Manus Hayne
Research Subjects
Manus Hayne has a total of 273 co-authors. Here are the top co-authors by number of shared publications:
V. V. Moshchalkov
Institute for Nanoscale Physics and Chemistry (INPAC), Katholieke Universiteit Leuven, Celestijnenlaan 200D, B-3001 Leuven, Belgium
29 shared publications- Mohammed Henini
University of Nottingham
16 shared publications - Peter David Hodgson
Department of Physics, Lancaster University 2 , Lancaster LA1 4YB
16 shared publications - Qiandong Zhuang
Physics Department, Lancaster University, Lancaster, UK
13 shared publications - Robert James Young
Lancaster University
13 shared publications A. Usher
Department of Physics, University of Exeter, Stocker Road, Exeter EX4 4QL, United Kingdom
10 shared publications
Latest Publications
- Open Access
- 23 March 2026
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 139 (12)
Artificial synapse based on ULTRARAM memory device for neuromorphic applications
The memory demands of large-scale deep neural networks (DNNs) require synaptic weight values to be stored and updated in off-chip memory, such as dynamic random-access memory, which reduces energy efficiency and increases training time. Monolithic crossbar or pseudo-crossbar arrays using analog non-volatile memories, which can store and update weights on-chip, present an opportunity to efficiently accelerate DNN training. In this article, we present on-chip training and inference of a neural network using an ULTRARAM memory device-based synaptic array and complementary metal–oxide–semiconductor (CMOS) peripheral circuits. ULTRARAM is a promising emerging memory exhibiting high endurance (>107 P/E cycles), ultrahigh retention (>1000 years), and ultralow switching energy per unit area. A physics-based compact model of ULTRARAM memory device has been proposed to capture the real-time trapping/de-trapping of charges in the floating gate and utilized for the synapse simulations. A circuit-level macro-model is employed to evaluate and benchmark the on-chip learning performance in terms of area, latency, energy, and accuracy of an ULTRARAM synaptic core. In comparison with CMOS-based design, it demonstrates an overall improvement in area and energy by 1.8× and 1.52×, respectively, with 91% of training accuracy.
- 23 March 2026
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 139 (12)
Open AccessArtificial synapse based on ULTRARAM memory device for neuromorphic applications
The memory demands of large-scale deep neural networks (DNNs) require synaptic weight values to be stored and updated in off-chip memory, such as dynamic random-access memory, which reduces energy efficiency and increases training time. Monolithic crossbar or pseudo-crossbar arrays using analog non-volatile memories, which can store and update weights on-chip, present an opportunity to efficiently accelerate DNN training. In this article, we present on-chip training and inference of a neural network using an ULTRARAM memory device-based synaptic array and complementary metal–oxide–semiconductor (CMOS) peripheral circuits. ULTRARAM is a promising emerging memory exhibiting high endurance (>107 P/E cycles), ultrahigh retention (>1000 years), and ultralow switching energy per unit area. A physics-based compact model of ULTRARAM memory device has been proposed to capture the real-time trapping/de-trapping of charges in the floating gate and utilized for the synapse simulations. A circuit-level macro-model is employed to evaluate and benchmark the on-chip learning performance in terms of area, latency, energy, and accuracy of an ULTRARAM synaptic core. In comparison with CMOS-based design, it demonstrates an overall improvement in area and energy by 1.8× and 1.52×, respectively, with 91% of training accuracy.
- Open Access
Artificial Synapse based on ULTRARAM Memory Device for Neuromorphic Applications
The memory demands of large-scale deep neural networks (DNNs) require synaptic weight values to be stored and updated in off-chip memory like dynamic random-access memory, which reduces energy efficiency and increases training time. Monolithic crossbar or pseudo-crossbar arrays using analog non-volatile memories, which can store and update weights on-chip, present an opportunity to efficiently accelerate DNN training. In this article, we present on-chip training and inference of a neural network using an ULTRARAM memory device-based synaptic array and complementary metal-oxide-semiconductor (CMOS) peripheral circuits. ULTRARAM is a promising emerging memory exhibiting high endurance (>10^7 P/E cycles), ultra-high retention (>1000 years), and ultra-low switching energy per unit area. A physics-based compact model of ULTRARAM memory device has been proposed to capture the real-time trapping/de-trapping of charges in the floating gate (FG) and utilized for the synapse simulations. A circuit-level macro-model is employed to evaluate and benchmark the on-chip learning performance in terms of area, latency, energy, and accuracy of an ULTRARAM synaptic core. In comparison to CMOS-based design, it demonstrates an overall improvement in area and energy by 1.8x and 1.52x, respectively, with 91% of training accuracy.
Open AccessArtificial Synapse based on ULTRARAM Memory Device for Neuromorphic Applications
The memory demands of large-scale deep neural networks (DNNs) require synaptic weight values to be stored and updated in off-chip memory like dynamic random-access memory, which reduces energy efficiency and increases training time. Monolithic crossbar or pseudo-crossbar arrays using analog non-volatile memories, which can store and update weights on-chip, present an opportunity to efficiently accelerate DNN training. In this article, we present on-chip training and inference of a neural network using an ULTRARAM memory device-based synaptic array and complementary metal-oxide-semiconductor (CMOS) peripheral circuits. ULTRARAM is a promising emerging memory exhibiting high endurance (>10^7 P/E cycles), ultra-high retention (>1000 years), and ultra-low switching energy per unit area. A physics-based compact model of ULTRARAM memory device has been proposed to capture the real-time trapping/de-trapping of charges in the floating gate (FG) and utilized for the synapse simulations. A circuit-level macro-model is employed to evaluate and benchmark the on-chip learning performance in terms of area, latency, energy, and accuracy of an ULTRARAM synaptic core. In comparison to CMOS-based design, it demonstrates an overall improvement in area and energy by 1.8x and 1.52x, respectively, with 91% of training accuracy.
- Open Access
- 2 September 2025
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 138 (9)
Physics, modeling, and benchmarking of ULTRARAM: A compound semiconductor-based memory device
ULTRARAM is a promising emerging memory exhibiting high endurance, long retention, and ultra-low switching energy per unit area. This compound semiconductor-based non-volatile memory utilizes triple barrier resonant tunneling (TBRT) through InAs/AlSb heterostructures. In this work, we have proposed a physics-based compact model of ULTRARAM memory device that captures the real-time trapping/de-trapping of charges in the floating gate and used to calculate the device characteristics. In addition, we have performed numerical simulations of the TBRT stack using the Schrödinger equation inside the quantum well with mono-layer (∼0.6 nm) variations to check the impact on memory characteristics of the device. Array-level simulations and benchmarking highlight the promise of this technology.
- 2 September 2025
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 138 (9)
Open AccessPhysics, modeling, and benchmarking of ULTRARAM: A compound semiconductor-based memory device
ULTRARAM is a promising emerging memory exhibiting high endurance, long retention, and ultra-low switching energy per unit area. This compound semiconductor-based non-volatile memory utilizes triple barrier resonant tunneling (TBRT) through InAs/AlSb heterostructures. In this work, we have proposed a physics-based compact model of ULTRARAM memory device that captures the real-time trapping/de-trapping of charges in the floating gate and used to calculate the device characteristics. In addition, we have performed numerical simulations of the TBRT stack using the Schrödinger equation inside the quantum well with mono-layer (∼0.6 nm) variations to check the impact on memory characteristics of the device. Array-level simulations and benchmarking highlight the promise of this technology.
- Open Access
- 16 May 2025
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 137 (19)
Iso-electronic Sb impurities in GaAs studied by cross-sectional scanning tunneling microscopy
Highly mismatched III/V alloys, such as GaAsN and GaAsSb, are known to suffer from segregation and clustering effects, which often limit their application in devices. In this cross-sectional scanning tunneling microscopy (X-STM) study, we explore the atomic-scale behavior of iso-electronic Sb doping atoms in MBE-grown dilute GaAs1-xSbx (0.01 < x < 0.03). We found that Sb atoms up to four layers below the cleavage surface can be identified in filled-state X-STM images. They appear with diverse anisotropic contrasts, depending on the depth of the Sb atom. These features are classified and are related to their depth below the cleavage surface through careful symmetry considerations. We show that the depth-dependent contrast of Sb atoms in filled-state imaging is determined by both topographic effects (lattice deformation due to the large Sb atom) and electronic effects (resonances of Sb atoms in the valence band). This study shows that in MBE-grown GaAsSb alloys, the Sb atoms can be rapidly incorporated, in which case the GaAsSb layers suffer little from segregation and sharp interfaces can be obtained. Additionally, short-range ordering of Sb, which can be uniquely studied by X-STM, has been analyzed in terms of nearest-neighbor-pair formation, and we find that in MBE-grown GaAsSb materials, a tendency to form Sb pairs or clusters can be suppressed. This opens the route to create high-quality devices based on the highly mismatched GaAsSb alloy.
- 16 May 2025
- Published by AIP Publishing in Journal of Applied Physics
- Vol. 137 (19)
Open AccessIso-electronic Sb impurities in GaAs studied by cross-sectional scanning tunneling microscopy
Highly mismatched III/V alloys, such as GaAsN and GaAsSb, are known to suffer from segregation and clustering effects, which often limit their application in devices. In this cross-sectional scanning tunneling microscopy (X-STM) study, we explore the atomic-scale behavior of iso-electronic Sb doping atoms in MBE-grown dilute GaAs1-xSbx (0.01 < x < 0.03). We found that Sb atoms up to four layers below the cleavage surface can be identified in filled-state X-STM images. They appear with diverse anisotropic contrasts, depending on the depth of the Sb atom. These features are classified and are related to their depth below the cleavage surface through careful symmetry considerations. We show that the depth-dependent contrast of Sb atoms in filled-state imaging is determined by both topographic effects (lattice deformation due to the large Sb atom) and electronic effects (resonances of Sb atoms in the valence band). This study shows that in MBE-grown GaAsSb alloys, the Sb atoms can be rapidly incorporated, in which case the GaAsSb layers suffer little from segregation and sharp interfaces can be obtained. Additionally, short-range ordering of Sb, which can be uniquely studied by X-STM, has been analyzed in terms of nearest-neighbor-pair formation, and we find that in MBE-grown GaAsSb materials, a tendency to form Sb pairs or clusters can be suppressed. This opens the route to create high-quality devices based on the highly mismatched GaAsSb alloy.
- Open Access
- 15 November 2024
- Published by IOP Publishing in Semiconductor Science and Technology
- Vol. 39 (12) , 125011
Au/Ni/Au as a contact for p-type GaAs
An easy to fabricate ohmic-contact to moderately-doped p-type GaAs has been achieved. The tri-layer Au/Ni/Au contact is deposited by thermal evaporation, followed by rapid thermal annealing in nitrogen atmosphere. A series of annealing times and temperatures are explored to determine the influence of annealing conditions on the low-resistance ohmic contacts. The resulting contacts show more than three orders of magnitude reduction in contact resistance compared to alternative Ti/Au depositions.
- 15 November 2024
- Published by IOP Publishing in Semiconductor Science and Technology
- Vol. 39 (12) , 125011
Open AccessAu/Ni/Au as a contact for p-type GaAs
An easy to fabricate ohmic-contact to moderately-doped p-type GaAs has been achieved. The tri-layer Au/Ni/Au contact is deposited by thermal evaporation, followed by rapid thermal annealing in nitrogen atmosphere. A series of annealing times and temperatures are explored to determine the influence of annealing conditions on the low-resistance ohmic contacts. The resulting contacts show more than three orders of magnitude reduction in contact resistance compared to alternative Ti/Au depositions.
- 10 April 2022
- Published by Wiley in Advanced Electronic Materials
- Vol. 8 (4)
ULTRARAM: A Low‐Energy, High‐Endurance, Compound‐Semiconductor Memory on Silicon (Adv. Electron. Mater. 4/2022)
- 10 April 2022
- Published by Wiley in Advanced Electronic Materials
- Vol. 8 (4)
ULTRARAM: A Low‐Energy, High‐Endurance, Compound‐Semiconductor Memory on Silicon (Adv. Electron. Mater. 4/2022)
- Open Access
- 5 January 2022
- Published by Wiley in Advanced Electronic Materials
- Vol. 8 (4) , 2101103
ULTRARAM: A Low‐Energy, High‐Endurance, Compound‐Semiconductor Memory on Silicon
ULTRARAM is a nonvolatile memory with the potential to achieve fast, ultralow‐energy electron storage in a floating gate accessed through a triple‐barrier resonant tunneling heterostructure. Here its implementation is reported on a Si substrate; a vital step toward cost‐effective mass production. Sample growth using molecular beam epitaxy commences with deposition of an AlSb nucleation layer to seed the growth of a GaSb buffer layer, followed by the III–V memory epilayers. Fabricated single‐cell memories show clear 0/1 logic‐state contrast after ≤10 ms duration program/erase pulses of ≈2.5 V, a remarkably fast switching speed for 10 and 20 µm devices. Furthermore, the combination of low voltage and small device capacitance per unit area results in a switching energy that is orders of magnitude lower than dynamic random access memory and flash, for a given cell size. Extended testing of devices reveals retention in excess of 1000 years and degradation‐free endurance of over 107 program/erase cycles, surpassing very recent results for similar devices on GaAs substrates.
- 5 January 2022
- Published by Wiley in Advanced Electronic Materials
- Vol. 8 (4) , 2101103
Open AccessULTRARAM: A Low‐Energy, High‐Endurance, Compound‐Semiconductor Memory on Silicon
ULTRARAM is a nonvolatile memory with the potential to achieve fast, ultralow‐energy electron storage in a floating gate accessed through a triple‐barrier resonant tunneling heterostructure. Here its implementation is reported on a Si substrate; a vital step toward cost‐effective mass production. Sample growth using molecular beam epitaxy commences with deposition of an AlSb nucleation layer to seed the growth of a GaSb buffer layer, followed by the III–V memory epilayers. Fabricated single‐cell memories show clear 0/1 logic‐state contrast after ≤10 ms duration program/erase pulses of ≈2.5 V, a remarkably fast switching speed for 10 and 20 µm devices. Furthermore, the combination of low voltage and small device capacitance per unit area results in a switching energy that is orders of magnitude lower than dynamic random access memory and flash, for a given cell size. Extended testing of devices reveals retention in excess of 1000 years and degradation‐free endurance of over 107 program/erase cycles, surpassing very recent results for similar devices on GaAs substrates.
- Open Access
- 1 October 2021
- Published by Springer Nature in Research Square
ULTRARAM™: a low-energy, high-endurance, compound-semiconductor memory on silicon
ULTRARAM is a non-volatile memory with the potential to achieve fast, ultralow-energy electron storage in a floating gate accessed through a triple-barrier resonant tunneling heterostructure. Here we report its implementation on a Si substrate; a vital step towards cost-effective mass production. Sample growth using molecular beam epitaxy commenced with deposition of an AlSb nucleation layer to seed the growth of a GaSb buffer layer, followed by the III-V memory epilayers. Fabricated single-cell memories show clear 0/1 logic-state contrast after ≤10-ms duration program/erase pulses of ~2.5 V, a remarkably fast switching speed for 10- and 20-µm devices. Furthermore, the combination of low voltage and small device capacitance per unit area results in a switching energy that is orders of magnitude lower than dynamic random access memory and flash, for a given cell size. Extended testing of devices revealed retention in excess of 1000 years, and degradation-free endurance of over 107 program/erase cycles, surpassing very recent results for similar devices on GaAs substrates.
- 1 October 2021
- Published by Springer Nature in Research Square
Open AccessULTRARAM™: a low-energy, high-endurance, compound-semiconductor memory on silicon
ULTRARAM is a non-volatile memory with the potential to achieve fast, ultralow-energy electron storage in a floating gate accessed through a triple-barrier resonant tunneling heterostructure. Here we report its implementation on a Si substrate; a vital step towards cost-effective mass production. Sample growth using molecular beam epitaxy commenced with deposition of an AlSb nucleation layer to seed the growth of a GaSb buffer layer, followed by the III-V memory epilayers. Fabricated single-cell memories show clear 0/1 logic-state contrast after ≤10-ms duration program/erase pulses of ~2.5 V, a remarkably fast switching speed for 10- and 20-µm devices. Furthermore, the combination of low voltage and small device capacitance per unit area results in a switching energy that is orders of magnitude lower than dynamic random access memory and flash, for a given cell size. Extended testing of devices revealed retention in excess of 1000 years, and degradation-free endurance of over 107 program/erase cycles, surpassing very recent results for similar devices on GaAs substrates.
- Open Access
- 25 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 68 (5) , 2271-2274
ULTRARAM: Toward the Development of a III–V Semiconductor, Nonvolatile, Random Access Memory
ULTRARAM is a III-V compound semiconductor memory concept that exploits quantum resonant tunneling to achieve nonvolatility at extremely low switching energy per unit area. Prototype devices are fabricated in a 2 x 2 memory array formation on GaAs substrates. The devices show 0/1 state contrast from program/erase (P/E) cycles with 2.5 V pulses of 500-μs duration, a remarkable switching speed for a 20 μm gate length. Memory retention is tested for 8 x 10⁴ s, whereby the 0/1 states show adequate contrast throughout, whilst performing 8 x 10⁴ readout operations. Further reliability is demonstrated via program-read-erase-read endurance cycling for 10⁶ cycles with 0/1 contrast. A half-voltage array architecture proposed in our previous work is experimentally realized, with an outstandingly small disturb rate over 10⁵ half-voltage cycles.
- 25 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 68 (5) , 2271-2274
Open AccessULTRARAM: Toward the Development of a III–V Semiconductor, Nonvolatile, Random Access Memory
ULTRARAM is a III-V compound semiconductor memory concept that exploits quantum resonant tunneling to achieve nonvolatility at extremely low switching energy per unit area. Prototype devices are fabricated in a 2 x 2 memory array formation on GaAs substrates. The devices show 0/1 state contrast from program/erase (P/E) cycles with 2.5 V pulses of 500-μs duration, a remarkable switching speed for a 20 μm gate length. Memory retention is tested for 8 x 10⁴ s, whereby the 0/1 states show adequate contrast throughout, whilst performing 8 x 10⁴ readout operations. Further reliability is demonstrated via program-read-erase-read endurance cycling for 10⁶ cycles with 0/1 contrast. A half-voltage array architecture proposed in our previous work is experimentally realized, with an outstandingly small disturb rate over 10⁵ half-voltage cycles.
- Open Access
- 25 March 2020
- Published by IOP Publishing in Semiconductor Science and Technology
- Vol. 35 (5) , 055003
A detailed comparison of measured and simulated optical properties of a short-period GaAs/AlxGa1−xAs distributed Bragg reflector
A 6-period GaAs/Al0.9Ga0.1As distributed Bragg reflector (DBR) has been grown and its optical properties have been both measured and simulated. Incremental improvements were made to the simulation, allowing it to account for internal consistency error, incorrect layer thicknesses, and absorption due to substrate doping to improve simulation accuracy. A compositional depth profile using secondary-ion mass spectrometry (SIMS) has been taken and shows that the Al fraction averages 88.0% ± 0.3%. It is found that the amplitude of the transmission is significantly affected by absorption in the n-doped GaAs substrate, even though the energy of the transmitted light is well below the GaAs band gap. The wavelength of the features in the transmission spectrum are mostly affected by DBR layer thicknesses. On the other hand, the transmission spectrum is found to be relatively tolerant to changes to Al fraction.
- 25 March 2020
- Published by IOP Publishing in Semiconductor Science and Technology
- Vol. 35 (5) , 055003
Open AccessA detailed comparison of measured and simulated optical properties of a short-period GaAs/AlxGa1−xAs distributed Bragg reflector
A 6-period GaAs/Al0.9Ga0.1As distributed Bragg reflector (DBR) has been grown and its optical properties have been both measured and simulated. Incremental improvements were made to the simulation, allowing it to account for internal consistency error, incorrect layer thicknesses, and absorption due to substrate doping to improve simulation accuracy. A compositional depth profile using secondary-ion mass spectrometry (SIMS) has been taken and shows that the Al fraction averages 88.0% ± 0.3%. It is found that the amplitude of the transmission is significantly affected by absorption in the n-doped GaAs substrate, even though the energy of the transmitted light is well below the GaAs band gap. The wavelength of the features in the transmission spectrum are mostly affected by DBR layer thicknesses. On the other hand, the transmission spectrum is found to be relatively tolerant to changes to Al fraction.