To achieve better p-type contact, intermediate layers of graphene and h-BN are used to mitigate the FLP effect, showing significant improvement. The F-adsorbed WS2-graphene-Pt interface demonstrates excellent p-type contact with a substantial reduction in the hole Schottky barrier height, making it ideal for efficient WS2-based p-type MOS transistors in CMOS technology.
- 13 December 2024
- Published by American Chemical Society (ACS) in ACS Omega
- Vol. 9  (51) , 50634-50641
Open AccessExploring p-Type Contact for Monolayer WS2 FETs Using Halogen Doping and Intermediate Layers
This work presents a density functional theory (DFT) study of substitutional and adsorption-based halogen (I or F) doping of WS2-based transistors to enhance their contact properties. Substitutional doping of the WS2 monolayer with halogens results in n-type behavior, while halogen adsorption on the surface of the WS2 monolayer induces p-type behavior. This is attributed to differing directions of charge flow, as supported by the Mulliken analysis. However, due to Fermi-level pinning (FLP) at the WS2-metal interface, the p-type behavior resulting from halogen adsorption is not very prominent. To achieve better p-type contact, intermediate layers of graphene and h-BN are used to mitigate the FLP effect, showing significant improvement. The F-adsorbed WS2-graphene-Pt interface demonstrates excellent p-type contact with a substantial reduction in the hole Schottky barrier height, making it ideal for efficient WS2-based p-type MOS transistors in CMOS technology.
- 25 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71  (12) , 7889-7895
Training-Free Parameter Extraction for Compact Device Models Using Sequential Bayesian Optimization With Adaptive Sampling
This work presents a computationally efficient approach for extracting compact model parameters with minimal training requirements. Bayesian optimization (BO) is employed in multiple stages to predict the optimum compact model parameters. Initially, the methodology is applied to the MIT virtual source model (MVS 2.0) for extremely thin silicon-on-insulator (ETSOI) devices, nanosheet FETs (NsFETs), and $\text{MoS}_{\text{2}}$ -based 2-D material-based FETs (2DFETs). Subsequently, it is demonstrated on the Berkeley short-channel IGFET model (BSIM) common multigate (CMG) compact model for NsFETs. Through sequential processing, adaptive sampling, successive domain reduction, and fine-tuned objective functions, the framework achieves precise and efficient fitting of both global and local model parameters across a range of devices, all in a reduced number of iterations, irrespective of the compact model used.
- 25 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71  (12) , 7889-7895
Training-Free Parameter Extraction for Compact Device Models Using Sequential Bayesian Optimization With Adaptive Sampling
This work presents a computationally efficient approach for extracting compact model parameters with minimal training requirements. Bayesian optimization (BO) is employed in multiple stages to predict the optimum compact model parameters. Initially, the methodology is applied to the MIT virtual source model (MVS 2.0) for extremely thin silicon-on-insulator (ETSOI) devices, nanosheet FETs (NsFETs), and $\text{MoS}_{\text{2}}$ -based 2-D material-based FETs (2DFETs). Subsequently, it is demonstrated on the Berkeley short-channel IGFET model (BSIM) common multigate (CMG) compact model for NsFETs. Through sequential processing, adaptive sampling, successive domain reduction, and fine-tuned objective functions, the framework achieves precise and efficient fitting of both global and local model parameters across a range of devices, all in a reduced number of iterations, irrespective of the compact model used.
- 30 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71  (11) , 6514-6519
Reliability of Tunneling Regime for Silicon on Insulator-Based Neuron
Low-power operations are essential for implementing large spiking neural networks (SNNs) in real-world applications. An area and energy-efficient demonstration of a functional liquid state machine (LSM) for spoken word detection using the band-to-band-tunneling (BTBT)-based neuron was proposed earlier. For a product-level demonstration of the BTBT regime operating chips, the variability and reliability of neurons emerge as noteworthy concerns for neuromorphic processors due to the potential implications for performance degradation over time. In this work, we characterize and compare the reliability of partially depleted (PD) silicon on insulator (SOI) transistors in the BTBT and on-regime ( ${I}_{\text {ON}}$ ) regimes. The drain current fractional degradation ( ${\Delta }{I}_{\text {D}}/{I}_{\text {D}}$ ) increases with an increase in voltage and thermal stress in the BTBT and ${I}_{\text {ON}}$ regime. The reaction-diffusion-drift (RDD) framework is used to estimate device degradation under operating bias conditions. At operating bias, a ~17% fractional degradation is observed in the BTBT regime operation for ten years, comparable to the degradation observed in the ${I}_{\text {ON}}$ regime. Finally, we analyzed the impact of device degradation on the SNN performance.
- 30 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71  (11) , 6514-6519
Reliability of Tunneling Regime for Silicon on Insulator-Based Neuron
Low-power operations are essential for implementing large spiking neural networks (SNNs) in real-world applications. An area and energy-efficient demonstration of a functional liquid state machine (LSM) for spoken word detection using the band-to-band-tunneling (BTBT)-based neuron was proposed earlier. For a product-level demonstration of the BTBT regime operating chips, the variability and reliability of neurons emerge as noteworthy concerns for neuromorphic processors due to the potential implications for performance degradation over time. In this work, we characterize and compare the reliability of partially depleted (PD) silicon on insulator (SOI) transistors in the BTBT and on-regime ( ${I}_{\text {ON}}$ ) regimes. The drain current fractional degradation ( ${\Delta }{I}_{\text {D}}/{I}_{\text {D}}$ ) increases with an increase in voltage and thermal stress in the BTBT and ${I}_{\text {ON}}$ regime. The reaction-diffusion-drift (RDD) framework is used to estimate device degradation under operating bias conditions. At operating bias, a ~17% fractional degradation is observed in the BTBT regime operation for ten years, comparable to the degradation observed in the ${I}_{\text {ON}}$ regime. Finally, we analyzed the impact of device degradation on the SNN performance.
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71  (9) , 5754-5760
Exploring Sheet Thickness Scaling and Substrate Orientation for Maximizing Nanosheet pFET Performance
We explored the performance of p-type nanosheet FETs (NsFETs) with sheet thickness scaling using a well-calibrated subband BTE solver that accounts for quantum confinement. Our investigation revealed that despite enhancements in gate electrostatics, the confined pFETs exhibit significantly reduced hole mobility due to increased phonon and surface roughness scattering (SRS). It is also found that introducing uniaxial compressive stress into the channel with an ideally flat surface (very low surface roughness) could boost the pFET on-current by approximately 2.5 times. Furthermore, by integrating p-type NsFETs on $\{ 110\}$ substrate, rather than on conventional $\{ 100\}$ substrate, it is likely to yield superior hole mobility and overall device performance, particularly at scaled sheet thicknesses.
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71  (9) , 5754-5760
Exploring Sheet Thickness Scaling and Substrate Orientation for Maximizing Nanosheet pFET Performance
We explored the performance of p-type nanosheet FETs (NsFETs) with sheet thickness scaling using a well-calibrated subband BTE solver that accounts for quantum confinement. Our investigation revealed that despite enhancements in gate electrostatics, the confined pFETs exhibit significantly reduced hole mobility due to increased phonon and surface roughness scattering (SRS). It is also found that introducing uniaxial compressive stress into the channel with an ideally flat surface (very low surface roughness) could boost the pFET on-current by approximately 2.5 times. Furthermore, by integrating p-type NsFETs on $\{ 110\}$ substrate, rather than on conventional $\{ 100\}$ substrate, it is likely to yield superior hole mobility and overall device performance, particularly at scaled sheet thicknesses.