Showing 1–50 of 60 results for author: Portal, J

Versatile CMOS Analog LIF Neuron for Memristor-Integrated Neuromorphic Circuits

Authors: Nikhil Garg, Davide Florini, Patrick Dufour, Eloir Muhr, Mathieu Faye, Marc Bocquet, Damien Querlioz, Yann Beilliard, Dominique Drouin, Fabien Alibart, Jean-Michel Portal

Abstract: Heterogeneous systems with analog CMOS circuits integrated with nanoscale memristive devices enable efficient deployment of neural networks on neuromorphic hardware. CMOS Neuron with low footprint can emulate slow temporal dynamics by operating with extremely low current levels. Nevertheless, the current read from the memristive synapses can be higher by several orders of magnitude, and performing… ▽ More Heterogeneous systems with analog CMOS circuits integrated with nanoscale memristive devices enable efficient deployment of neural networks on neuromorphic hardware. CMOS Neuron with low footprint can emulate slow temporal dynamics by operating with extremely low current levels. Nevertheless, the current read from the memristive synapses can be higher by several orders of magnitude, and performing impedance matching between neurons and synapses is mandatory. In this paper, we implement an analog leaky integrate and fire (LIF) neuron with a voltage regulator and current attenuator for interfacing CMOS neurons with memristive synapses. In addition, the neuron design proposes a dual leakage that could enable the implementation of local learning rules such as voltage-dependent synaptic plasticity. We also propose a connection scheme to implement adaptive LIF neurons based on two-neuron interaction. The proposed circuits can be used to interface with a variety of synaptic devices and process signals of diverse temporal dynamics. △ Less

Submitted 28 June, 2024; originally announced June 2024.

Comments: Accepted to International Conference on Neuromorphic Systems (ICONS 2024)

The Logarithmic Memristor-Based Bayesian Machine

Authors: Clément Turck, Kamel-Eddine Harabi, Adrien Pontlevy, Théo Ballet, Tifenn Hirtzlin, Elisa Vianello, Raphaël Laurent, Jacques Droulez, Pierre Bessière, Marc Bocquet, Jean-Michel Portal, Damien Querlioz

Abstract: The demand for explainable and energy-efficient artificial intelligence (AI) systems for edge computing has led to significant interest in electronic systems dedicated to Bayesian inference. Traditional designs of such systems often rely on stochastic computing, which offers high energy efficiency but suffers from latency issues and struggles with low-probability values. In this paper, we introduc… ▽ More The demand for explainable and energy-efficient artificial intelligence (AI) systems for edge computing has led to significant interest in electronic systems dedicated to Bayesian inference. Traditional designs of such systems often rely on stochastic computing, which offers high energy efficiency but suffers from latency issues and struggles with low-probability values. In this paper, we introduce the logarithmic memristor-based Bayesian machine, an innovative design that leverages the unique properties of memristors and logarithmic computing as an alternative to stochastic computing. We present a prototype machine fabricated in a hybrid CMOS/hafnium-oxide memristor process. We validate the versatility and robustness of our system through experimental validation and extensive simulations in two distinct applications: gesture recognition and sleep stage classification. The logarithmic approach simplifies the computational model by converting multiplications into additions and enhances the handling of low-probability events, which are crucial in time-dependent tasks. Our results demonstrate that the logarithmic Bayesian machine achieves superior performance in terms of accuracy and energy efficiency compared to its stochastic counterpart, particularly in scenarios involving complex probabilistic models. This work paves the way for the deployment of advanced AI capabilities in edge devices, where power efficiency and reliability are paramount. △ Less

Submitted 5 June, 2024; originally announced June 2024.

Powering AI at the Edge: A Robust, Memristor-based Binarized Neural Network with Near-Memory Computing and Miniaturized Solar Cell

Authors: Fadi Jebali, Atreya Majumdar, Clément Turck, Kamel-Eddine Harabi, Mathieu-Coumba Faye, Eloi Muhr, Jean-Pierre Walder, Oleksandr Bilousov, Amadeo Michaud, Elisa Vianello, Tifenn Hirtzlin, François Andrieu, Marc Bocquet, Stéphane Collin, Damien Querlioz, Jean-Michel Portal

Abstract: Memristor-based neural networks provide an exceptional energy-efficient platform for artificial intelligence (AI), presenting the possibility of self-powered operation when paired with energy harvesters. However, most memristor-based networks rely on analog in-memory computing, necessitating a stable and precise power supply, which is incompatible with the inherently unstable and unreliable energy… ▽ More Memristor-based neural networks provide an exceptional energy-efficient platform for artificial intelligence (AI), presenting the possibility of self-powered operation when paired with energy harvesters. However, most memristor-based networks rely on analog in-memory computing, necessitating a stable and precise power supply, which is incompatible with the inherently unstable and unreliable energy harvesters. In this work, we fabricated a robust binarized neural network comprising 32,768 memristors, powered by a miniature wide-bandgap solar cell optimized for edge applications. Our circuit employs a resilient digital near-memory computing approach, featuring complementarily programmed memristors and logic-in-sense-amplifier. This design eliminates the need for compensation or calibration, operating effectively under diverse conditions. Under high illumination, the circuit achieves inference performance comparable to that of a lab bench power supply. In low illumination scenarios, it remains functional with slightly reduced accuracy, seamlessly transitioning to an approximate computing mode. Through image classification neural network simulations, we demonstrate that misclassified images under low illumination are primarily difficult-to-classify cases. Our approach lays the groundwork for self-powered AI and the creation of intelligent sensors for various applications in health, safety, and environment monitoring. △ Less

Submitted 22 May, 2023; originally announced May 2023.

A Multimode Hybrid Memristor-CMOS Prototyping Platform Supporting Digital and Analog Projects

Authors: Kamel-Eddine Harabi, Clement Turck, Marie Drouhin, Adrien Renaudineau, Thomas Bersani--Veroni, Damien Querlioz, Tifenn Hirtzlin, Elisa Vianello, Marc Bocquet, Jean-Michel Portal

Abstract: We present an integrated circuit fabricated in a process co-integrating CMOS and hafnium-oxide memristor technology, which provides a prototyping platform for projects involving memristors. Our circuit includes the periphery circuitry for using memristors within digital circuits, as well as an analog mode with direct access to memristors. The platform allows optimizing the conditions for reading a… ▽ More We present an integrated circuit fabricated in a process co-integrating CMOS and hafnium-oxide memristor technology, which provides a prototyping platform for projects involving memristors. Our circuit includes the periphery circuitry for using memristors within digital circuits, as well as an analog mode with direct access to memristors. The platform allows optimizing the conditions for reading and writing memristors, as well as developing and testing innovative memristor-based neuromorphic concepts. △ Less

Submitted 28 February, 2023; originally announced February 2023.

Voltage-Dependent Synaptic Plasticity (VDSP): Unsupervised probabilistic Hebbian plasticity rule based on neurons membrane potential

Authors: Nikhil Garg, Ismael Balafrej, Terrence C. Stewart, Jean Michel Portal, Marc Bocquet, Damien Querlioz, Dominique Drouin, Jean Rouat, Yann Beilliard, Fabien Alibart

Abstract: This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb's plasticity mechanism on neuromorphic hardware. The proposed VDSP learning rule updates the synaptic conductance on the spike of the postsynaptic neuron only, which reduces by a factor of two the number of updates with respect to standard… ▽ More This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb's plasticity mechanism on neuromorphic hardware. The proposed VDSP learning rule updates the synaptic conductance on the spike of the postsynaptic neuron only, which reduces by a factor of two the number of updates with respect to standard spike-timing-dependent plasticity (STDP). This update is dependent on the membrane potential of the presynaptic neuron, which is readily available as part of neuron implementation and hence does not require additional memory for storage. Moreover, the update is also regularized on synaptic weight and prevents explosion or vanishing of weights on repeated stimulation. Rigorous mathematical analysis is performed to draw an equivalence between VDSP and STDP. To validate the system-level performance of VDSP, we train a single-layer spiking neural network (SNN) for the recognition of handwritten digits. We report 85.01 $ \pm $ 0.76% (Mean $ \pm $ S.D.) accuracy for a network of 100 output neurons on the MNIST dataset. The performance improves when scaling the network size (89.93 $ \pm $ 0.41% for 400 output neurons, 90.56 $ \pm $ 0.27 for 500 neurons), which validates the applicability of the proposed learning rule for spatial pattern recognition tasks. Future work will consider more complicated tasks. Interestingly, the learning rule better adapts than STDP to the frequency of input signal and does not require hand-tuning of hyperparameters △ Less

Submitted 22 October, 2022; v1 submitted 21 March, 2022; originally announced March 2022.

Comments: Front. Neurosci., 21 October 2022 Sec. Neuromorphic Engineering

Journal ref: Front. Neurosci. 16:983950 (2022)

Experimental demonstration of Single-Level and Multi-Level-Cell RRAM-based In-Memory Computing with up to 16 parallel operations

Authors: E. Esmanhotto, T. Hirtzlin, N. Castellani, S. Martin, B. Giraud, F. Andrieu, J. F. Nodin, D. Querlioz, J-M. Portal, E. Vianello

Abstract: Crossbar arrays of resistive memories (RRAM) hold the promise of enabling In-Memory Computing (IMC), but essential challenges due to the impact of device imperfection and device endurance have yet to be overcome. In this work, we demonstrate experimentally an RRAM-based IMC logic concept with strong resilience to RRAM variability, even after one million endurance cycles. Our work relies on a gener… ▽ More Crossbar arrays of resistive memories (RRAM) hold the promise of enabling In-Memory Computing (IMC), but essential challenges due to the impact of device imperfection and device endurance have yet to be overcome. In this work, we demonstrate experimentally an RRAM-based IMC logic concept with strong resilience to RRAM variability, even after one million endurance cycles. Our work relies on a generalization of the concept of in-memory Scouting Logic, and we demonstrate it experimentally with up to 16 parallel devices (operands), a new milestone for RRAM in-memory logic. Moreover, we combine IMC with Multi-Level-Cell programming and demonstrate experimentally, for the first time, an IMC RRAM-based MLC 2-bit adder. △ Less

Submitted 3 March, 2022; originally announced March 2022.

Comments: Preprint for IRPS2022

A Memristor-Based Bayesian Machine

Authors: Kamel-Eddine Harabi, Tifenn Hirtzlin, Clément Turck, Elisa Vianello, Raphaël Laurent, Jacques Droulez, Pierre Bessière, Jean-Michel Portal, Marc Bocquet, Damien Querlioz

Abstract: In recent years, a considerable research effort has shown the energy benefits of implementing neural networks with memristors or other emerging memory technologies. However, for extreme-edge applications with high uncertainty, access to reduced amounts of data, and where explainable decisions are required, neural networks may not provide an acceptable form of intelligence. Bayesian reasoning can s… ▽ More In recent years, a considerable research effort has shown the energy benefits of implementing neural networks with memristors or other emerging memory technologies. However, for extreme-edge applications with high uncertainty, access to reduced amounts of data, and where explainable decisions are required, neural networks may not provide an acceptable form of intelligence. Bayesian reasoning can solve these concerns, but it is computationally expensive and, unlike neural networks, does not translate naturally to memristor-based architectures. In this work, we introduce, demonstrate experimentally on a fully fabricated hybrid CMOS-memristor system, and analyze a Bayesian machine designed for highly-energy efficient Bayesian reasoning. The architecture of the machine is obtained by writing Bayes' law in a way making its implementation natural by the principles of distributed memory and stochastic computing, allowing the circuit to function using solely local memory and minimal data movement. Measurements on a fabricated small-scale Bayesian machine featuring 2,048 memristors and 30,080 transistors show the viability of this approach and the possibility of overcoming the challenges associated with its design: the inherent imperfections of memristors, as well as the need to distribute very locally higher-than-nominal supply voltages. The design of a scaled-up version of the machine shows its outstanding energy efficiency on a real-life gesture recognition task: a gesture can be recognized using 5,000 times less energy than using a microcontroller unit. The Bayesian machine also features several desirable features, e.g., instant on/off operation, compatibility with low supply voltages, and resilience to single-event upsets. These results open the road for Bayesian reasoning as an attractive way for energy-efficient, robust, and explainable intelligence at the edge. △ Less

Submitted 20 December, 2021; originally announced December 2021.

Model of the Weak Reset Process in HfOx Resistive Memory for Deep Learning Frameworks

Authors: Atreya Majumdar, Marc Bocquet, Tifenn Hirtzlin, Axel Laborieux, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: The implementation of current deep learning training algorithms is power-hungry, owing to data transfer between memory and logic units. Oxide-based RRAMs are outstanding candidates to implement in-memory computing, which is less power-intensive. Their weak RESET regime, is particularly attractive for learning, as it allows tuning the resistance of the devices with remarkable endurance. However, th… ▽ More The implementation of current deep learning training algorithms is power-hungry, owing to data transfer between memory and logic units. Oxide-based RRAMs are outstanding candidates to implement in-memory computing, which is less power-intensive. Their weak RESET regime, is particularly attractive for learning, as it allows tuning the resistance of the devices with remarkable endurance. However, the resistive change behavior in this regime suffers many fluctuations and is particularly challenging to model, especially in a way compatible with tools used for simulating deep learning. In this work, we present a model of the weak RESET process in hafnium oxide RRAM and integrate this model within the PyTorch deep learning framework. Validated on experiments on a hybrid CMOS/RRAM technology, our model reproduces both the noisy progressive behavior and the device-to-device (D2D) variability. We use this tool to train Binarized Neural Networks for the MNIST handwritten digit recognition task and the CIFAR-10 object classification task. We simulate our model with and without various aspects of device imperfections to understand their impact on the training process and identify that the D2D variability is the most detrimental aspect. The framework can be used in the same manner for other types of memories to identify the device imperfections that cause the most degradation, which can, in turn, be used to optimize the devices to reduce the impact of these imperfections. △ Less

Submitted 2 September, 2021; v1 submitted 2 July, 2021; originally announced July 2021.

Ultra-High-density 3D vertical RRAM with stacked JunctionLess nanowires for In-Memory-Computing applications

Authors: M. Ezzadeen, D. Bosch, B. Giraud, S. Barraud, J. -P. Noel, D. Lattard, J. Lacord, J. -M. Portal, F. Andrieu

Abstract: The Von-Neumann bottleneck is a clear limitation for data-intensive applications, bringing in-memory computing (IMC) solutions to the fore. Since large data sets are usually stored in nonvolatile memory (NVM), various solutions have been proposed based on emerging memories, such as OxRAM, that rely mainly on area hungry, one transistor (1T) one OxRAM (1R) bit-cell. To tackle this area issue, while… ▽ More The Von-Neumann bottleneck is a clear limitation for data-intensive applications, bringing in-memory computing (IMC) solutions to the fore. Since large data sets are usually stored in nonvolatile memory (NVM), various solutions have been proposed based on emerging memories, such as OxRAM, that rely mainly on area hungry, one transistor (1T) one OxRAM (1R) bit-cell. To tackle this area issue, while keeping the programming control provided by 1T1R bit-cell, we propose to combine gate-all-around stacked junctionless nanowires (1JL) and OxRAM (1R) technology to create a 3-D memory pillar with ultrahigh density. Nanowire junctionless transistors have been fabricated, characterized, and simulated to define current conditions for the whole pillar. Finally, based on Simulation Program with Integrated Circuit Emphasis (SPICE) simulations, we demonstrated successfully scouting logic operations up to three-pillar layers, with one operand per layer. △ Less

Submitted 30 November, 2020; originally announced December 2020.

Implementation of Ternary Weights with Resistive RAM Using a Single Sense Operation per Synapse

Authors: Axel Laborieux, Marc Bocquet, Tifenn Hirtzlin, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a significant lead for reducing the energy consumption of artificial intelligence. To achieve maximum energy efficiency in such systems, logic and memory should be integrated as tightly as possible. In this work, we focus on the case of ternary neural networks, w… ▽ More The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a significant lead for reducing the energy consumption of artificial intelligence. To achieve maximum energy efficiency in such systems, logic and memory should be integrated as tightly as possible. In this work, we focus on the case of ternary neural networks, where synaptic weights assume ternary values. We propose a two-transistor/two-resistor memory architecture employing a precharge sense amplifier, where the weight value can be extracted in a single sense operation. Based on experimental measurements on a hybrid 130 nm CMOS/RRAM chip featuring this sense amplifier, we show that this technique is particularly appropriate at low supply voltage, and that it is resilient to process, voltage, and temperature variations. We characterize the bit error rate in our scheme. We show based on neural network simulation on the CIFAR-10 image recognition task that the use of ternary neural networks significantly increases neural network performance, with regards to binary ones, which are often preferred for inference hardware. We finally evidence that the neural network is immune to the type of bit errors observed in our scheme, which can therefore be used without error correction. △ Less

Submitted 14 October, 2020; v1 submitted 26 July, 2020; originally announced July 2020.

Comments: arXiv admin note: substantial text overlap with arXiv:2005.01973

Embracing the Unreliability of Memory Devices for Neuromorphic Computing

Authors: Marc Bocquet, Tifenn Hirtzlin, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: The emergence of resistive non-volatile memories opens the way to highly energy-efficient computation near- or in-memory. However, this type of computation is not compatible with conventional ECC, and has to deal with device unreliability. Inspired by the architecture of animal brains, we present a manufactured differential hybrid CMOS/RRAM memory architecture suitable for neural network implement… ▽ More The emergence of resistive non-volatile memories opens the way to highly energy-efficient computation near- or in-memory. However, this type of computation is not compatible with conventional ECC, and has to deal with device unreliability. Inspired by the architecture of animal brains, we present a manufactured differential hybrid CMOS/RRAM memory architecture suitable for neural network implementation that functions without formal ECC. We also show that using low-energy but error-prone programming conditions only slightly reduces network accuracy. △ Less

Submitted 13 July, 2020; originally announced July 2020.

In-Memory Resistive RAM Implementation of Binarized Neural Networks for Medical Applications

Authors: Bogdan Penkovsky, Marc Bocquet, Tifenn Hirtzlin, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: The advent of deep learning has considerably accelerated machine learning development. The deployment of deep neural networks at the edge is however limited by their high memory and energy consumption requirements. With new memory technology available, emerging Binarized Neural Networks (BNNs) are promising to reduce the energy impact of the forthcoming machine learning hardware generation, enabli… ▽ More The advent of deep learning has considerably accelerated machine learning development. The deployment of deep neural networks at the edge is however limited by their high memory and energy consumption requirements. With new memory technology available, emerging Binarized Neural Networks (BNNs) are promising to reduce the energy impact of the forthcoming machine learning hardware generation, enabling machine learning on the edge devices and avoiding data transfer over the network. In this work, after presenting our implementation employing a hybrid CMOS - hafnium oxide resistive memory technology, we suggest strategies to apply BNNs to biomedical signals such as electrocardiography and electroencephalography, keeping accuracy level and reducing memory requirements. We investigate the memory-accuracy trade-off when binarizing whole network and binarizing solely the classifier part. We also discuss how these results translate to the edge-oriented Mobilenet~V1 neural network on the Imagenet task. The final goal of this research is to enable smart autonomous healthcare devices. △ Less

Submitted 20 June, 2020; originally announced June 2020.

Low Power In-Memory Implementation of Ternary Neural Networks with Resistive RAM-Based Synapse

Authors: Axel Laborieux, Marc Bocquet, Tifenn Hirtzlin, Jacques-Olivier Klein, Liza Herrera Diez, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a major lead for reducing the energy consumption of artificial intelligence (AI). Multiple works have for example proposed in-memory architectures to implement low power binarized neural networks. These simple neural networks, where synaptic weights and neuronal… ▽ More The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a major lead for reducing the energy consumption of artificial intelligence (AI). Multiple works have for example proposed in-memory architectures to implement low power binarized neural networks. These simple neural networks, where synaptic weights and neuronal activations assume binary values, can indeed approach state-of-the-art performance on vision tasks. In this work, we revisit one of these architectures where synapses are implemented in a differential fashion to reduce bit errors, and synaptic weights are read using precharge sense amplifiers. Based on experimental measurements on a hybrid 130 nm CMOS/RRAM chip and on circuit simulation, we show that the same memory array architecture can be used to implement ternary weights instead of binary weights, and that this technique is particularly appropriate if the sense amplifier is operated in near-threshold regime. We also show based on neural network simulation on the CIFAR-10 image recognition task that going from binary to ternary neural networks significantly increases neural network performance. These results highlight that AI circuits function may sometimes be revisited when operated in low power regimes. △ Less

Submitted 5 May, 2020; originally announced May 2020.

Implementing Binarized Neural Networks with Magnetoresistive RAM without Error Correction

Authors: Tifenn Hirtzlin, Bogdan Penkovsky, Jacques-Olivier Klein, Nicolas Locatelli, Adrien F. Vincent, Marc Bocquet, Jean-Michel Portal, Damien Querlioz

Abstract: One of the most exciting applications of Spin Torque Magnetoresistive Random Access Memory (ST-MRAM) is the in-memory implementation of deep neural networks, which could allow improving the energy efficiency of Artificial Intelligence by orders of magnitude with regards to its implementation on computers and graphics cards. In particular, ST-MRAM could be ideal for implementing Binarized Neural Ne… ▽ More One of the most exciting applications of Spin Torque Magnetoresistive Random Access Memory (ST-MRAM) is the in-memory implementation of deep neural networks, which could allow improving the energy efficiency of Artificial Intelligence by orders of magnitude with regards to its implementation on computers and graphics cards. In particular, ST-MRAM could be ideal for implementing Binarized Neural Networks (BNNs), a type of deep neural networks discovered in 2016, which can achieve state-of-the-art performance with a highly reduced memory footprint with regards to conventional artificial intelligence approaches. The challenge of ST-MRAM, however, is that it is prone to write errors and usually requires the use of error correction. In this work, we show that these bit errors can be tolerated by BNNs to an outstanding level, based on examples of image recognition tasks (MNIST, CIFAR-10 and ImageNet): bit error rates of ST-MRAM up to 0.1% have little impact on recognition accuracy. The requirements for ST-MRAM are therefore considerably relaxed for BNNs with regards to traditional applications. By consequence, we show that for BNNs, ST-MRAMs can be programmed with weak (low-energy) programming conditions, without error correcting codes. We show that this result can allow the use of low energy and low area ST-MRAM cells, and show that the energy savings at the system level can reach a factor two. △ Less

Submitted 12 August, 2019; originally announced August 2019.

Digital Biologically Plausible Implementation of Binarized Neural Networks with Differential Hafnium Oxide Resistive Memory Arrays

Authors: Tifenn Hirtzlin, Marc Bocquet, Bogdan Penkovsky, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: The brain performs intelligent tasks with extremely low energy consumption. This work takes inspiration from two strategies used by the brain to achieve this energy efficiency: the absence of separation between computing and memory functions, and the reliance on low precision computation. The emergence of resistive memory technologies indeed provides an opportunity to co-integrate tightly logic an… ▽ More The brain performs intelligent tasks with extremely low energy consumption. This work takes inspiration from two strategies used by the brain to achieve this energy efficiency: the absence of separation between computing and memory functions, and the reliance on low precision computation. The emergence of resistive memory technologies indeed provides an opportunity to co-integrate tightly logic and memory in hardware. In parallel, the recently proposed concept of Binarized Neural Network, where multiplications are replaced by exclusive NOR (XNOR) logic gates, offers a way to implement artificial intelligence using very low precision computation. In this work, we therefore propose a strategy to implement low energy Binarized Neural Networks, which employs brain-inspired concepts, while retaining energy benefits from digital electronics. We design, fabricate and test a memory array, including periphery and sensing circuits, optimized for this in-memory computing scheme. Our circuit employs hafnium oxide resistive memory integrated in the back end of line of a 130 nanometer CMOS process, in a two transistors - two resistors cell, which allows performing the exclusive NOR operations of the neural network directly within the sense amplifiers. We show, based on extensive electrical measurements, that our design allows reducing the amount of bit errors on the synaptic weights, without the use of formal error correcting codes. We design a whole system using this memory array. We show on standard machine learning tasks (MNIST, CIFAR-10, ImageNet and an ECG task) that the system has an inherent resilience to bit errors. We evidence that its energy consumption is attractive compared to more standard approaches, and that it can use the memory devices in regimes where they exhibit particularly low programming energy and high endurance. △ Less

Submitted 7 December, 2019; v1 submitted 12 August, 2019; originally announced August 2019.

Stochastic Computing for Hardware Implementation of Binarized Neural Networks

Authors: Tifenn Hirtzlin, Bogdan Penkovsky, Marc Bocquet, Jacques-Olivier Klein, Jean-Michel Portal, Damien Querlioz

Abstract: Binarized Neural Networks, a recently discovered class of neural networks with minimal memory requirements and no reliance on multiplication, are a fantastic opportunity for the realization of compact and energy efficient inference hardware. However, such neural networks are generally not entirely binarized: their first layer remains with fixed point input. In this work, we propose a stochastic co… ▽ More Binarized Neural Networks, a recently discovered class of neural networks with minimal memory requirements and no reliance on multiplication, are a fantastic opportunity for the realization of compact and energy efficient inference hardware. However, such neural networks are generally not entirely binarized: their first layer remains with fixed point input. In this work, we propose a stochastic computing version of Binarized Neural Networks, where the input is also binarized. Simulations on the example of the Fashion-MNIST and CIFAR-10 datasets show that such networks can approach the performance of conventional Binarized Neural Networks. We evidence that the training procedure should be adapted for use with stochastic computing. Finally, the ASIC implementation of our scheme is investigated, in a system that closely associates logic and memory, implemented by Spin Torque Magnetoresistive Random Access Memory. This analysis shows that the stochastic computing approach can allow considerable savings with regards to conventional Binarized Neural networks in terms of area (62% area reduction on the Fashion-MNIST task). It can also allow important savings in terms of energy consumption, if we accept reasonable reduction of accuracy: for example a factor 2.1 can be saved, with the cost of 1.4% in Fashion-MNIST test accuracy. These results highlight the high potential of Binarized Neural Networks for hardware implementation, and that adapting them to hardware constrains can provide important benefits. △ Less

Submitted 3 June, 2019; originally announced June 2019.

Outstanding Bit Error Tolerance of Resistive RAM-Based Binarized Neural Networks

Authors: Tifenn Hirtzlin, Marc Bocquet, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: Resistive random access memories (RRAM) are novel nonvolatile memory technologies, which can be embedded at the core of CMOS, and which could be ideal for the in-memory implementation of deep neural networks. A particularly exciting vision is using them for implementing Binarized Neural Networks (BNNs), a class of deep neural networks with a highly reduced memory footprint. The challenge of resist… ▽ More Resistive random access memories (RRAM) are novel nonvolatile memory technologies, which can be embedded at the core of CMOS, and which could be ideal for the in-memory implementation of deep neural networks. A particularly exciting vision is using them for implementing Binarized Neural Networks (BNNs), a class of deep neural networks with a highly reduced memory footprint. The challenge of resistive memory, however, is that they are prone to device variation, which can lead to bit errors. In this work we show that BNNs can tolerate these bit errors to an outstanding level, through simulations of networks on the MNIST and CIFAR10 tasks. If a standard BNN is used, up to 10^-4 bit error rate can be tolerated with little impact on recognition performance on both MNIST and CIFAR10. We then show that by adapting the training procedure to the fact that the BNN will be operated on error-prone hardware, this tolerance can be extended to a bit error rate of 4x10^-2. The requirements for RRAM are therefore a lot less stringent for BNNs than more traditional applications. We show, based on experimental measurements on a RRAM HfO2 technology, that this result can allow reduce RRAM programming energy by a factor 30. △ Less

Submitted 7 April, 2019; originally announced April 2019.

In-Memory and Error-Immune Differential RRAM Implementation of Binarized Deep Neural Networks

Authors: Marc Bocquet, Tifenn Hirztlin, Jacques-Olivier Klein, Etienne Nowak, Elisa Vianello, Jean-Michel Portal, Damien Querlioz

Abstract: RRAM-based in-Memory Computing is an exciting road for implementing highly energy efficient neural networks. This vision is however challenged by RRAM variability, as the efficient implementation of in-memory computing does not allow error correction. In this work, we fabricated and tested a differential HfO2-based memory structure and its associated sense circuitry, which are ideal for in-memory… ▽ More RRAM-based in-Memory Computing is an exciting road for implementing highly energy efficient neural networks. This vision is however challenged by RRAM variability, as the efficient implementation of in-memory computing does not allow error correction. In this work, we fabricated and tested a differential HfO2-based memory structure and its associated sense circuitry, which are ideal for in-memory computing. For the first time, we show that our approach achieves the same reliability benefits as error correction, but without any CMOS overhead. We show, also for the first time, that it can naturally implement Binarized Deep Neural Networks, a very recent development of Artificial Intelligence, with extreme energy efficiency, and that the system is fully satisfactory for image recognition applications. Finally, we evidence how the extra reliability provided by the differential memory allows programming the devices in low voltage conditions, where they feature high endurance of billions of cycles. △ Less

Submitted 7 February, 2019; originally announced February 2019.

Fine structure of phonon replicas in a tunnel spectrum of a GaAs quantum well

Authors: V. G. Krishtop, V. G. Popov, M. Henini, Yu. Krupko, J. -C. Portal

Abstract: A fine structure of phonon replicas in the current-voltage characteristic of a resonant-tunneling diode has been investigated experimentally. A detailed study of the diode I-V curves in magnetic fields of different orientations has allowed to determine the origin of the features in the fine structure. The voltage positions of the features are shown to coincide with calculated that in the frame of… ▽ More A fine structure of phonon replicas in the current-voltage characteristic of a resonant-tunneling diode has been investigated experimentally. A detailed study of the diode I-V curves in magnetic fields of different orientations has allowed to determine the origin of the features in the fine structure. The voltage positions of the features are shown to coincide with calculated that in the frame of two models: LO-phonon assisted tunneling and resonant tunneling of polarons. △ Less

Submitted 2 February, 2014; originally announced February 2014.

Comments: 6 pages, 7 figures

The resistance of 2D Topological insulator in the absence of the quantized transport

Authors: G. M. Gusev, Z. D. Kvon, E. B Olshanetsky, A. D. Levin, Y. Krupko, J. C. Portal, N. N. Mikhailov, S. A. Dvoretsky

Abstract: We report unconventional transport properties of HgTe wells with inverted band structure: the resistance does not show insulating behavior even when it is of the order of $10^2\times h/2e^{2}$. The system is expected to be a two-dimensional topological insulator with a dominant edge state contribution. The results are inconsistent with theoretical models developed within the framework of the helic… ▽ More We report unconventional transport properties of HgTe wells with inverted band structure: the resistance does not show insulating behavior even when it is of the order of $10^2\times h/2e^{2}$. The system is expected to be a two-dimensional topological insulator with a dominant edge state contribution. The results are inconsistent with theoretical models developed within the framework of the helical Luttinger liquid. △ Less

Submitted 20 August, 2013; originally announced August 2013.

Comments: 5 pages, 4 figures

Unconventional Hall effect near charge neutrality point in a two-dimensional electron-hole system

Authors: O. E. Raichev, G. M. Gusev, E. B. Olshanetsky, Z. D. Kvon, N. N. Mikhailov, S. A. Dvoretsky, J. C. Portal

Abstract: The transport properties of the two-dimensional system in HgTe-based quantum wells containing simultaneously electrons and holes of low densities are examined. The Hall resistance, as a function of perpendicular magnetic field, reveals an unconventional behavior, different from the classical N-shaped dependence typical for bipolar systems with electron-hole asymmetry. The quantum features of magne… ▽ More The transport properties of the two-dimensional system in HgTe-based quantum wells containing simultaneously electrons and holes of low densities are examined. The Hall resistance, as a function of perpendicular magnetic field, reveals an unconventional behavior, different from the classical N-shaped dependence typical for bipolar systems with electron-hole asymmetry. The quantum features of magnetotransport are explained by means of numerical calculation of the Landau level spectrum based on the Kane Hamiltonian. The origin of the quantum Hall plateau σxy = 0 near the charge neutrality point is attributed to special features of Landau quantization in our system. △ Less

Submitted 26 October, 2012; originally announced October 2012.

Comments: 8 pages, 7 figures

Journal ref: Physical Review B 86, 155320 (2012)

Non-linear transport phenomena in a two-subband system

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: We study non-linear transport phenomena in a high-mobility bilayer system with two closely spaced populated electronic subbands in a perpendicular magnetic field. For a moderate direct current excitation, we observe zero-differential-resistance states with a characteristic 1/B periodicity. We investigate, both experimentally and theoretically, the Hall field-induced resistance oscillations which m… ▽ More We study non-linear transport phenomena in a high-mobility bilayer system with two closely spaced populated electronic subbands in a perpendicular magnetic field. For a moderate direct current excitation, we observe zero-differential-resistance states with a characteristic 1/B periodicity. We investigate, both experimentally and theoretically, the Hall field-induced resistance oscillations which modulate the high-frequency magneto-intersubband oscillations in our system if we increase the current. We also observe and describe the influence of direct current on the magnetoresistance in the presence of microwave irradiation. △ Less

Submitted 10 October, 2011; originally announced October 2011.

Comments: 8 pages, 6 figures

Journal ref: Phys. Rev. B 84, 165303 (2011)

Transport in disordered two-dimensional topological insulator

Authors: G. M. Gusev, Z. D. Kvon, O. A. Shegai, N. N. Mikhailov, S. A. Dvoretsky, J. C. Portal

Abstract: We study experimentally the transport properties of "inverted" semiconductor HgTe-based quantum well, which is related to the two-dimensional topological insulator, in diffusive transport regime. We perform nonlocal electrical measurements in the absence of the magnetic field and observe large signal due to the edge states. It demonstrates, that the edge states can propagate over long distance 1… ▽ More We study experimentally the transport properties of "inverted" semiconductor HgTe-based quantum well, which is related to the two-dimensional topological insulator, in diffusive transport regime. We perform nonlocal electrical measurements in the absence of the magnetic field and observe large signal due to the edge states. It demonstrates, that the edge states can propagate over long distance 1 mm, and, therefore, there is no difference between local and non local electrical measurements in topological insulator. In the presence of the in-plane magnetic field we find strong decrease of the local resistance and complete suppression of the nonlocal resistance. We attribute this observation to the transition between topological insulator and bulk metal induced by the in-plane magnetic field. △ Less

Submitted 9 June, 2011; originally announced June 2011.

Comments: 4.5 pages, 4 figures

Microwave-induced Hall resistance in bilayer electron systems

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, S. Krämer, A. K. Bakarov, J. C. Portal

Abstract: The influence of microwave irradiation on dissipative and Hall resistance in high-quality bilayer electron systems is investigated experimentally. We observe a deviation from odd symmetry under magnetic field reversal in the microwave-induced Hall resistance $ΔR_{xy}$ whereas the dissipative resistance $ΔR_{xx}$ obeys even symmetry. Studies of $ΔR_{xy}$ as a function of the microwave electric fiel… ▽ More The influence of microwave irradiation on dissipative and Hall resistance in high-quality bilayer electron systems is investigated experimentally. We observe a deviation from odd symmetry under magnetic field reversal in the microwave-induced Hall resistance $ΔR_{xy}$ whereas the dissipative resistance $ΔR_{xx}$ obeys even symmetry. Studies of $ΔR_{xy}$ as a function of the microwave electric field and polarization exhibit a strong and non-trivial power and polarization dependence. The obtained results are discussed in connection to existing theoretical models of microwave-induced photoconductivity. △ Less

Submitted 17 May, 2011; originally announced May 2011.

Comments: 8 pages, 5 figures

Journal ref: Physical Review B 83, 195317 (2011)

Evidence for zero-differential resistance states in electronic bilayers

Authors: G. M. Gusev, S. Wiedmann, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: We observe zero-differential resistance states at low temperatures and moderate direct currents in a bilayer electron system formed by a wide quantum well. Several regions of vanishing resistance evolve from the inverted peaks of magneto-intersubband oscillations as the current increases. The experiment, supported by a theoretical analysis, suggests that the origin of this phenomenon is based on i… ▽ More We observe zero-differential resistance states at low temperatures and moderate direct currents in a bilayer electron system formed by a wide quantum well. Several regions of vanishing resistance evolve from the inverted peaks of magneto-intersubband oscillations as the current increases. The experiment, supported by a theoretical analysis, suggests that the origin of this phenomenon is based on instability of homogeneous current flow under conditions of negative differential resistivity which leads to formation of current domains in our sample, similar to the case of single-layer systems. △ Less

Submitted 26 January, 2011; originally announced January 2011.

Comments: 4 pages, 4 figures

Journal ref: Phys. Rev. B, 2011, v.83, 041306(R)

Thermally activated intersubband scattering and oscillating magnetoresistance in quantum wells

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: Experimental studies of magnetoresistance in high-mobility wide quantum wells reveal oscillations which appear with an increase in temperature to 10 K and whose period is close to that of Shubnikov-de Haas oscillations. The observed phenomenon is identified as magnetointersubband oscillations caused by the scattering of electrons between two occupied subbands and the third subband which becomes oc… ▽ More Experimental studies of magnetoresistance in high-mobility wide quantum wells reveal oscillations which appear with an increase in temperature to 10 K and whose period is close to that of Shubnikov-de Haas oscillations. The observed phenomenon is identified as magnetointersubband oscillations caused by the scattering of electrons between two occupied subbands and the third subband which becomes occupied as a result of thermal activation. These small-period oscillations are less sensitive to thermal suppression than the largeperiod magnetointersubband oscillations caused by the scattering between the first and the second subbands. Theoretical study, based on consideration of electron scattering near the edge of the third subband, gives a reasonable explanation of our experimental findings. △ Less

Submitted 28 October, 2010; originally announced October 2010.

Comments: 9 pages, 5 figures

Journal ref: Physical Review B 82, 165333 (2010)

Microwave zero-resistance states in a bilayer electron system

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: Magnetotransport measurements on a high-mobility electron bilayer system formed in a wide GaAs quantum well reveal vanishing dissipative resistance under continuous microwave irradiation. Profound zero-resistance states (ZRS) appear even in the presence of additional intersubband scattering of electrons. We study the dependence of photoresistance on frequency, microwave power. and temperature. Exp… ▽ More Magnetotransport measurements on a high-mobility electron bilayer system formed in a wide GaAs quantum well reveal vanishing dissipative resistance under continuous microwave irradiation. Profound zero-resistance states (ZRS) appear even in the presence of additional intersubband scattering of electrons. We study the dependence of photoresistance on frequency, microwave power. and temperature. Experimental results are compared with a theory demonstrating that the conditions for absolute negative resistivity correlate with the appearance of ZRS. △ Less

Submitted 8 July, 2010; originally announced July 2010.

Comments: 5 pages, 4 figures

Journal ref: PRL 105, 026804 (2010)

Magnetic field induced transition in a wide parabolic well superimposed with superlattice

Authors: G. M. Gusev, Yu. A. Pusep, A. K. Bakarov, A. I. Toropov, J. C. Portal

Abstract: We study a $Al_{x}Ga_{x-1}As$ parabolic quantum wells (PQW) with $GaAs/Al_{x}Ga_{x-1}As$ square superlattice. The magnetotransport in PQW with intentionally disordered short-period superlattice reveals a surprising transition from electrons distribution over whole parabolic well to independent-layer states with unequal density. The transition occurs in the perpendicular magnetic field at Landau fi… ▽ More We study a $Al_{x}Ga_{x-1}As$ parabolic quantum wells (PQW) with $GaAs/Al_{x}Ga_{x-1}As$ square superlattice. The magnetotransport in PQW with intentionally disordered short-period superlattice reveals a surprising transition from electrons distribution over whole parabolic well to independent-layer states with unequal density. The transition occurs in the perpendicular magnetic field at Landau filling factor $ν\approx3$ and is signaled by the appearance of the strong and developing fractional quantum Hall (FQH) states and by the enhanced slope of the Hall resistance. We attribute the transition to the possible electron localization in the x-y plane inside the lateral wells, and formation of the FQH states in the central well of the superlattice, driven by electron-electron interaction. △ Less

Submitted 7 April, 2010; originally announced April 2010.

Comments: 5 pages, 4 figures

Journal ref: Phys.Rev. B, 81, 165302 (2010)

Quantum Hall effect near the charge neutrality point in two-dimensional electron-hole system

Authors: G. M. Gusev, E. B. Olshanetsky, Z. D. Kvon, N. N. Mikhailov, S. A. Dvoretsky, J. C. Portal

Abstract: We study the transport properties of $HgTe$-based quantum wells containing simultaneously electrons and holes in magnetic field B. At the charge neutrality point (CNP) with nearly equal electron and hole densities, the resistance is found to increase very strongly with B while the Hall resistivity turns to zero. This behavior results in a wide plateau in the Hall conductivity $σ_{xy}\approx 0$ and… ▽ More We study the transport properties of $HgTe$-based quantum wells containing simultaneously electrons and holes in magnetic field B. At the charge neutrality point (CNP) with nearly equal electron and hole densities, the resistance is found to increase very strongly with B while the Hall resistivity turns to zero. This behavior results in a wide plateau in the Hall conductivity $σ_{xy}\approx 0$ and in a minimum of diagonal conductivity $σ_{xx}$ at $ν=ν_p-ν_n=0$, where $ν_n$ and $ν_p$ are the electron and hole Landau filling factors. We suggest that the transport at the CNP point is determined by electron-hole "snake states" propagating along the $ν=0$ lines. Our observations are qualitatively similar to the quantum Hall effect in graphene as well as to the transport in random magnetic field with zero mean value. △ Less

Submitted 30 March, 2010; originally announced March 2010.

Comments: 4.5 pages, 4 figures

Journal ref: Phys. Rev. Lett, 2010, v.104, 166401

Crossover between distinct mechanisms of microwave photoresistance in bilayer systems

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: We report on temperature-dependent magnetoresistance measurements in balanced double quantum wells exposed to microwave irradiation for various frequencies. We have found that the resistance oscillations are described by the microwave-induced modification of electron distribution function limited by inelastic scattering (inelastic mechanism), up to a temperature of T*~4 K. With increasing temper… ▽ More We report on temperature-dependent magnetoresistance measurements in balanced double quantum wells exposed to microwave irradiation for various frequencies. We have found that the resistance oscillations are described by the microwave-induced modification of electron distribution function limited by inelastic scattering (inelastic mechanism), up to a temperature of T*~4 K. With increasing temperature, a strong deviation of the oscillation amplitudes from the behavior predicted by this mechanism is observed, presumably indicating a crossover to another mechanism of microwave photoresistance, with similar frequency dependence. Our analysis shows that this deviation cannot be fully understood in terms of contribution from the mechanisms discussed in theory. △ Less

Submitted 11 February, 2010; originally announced February 2010.

Comments: 7 pages, 4 figures

Journal ref: Phys. Rev. B 81, 085311 (2010)

Magnetoresistance oscillations in multilayer systems - triple quantum wells

Authors: S. Wiedmann, N. C. Mamani, G. M. Gusev, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: Magnetoresistance of two-dimensional electron systems with several occupied subbands oscillates owing to periodic modulation of the probability of intersubband transitions by the quantizing magnetic field. In addition to previous investigations of these magneto-intersubband (MIS) oscillations in two-subband systems, we report on both experimental and theoretical studies of such a phenomenon in t… ▽ More Magnetoresistance of two-dimensional electron systems with several occupied subbands oscillates owing to periodic modulation of the probability of intersubband transitions by the quantizing magnetic field. In addition to previous investigations of these magneto-intersubband (MIS) oscillations in two-subband systems, we report on both experimental and theoretical studies of such a phenomenon in three-subband systems realized in triple quantum wells. We show that the presence of more than two subbands leads to a qualitatively different MIS oscillation picture, described as a superposition of several oscillating contributions. Under a continuous microwave irradiation, the magnetoresistance of triple-well systems exhibits an interference of MIS oscillations and microwaveinduced resistance oscillations. The theory explaining these phenomena is presented in the general form, valid for an arbitrary number of subbands. A comparison of theory and experiment allows us to extract temperature dependence of quantum lifetime of electrons and to confirm the applicability of the inelastic mechanism of microwave photoresistance for the description of magnetotransport in multilayer systems. △ Less

Submitted 7 December, 2009; originally announced December 2009.

Comments: 10 pages, 5 figures

Journal ref: Phys. Rev. B 80, 245306 (2009)

High order fractional microwave induced resistance oscillations in 2D systems

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, A. K. Bakarov, J. C. Portal

Abstract: We report on the observation of microwave-induced resistance oscillations associated with the fractional ratio n/m of the microwave irradiation frequency to the cyclotron frequency for m up to 8 in a two-dimensional electron system with high electron density. The features are quenched at high microwave frequencies independent of the fractional order m. We analyze temperature, power, and frequenc… ▽ More We report on the observation of microwave-induced resistance oscillations associated with the fractional ratio n/m of the microwave irradiation frequency to the cyclotron frequency for m up to 8 in a two-dimensional electron system with high electron density. The features are quenched at high microwave frequencies independent of the fractional order m. We analyze temperature, power, and frequency dependencies of the magnetoresistance oscillations and discuss them in connection with existing theories. △ Less

Submitted 30 July, 2009; originally announced July 2009.

Comments: 5 pages, 5 figures

Journal ref: Phys. Rev. B 80, 035317 (2009)

Interference oscillations of microwave photoresistance in double quantum wells

Authors: S. Wiedmann, G. M. Gusev, O. E. Raichev, T. E. Lamas, A. K. Bakarov, J. C. Portal

Abstract: We observe oscillatory magnetoresistance in double quantum wells under microwave irradiation. The results are explained in terms of the influence of subband coupling on the frequency-dependent photoinduced part of the electron distribution function. As a consequence, the magnetoresistance demonstrates the interference of magneto-intersubband oscillations and conventional microwave- induced resis… ▽ More We observe oscillatory magnetoresistance in double quantum wells under microwave irradiation. The results are explained in terms of the influence of subband coupling on the frequency-dependent photoinduced part of the electron distribution function. As a consequence, the magnetoresistance demonstrates the interference of magneto-intersubband oscillations and conventional microwave- induced resistance oscillations. △ Less

Submitted 29 September, 2008; originally announced September 2008.

Comments: 5 pages, 3 figures

Journal ref: Physical Review B 78, 121301(R) 2008

Coulomb pseudogap in elastic 2D-2D electron tunneling in a quantizing magnetic field

Authors: V. G. Popov, O. N. Makarovskii, V. Renard, L. Eaves, J. -C. Portal

Abstract: The electron tunneling is experimentally studied between two-dimensional electron gases (2DEGs) formed in a single-doped-barrier heterostructure in the magnetic fields directed perpendicular to the 2DEGs planes. It is well known that the quantizing magnetic field induces the Coulomb pseudogap suppressing the electron tunneling at Fermi level. In this paper we firstly present the experimental res… ▽ More The electron tunneling is experimentally studied between two-dimensional electron gases (2DEGs) formed in a single-doped-barrier heterostructure in the magnetic fields directed perpendicular to the 2DEGs planes. It is well known that the quantizing magnetic field induces the Coulomb pseudogap suppressing the electron tunneling at Fermi level. In this paper we firstly present the experimental results revealing the pseudogap in the electron tunneling assisted by elastic electron scattering on disorder. △ Less

Submitted 22 August, 2008; originally announced August 2008.

Comments: 5 pages, 2 figures, Proc. of the ICPS-2008

Boundary-mediated electron-electron interactions in quantum point contacts

Authors: Vincent Thomas Francois Renard, O. A. Tkachenko, V. A. Tkachenko, T. Ota, N. Kumada, J. -C. Portal, Y. Hirayama

Abstract: An unusual increase of the conductance with temperature is observed in clean quantum point contacts for conductances larger than 2e^2/h. At the same time a positive magnetoresistance arises at high temperatures. A model accounting for electron-electron interactions mediated by bound- aries (scattering on Friedel oscillations) qualitatively describes the observation. It is supported by numerical… ▽ More An unusual increase of the conductance with temperature is observed in clean quantum point contacts for conductances larger than 2e^2/h. At the same time a positive magnetoresistance arises at high temperatures. A model accounting for electron-electron interactions mediated by bound- aries (scattering on Friedel oscillations) qualitatively describes the observation. It is supported by numerical simulation at zero magnetic field. △ Less

Submitted 7 April, 2008; v1 submitted 25 March, 2008; originally announced March 2008.

Comments: To appear in Phys. Rev. Lett Updated version of Fig. 2

Journal ref: Physical Review Letters 100 (2008) 186801

Experimental investigation of the ratchet effect in a two-dimensional electron system with broken spatial inversion symmetry

Authors: S. Sassine, Yu. Krupko, J. -C. Portal, Z. D. Kvon, R. Murali, K. P. Martin, G. Hill, A. D. Wieck

Abstract: We report on experimental evidence of directed electron transport, induced by external linear-polarized microwave irradiation, in a two-dimensional spatially-periodic asymmetrical system called ratchet. The broken spatial symmetry was introduced in a high mobility two-dimensional electron gas based on AlGaAs/GaAs heterojunction, by patterning an array of artificial semi-discs-shaped antidots. We… ▽ More We report on experimental evidence of directed electron transport, induced by external linear-polarized microwave irradiation, in a two-dimensional spatially-periodic asymmetrical system called ratchet. The broken spatial symmetry was introduced in a high mobility two-dimensional electron gas based on AlGaAs/GaAs heterojunction, by patterning an array of artificial semi-discs-shaped antidots. We show that the direction of the transport is efficiently changed by microwave polarization. The dependence of the effect on magnetic field and temperature is investigated. This represents a significant step towards the realization of new microwave detectors and current generators. △ Less

Submitted 28 February, 2008; originally announced February 2008.

A New Embedded Measurement Structure for eDRAM Capacitor

Authors: L. Lopez, J. M. Portal, D. Nee

Abstract: The embedded DRAM (eDRAM) is more and more used in System On Chip (SOC). The integration of the DRAM capacitor process into a logic process is challenging to get satisfactory yields. The specific process of DRAM capacitor and the low capacitance value (~30F) of this device induce problems of process monitoring and failure analysis. We propose a new test structure to measure the capacitance value… ▽ More The embedded DRAM (eDRAM) is more and more used in System On Chip (SOC). The integration of the DRAM capacitor process into a logic process is challenging to get satisfactory yields. The specific process of DRAM capacitor and the low capacitance value (~30F) of this device induce problems of process monitoring and failure analysis. We propose a new test structure to measure the capacitance value of each DRAM cell capacitor in a DRAM array. This concept has been validated by simulation on a 0.18$μ$m eDRAM technology. △ Less

Submitted 25 October, 2007; originally announced October 2007.

Comments: Submitted on behalf of EDAA (http://www.edaa.com/)

Journal ref: Dans Design, Automation and Test in Europe - DATE'05, Munich : Allemagne (2005)

Inducement and suppression of Coulomb effects in elastic 2D-2D electron tunnelling in a quantizing magnetic field

Authors: V. G. Popov, Yu. V. Dubrovskii, J. -C. Portal

Abstract: Tunnelling between two-dimensional electron systems has been studied in the magnetic field perpendicular to the systems planes. The satellite conductance peaks of the main resonance have been observed due to the electron tunnelling assisted by the elastic scattering on impurities in the barrier layer. These peaks are shown to shift to the higher voltage due to the Coulomb pseudogap in the interm… ▽ More Tunnelling between two-dimensional electron systems has been studied in the magnetic field perpendicular to the systems planes. The satellite conductance peaks of the main resonance have been observed due to the electron tunnelling assisted by the elastic scattering on impurities in the barrier layer. These peaks are shown to shift to the higher voltage due to the Coulomb pseudogap in the intermediate fields. In the high magnetic fields the pseudogap shift is disappeared. △ Less

Submitted 9 July, 2007; originally announced July 2007.

Comments: 15th Int. Symp. "Nanostructures: Physics and Technology" Novosibirsk, Russia, June 25-29, 2007 Ioffe Institute

Electron transport through antidot superlattices in $Si/SiGe$ heterostructures: new magnetoresistance resonances in lattices with large diameter antidots

Authors: E. B. Olshanetsky, Vincent Thomas Francois Renard, Z. D. Kvon, J. -C. Portal, J. -M. Hartmann

Abstract: In the present work we have investigated the transport properties in a number of Si/SiGe samples with square antidot lattices of different periods. In samples with lattice periods equal to 700 nm and 850 nm we have observed the conventional low-field commensurability magnetoresistance peaks consistent with the previous observations in GaAs/AlGaAs and Si/SiGe samples with antidot lattices. In sam… ▽ More In the present work we have investigated the transport properties in a number of Si/SiGe samples with square antidot lattices of different periods. In samples with lattice periods equal to 700 nm and 850 nm we have observed the conventional low-field commensurability magnetoresistance peaks consistent with the previous observations in GaAs/AlGaAs and Si/SiGe samples with antidot lattices. In samples with a 600 nm lattice period a new series of well-developed magnetoresistance oscillations has been found beyond the last commensurability peak which are supposed to originate from periodic skipping orbits encircling an antidot with a particular number of bounds. △ Less

Submitted 21 September, 2006; originally announced September 2006.

Comments: To appear in EuroPhys. Lett

Journal ref: Europhysics Letters (EPL) 76, 4 (2006) 657

Quantum corrections to the conductivity and Hall coefficient of a two-dimensional electron gas in a dirty AlGaAs/GaAs/AlGaAs quantum well: from the diffusive to the ballistic regime

Authors: V. T. Renard, I. V. Gornyi, O. A. Tkachenko, V. A. Tkachenko, Z. D. Kvon, E. B. Olshanetsky, A. I. Toropov, J. -C. Portal

Abstract: We report an experimental study of quantum conductivity corrections in a low mobility, high density two-dimensional electron gas in a AlGaAs/GaAs/AlGaAs quantum well in a wide temperature range (1.5K - 110K). This temperature range covers both the diffusive and the ballistic interaction regimes for our samples. It has been therefore possible to study the crossover between these regimes for both… ▽ More We report an experimental study of quantum conductivity corrections in a low mobility, high density two-dimensional electron gas in a AlGaAs/GaAs/AlGaAs quantum well in a wide temperature range (1.5K - 110K). This temperature range covers both the diffusive and the ballistic interaction regimes for our samples. It has been therefore possible to study the crossover between these regimes for both the longitudinal conductivity and the Hall effect. We perform a parameter free comparison of our experimental data for the longitudinal conductivity at zero magnetic field, the Hall coefficient, and the magnetoresistivity to the recent theories of interaction-induced corrections to the transport coefficients. A quantitative agreement between these theories and our experimental results has been found. △ Less

Submitted 19 May, 2005; originally announced May 2005.

Comments: extended version of cond-mat/0412463

Journal ref: Physical Review B 72 (2005) 075313

Directed electron transport through ballistic quantum dot under microwave radiation

Authors: Jing-qiao Zhang, Sergey Vitkalov, Z. D. Kvon, J. C. Portal, A. Wieck

Abstract: Rectification of microwave radiation by asymmetric, ballistic quantum dot is observed. The directed transport is studied at different frequency (1-40 GHz) temperatures (0.3K-6K)and magnetic field. Dramatic reduction of the rectification is found in magnetic fields at which the cyclotron (Larmor) radius of the electron orbits at Fermi level is smaller than the size of the quantum dot. It strongly… ▽ More Rectification of microwave radiation by asymmetric, ballistic quantum dot is observed. The directed transport is studied at different frequency (1-40 GHz) temperatures (0.3K-6K)and magnetic field. Dramatic reduction of the rectification is found in magnetic fields at which the cyclotron (Larmor) radius of the electron orbits at Fermi level is smaller than the size of the quantum dot. It strongly suggests the ballistic nature of the observed nonlinear phenomena. Both symmetric and anti-symmetric with respect to the magnetic field contributions to the directed transport are presented. We have found that the behavior of the symmetric part of the rectified voltage with the magnetic field is different significantly for microwaves with different frequencies. A ballistic model of the directed transport is proposed. △ Less

Submitted 4 April, 2005; originally announced April 2005.

Comments: 5 pages, 3 figures

Anisotropic positive magnetoresistance of a nonplanar 2D electron gas in a parallel magnetic field

Authors: A. V. Goran, A. A. Bykov, A. K. Bakarov, J. C. Portal

Abstract: We study the transport properties of a 2D electron gas in narrow GaAs quantum wells with AlAs/GaAs superlattice barriers. It is shown that the anisotropic positive magnetoresistance observed in selectively doped semiconductor structures in a parallel magnetic field is caused by the spatial modulation of the 2D electron gas. We study the transport properties of a 2D electron gas in narrow GaAs quantum wells with AlAs/GaAs superlattice barriers. It is shown that the anisotropic positive magnetoresistance observed in selectively doped semiconductor structures in a parallel magnetic field is caused by the spatial modulation of the 2D electron gas. △ Less

Submitted 22 February, 2005; originally announced February 2005.

Comments: 4 pages, 3 figures

Journal ref: JETP Letters, Vol.79, No.10, 2004, pp.495-498

Excited two-dimensional magnetopolaron states in quantum well of resonant tunnel junction

Authors: D. Yu. Ivanov, M. V. Chukalina, E. G. Takhtamirov, Yu. V. Dubrovskii, L. Eaves, V. A. Volkov, E. E. Vdovin, J. -C. Portal, D. K. Maude, M. Henini, G. Hill

Abstract: Tunnel spectroscopy is used to probe the electronic structure in GaAs quantum well of resonant tunnel junction over wide range of energies and magnetic fields normal to layers. Spin degenerated high Landau levels ($N=2\div7$) are found to be drastically renormalised near energies when the longitudinal optical-phonon ($\hbarω_{LO}$) and cyclotron energy ($\hbarω_{C}$) are satisfied condition… ▽ More Tunnel spectroscopy is used to probe the electronic structure in GaAs quantum well of resonant tunnel junction over wide range of energies and magnetic fields normal to layers. Spin degenerated high Landau levels ($N=2\div7$) are found to be drastically renormalised near energies when the longitudinal optical-phonon ($\hbarω_{LO}$) and cyclotron energy ($\hbarω_{C}$) are satisfied condition $\hbarω_{LO}=m\hbarω_{C}$, where $m=1,2,3$. This renormalisation is attributed to formation of resonant magnetopolarons, i.e. mixing of high index Landau levels by strong interaction of electrons at Landau level states with LO-phonons. △ Less

Submitted 13 January, 2005; originally announced January 2005.

Comments: 4 pages, 3 eps figures, revtex4

Development of the tunnelling gap in disordered 2D electron system with magnetic field: observation of the soft-hard gap transition

Authors: Yu. V. Dubrovskii, V. A. Volkov, L. Eaves, E. E. Vdovin, O. N. Makarovskii, J. -C. Portal, M. Henini, G. Hill

Abstract: Magnetic field suppression of the tunneling between disordered 2D electron systems in GaAs around zero bias voltage has been studied. Magnetic field B normal to the layers induces a dip in the tunneling density of states (TDOS) centered precisely at the Fermi level, i.e. soft tunneling gap. The soft gap has a linear form with finite TDOS diminishing with B at the Fermi level. Driven by magnetic… ▽ More Magnetic field suppression of the tunneling between disordered 2D electron systems in GaAs around zero bias voltage has been studied. Magnetic field B normal to the layers induces a dip in the tunneling density of states (TDOS) centered precisely at the Fermi level, i.e. soft tunneling gap. The soft gap has a linear form with finite TDOS diminishing with B at the Fermi level. Driven by magnetic field the transition soft-hard gap has been observed, i.e. the TDOS vanishes in the finite energy window around Fermi level at B>13 T. △ Less

Submitted 6 January, 2005; v1 submitted 5 January, 2005; originally announced January 2005.

Comments: Revtex4, 6 pages, 5 Postscript figures, to be submitted to Physical Review B

Quantum corrections to the conductivity and Hall coefficient of a 2D electron gas in a dirty AlGaAs/GaAs/AlGaAs quantum well: transition from diffusive to ballistic regime

Authors: Vincent Thomas Francois Renard, O. A. Tkachenko, Ze Don Kvon, E. B. Olshanetsky, A. I. Toropov, J. C Portal

Abstract: We report an experimental study of the quantum corrections to the longitudinal conductivity and the Hall coefficient of a low mobility, high density two-dimensional two-dimensional electron gas in a AlGaAs/GaAs/AlGaAs quantum well in a wide temperature range (1.5 K - 110 K). This temperature range covers both the diffusive and the ballistic interaction regimes for our samples. It was therefore p… ▽ More We report an experimental study of the quantum corrections to the longitudinal conductivity and the Hall coefficient of a low mobility, high density two-dimensional two-dimensional electron gas in a AlGaAs/GaAs/AlGaAs quantum well in a wide temperature range (1.5 K - 110 K). This temperature range covers both the diffusive and the ballistic interaction regimes for our samples. It was therefore possible to study the crossover region for the longitudinal conductivity and the Hall effect. △ Less

Submitted 17 December, 2004; originally announced December 2004.

Journal ref: Physical Review B 72 (2005) 075313

Large positive magneto-resistance in high mobility 2D electron gas : interplay of short and long range disorder

Authors: Vincent T. F Renard, Ze Don Kvon, G M Gusev, J. C Portal

Abstract: We have observed a large positive quasi-classical magneto-resistance (MR) in a high mobility 2D electron gas in AlGaAs/GaAs heterostructure. The magneto-resistance is non-saturating and increases with magnetic field as $ρ_{xx}\sim B^α (α=0.9-1.2)$. In antidot lattices a non-monotonic MR is observed. We show that in both cases this MR can be qualitatively described in terms of the theory recently… ▽ More We have observed a large positive quasi-classical magneto-resistance (MR) in a high mobility 2D electron gas in AlGaAs/GaAs heterostructure. The magneto-resistance is non-saturating and increases with magnetic field as $ρ_{xx}\sim B^α (α=0.9-1.2)$. In antidot lattices a non-monotonic MR is observed. We show that in both cases this MR can be qualitatively described in terms of the theory recently advanced by Polyakov et al (PRB, 64, 205306 (2001)). Their prediction is that such behavior as we observe may be the consequence of a concurrent existence of short and long range scattering potentials. △ Less

Submitted 9 March, 2004; v1 submitted 7 January, 2004; originally announced January 2004.

Comments: 7 jan 2004

Journal ref: Physical Review B 70 (2004) 033303

Conductivity of a 2DEG in Si/SiGe heterostructure near metal- insulator transition: role of the short and long range scattering potential

Authors: E. B. Olshanetsky, V. Renard, Z. D. Kvon, J. C. Portal, N. J. Woods, J. Zhang, J. J Harris

Abstract: We report the observation of a metal-insulator transition (MIT) in a two- dimensional electron gas (2DEG) in a Si/SiGe heterostructure at zero magnetic field. On going through the MIT we observe the corresponding evolution of the magnetic field induced transition between the insulating phase and the quantum Hall (QH) liquid state in the QH regime. Similar to the previous reports for a GaAs sampl… ▽ More We report the observation of a metal-insulator transition (MIT) in a two- dimensional electron gas (2DEG) in a Si/SiGe heterostructure at zero magnetic field. On going through the MIT we observe the corresponding evolution of the magnetic field induced transition between the insulating phase and the quantum Hall (QH) liquid state in the QH regime. Similar to the previous reports for a GaAs sample, we find that the critical magnetic field needed to produce the transition becomes zero at the critical electron density corresponding to the zero field MIT. The temperature dependence of the conductivity in a metallic-like state at zero field is compared with the theory of the interaction corrections at intermediate and ballistic regimes $k_{B}Tτ/\hbar\geq1$. The theory yields a good fit for the linear part of the curve. However the slope of that part of $σ_{xx}(T)$ is about two times smaller than that reported in other 2D systems with similar values of $r_s$. At the same time, the recent theory of magnetoresistance due to electron-electron interaction in the case of arbitrary $k_{B}Tτ/\hbar$, smooth disorder and classically strong fields does not seem to be quite adequate for the description of the parabolic magnetoresistance observed in our samples. We attribute these results to the fact that neither of these theories deals with the whole scattering potential in a sample but leaves either its long range or its short range component out of consideration. △ Less

Submitted 14 March, 2003; originally announced March 2003.

Comments: 8 pages, 5 figures

Metal-Insulator oscillations in a Two-dimensional Electron-Hole system

Authors: R. J. Nicholas, K. Takashina, M. Lakrimi, B. Kardynal, S. Khym, N. J. Mason, D. M. Symons, D. K. Maude, J. C. Portal

Abstract: The electrical transport properties of a bipolar InAs/GaSb system have been studied in magnetic field. The resistivity oscillates between insulating and metallic behaviour while the quantum Hall effect shows a digital character oscillating from 0 to 1 conducatance quantum e^2/h. The insulating behaviour is attributed to the formation of a total energy gap in the system. A novel looped edge state… ▽ More The electrical transport properties of a bipolar InAs/GaSb system have been studied in magnetic field. The resistivity oscillates between insulating and metallic behaviour while the quantum Hall effect shows a digital character oscillating from 0 to 1 conducatance quantum e^2/h. The insulating behaviour is attributed to the formation of a total energy gap in the system. A novel looped edge state picture is proposed associated with the appearance of a voltage between Hall probes which is symmetric on magnetic field reversal. △ Less

Submitted 11 April, 2000; v1 submitted 7 December, 1999; originally announced December 1999.

Comments: 4 pages, 5 Postscript figures: revised version

Proximity effects and Andreev reflection in mesoscopic SNS junction with perfect NS interfaces

Authors: Z. D. Kvon, T. I. Baturina, R. A. Donaton, M. R. Baklanov, K. Maex, E. B. Olshanetsky, A. E. Plotnikov, J. C. Portal

Abstract: Low temperature transport measurements on superconducting film - normal metal wire - superconducting film (SNS) junctions fabricated on the basis of 6 nm thick superconducting polycrystalline PtSi films are reported. The structures with the normal metal wires of two different lengths L=1.5 $μ$m and L=6$μ$m and the same widths W=0.3$μ$m are studied. Zero bias resistance dip related to pair curren… ▽ More Low temperature transport measurements on superconducting film - normal metal wire - superconducting film (SNS) junctions fabricated on the basis of 6 nm thick superconducting polycrystalline PtSi films are reported. The structures with the normal metal wires of two different lengths L=1.5 $μ$m and L=6$μ$m and the same widths W=0.3$μ$m are studied. Zero bias resistance dip related to pair current proximity effect is observed for all junctions whereas the subharmonic energy gap structure originating from phase coherent multiple Andreev reflections have occurs only in the SNS junctions with short wires. △ Less

Submitted 18 July, 1999; originally announced July 1999.

Comments: ReVTex, 4 pages, 4 eps figures included

Journal ref: Phys. Rev. B 61, 11340 (2000)

Searches for Skyrmions in the Limit of Zero g-Factor

Authors: D. R. Leadley, R. J. Nicholas, D. K. Maude, A. N. Utjuzh, J. C. Portal, J. J. Harris, C. T. Foxon

Abstract: Energy gaps have been measured for the ferromagnetic quantum Hall effect states at v=1 and 3 in GaAs/GaAlAs heterojunctions as a function of Zeeman energy, which is reduced to zero by applying hydrostatic pressures of up to 20kbar. At large Zeeman energy the gaps are consistent with spin wave excitations. For a low density sample the gap at v=1 decreases with increasing pressure and reaches a mi… ▽ More Energy gaps have been measured for the ferromagnetic quantum Hall effect states at v=1 and 3 in GaAs/GaAlAs heterojunctions as a function of Zeeman energy, which is reduced to zero by applying hydrostatic pressures of up to 20kbar. At large Zeeman energy the gaps are consistent with spin wave excitations. For a low density sample the gap at v=1 decreases with increasing pressure and reaches a minimum when the g-factor vanishes. At small Zeeman energy the excitation appears to consist of a large number of reversed spins and may be interpreted as a Skyrmion. The data also suggest Skyrmionic excitations take place at v=3. The width of the minimum at v=1 is found to decrease as the g-factor is reduced in a similar way for all samples. △ Less

Submitted 27 May, 1998; originally announced May 1998.

Comments: 14 pages, 9 figures. To appear in Semiconductor Science and Technology