DOI: 10.1063/1.4991917. NbOx based oscillation neuron for neuromorphic computing

NbOx based oscillation neuron for neuromorphic computing

10.1063/1.4991917

Crossref journal-article

AIP Publishing

Applied Physics Letters (317)

Abstract

In a neuromorphic computing system, the complex CMOS neuron circuits have been the bottleneck for efficient implementation of weighted sum operation. The phenomenon of metal-insulator-transition (MIT) in strongly correlated oxides, such as NbO2, has shown the oscillation behavior in recent experiments. In this work, we propose using a MIT device to function as a compact oscillation neuron, achieving the same functionality as the CMOS neuron but occupying a much smaller area. Pt/NbOx/Pt devices are fabricated, exhibiting the threshold switching I-V hysteresis. When the NbOx device is connected with an external resistor (i.e., the synapse), the neuron membrane voltage starts a self-oscillation. We experimentally demonstrate that the oscillation frequency is proportional to the conductance of the synapse, showing its feasibility for integrating the weighted sum current. The switching speed measurement indicates that the oscillation frequency could achieve &gt;33 MHz if parasitic capacitance can be eliminated.

Bibliography

Gao, L., Chen, P.-Y., & Yu, S. (2017). NbOx based oscillation neuron for neuromorphic computing. Applied Physics Letters, 111(10).

Authors 3

Ligang Gao (first)
Pai-Yu Chen (additional)
Shimeng Yu (additional)

References 24 Referenced 137

{'year': '2015', 'article-title': 'Device and system level design consideration for analog-non-volatile-memory based neuromorphic architectures', 'key': '2023061719132671500_c1'} / Device and system level design consideration for analog-non-volatile-memory based neuromorphic architectures (2015)
{'year': '2015', 'article-title': 'Large-scale neural networks implemented with non-volatile memory as the synaptic weight element: comparative performance analysis (accuracy, speed, and power)', 'key': '2023061719132671500_c2'} / Large-scale neural networks implemented with non-volatile memory as the synaptic weight element: comparative performance analysis (accuracy, speed, and power) (2015)
10.1088/0957-4484/24/38/382001 / Nanotechnology / Synaptic electronics: Materials, devices and applications (2013)
10.1021/nl904092h / Nano Lett. / Nanoscale memristor device as synapse in neuromorphic systems (2010)
10.1109/TED.2015.2440102 / IEEE Trans. Electron Devices / HfO2-based OxRAM devices as synapses for convolutional neural networks (2015)
10.1109/TED.2016.2526647 / IEEE Trans. Electron Devices / Neuromorphic learning and recognition with one-transistor-one-resistor synapses and bistable metal oxide RRAM (2016)
10.1088/0957-4484/26/45/455204 / Nanotechnology / Fully parallel write/read in resistive synaptic array for accelerating on-chip learning (2015)
10.1038/nature14441 / Nature / Training and operation of an integrated neuromorphic network based on metal-oxide memristors (2015)
{'year': '2015', 'article-title': 'Technology-design co-optimization of resistive cross-point array for accelerating learning algorithms on chip', 'key': '2023061719132671500_c9'} / Technology-design co-optimization of resistive cross-point array for accelerating learning algorithms on chip (2015)
10.1109/JETCAS.2015.2426495 / IEEE J. Emerging Sel. Top. Circuits Syst. / Parallel architecture with resistive crosspoint array for dictionary learning acceleration (2015)
10.1109/LED.2012.2229457 / IEEE Electron Device Lett. / Voltage-triggered ultrafast phase transition in vanadium dioxide switches (2013)
10.1088/0957-4484/23/21/215202 / Nanotechnology / Sub-100 fJ and sub-nanosecond thermally driven threshold switching in niobium oxide crosspoint nanodevices (2012)
{'year': '2015', 'article-title': 'High density neuromorphic system with Mo/Pr0.7Ca0.3MnO3 synapse and NbO2 IMT oscillator neuron', 'key': '2023061719132671500_c13'} / High density neuromorphic system with Mo/Pr0.7Ca0.3MnO3 synapse and NbO2 IMT oscillator neuron (2015)
10.1063/1.4921745 / Appl. Phys. Lett. / High-endurance megahertz electrical self-oscillation in Ti/NbOx bilayer structures (2015)
10.1038/nmat3510 / Nat. Mater. / A scalable neuristor built with Mott memristors (2013)
10.1063/1.2930959 / J. Appl. Phys. / High-efficiency voltage oscillation in VO2 planer-type junctions with infinite negative differential resistance (2008)
{'year': '2016', 'article-title': 'Ultra low power coupled oscillator arrays for computer vision applications', 'key': '2023061719132671500_c17'} / Ultra low power coupled oscillator arrays for computer vision applications (2016)
{'year': '2016', 'article-title': 'Compact oscillation neuron exploiting metal-insulator-transition for neuromorphic computing', 'key': '2023061719132671500_c18'} / Compact oscillation neuron exploiting metal-insulator-transition for neuromorphic computing (2016)
10.1088/0957-4484/22/47/475702 / Nanotechnology / Diode-less bilayer oxide (WOx–NbOx) device for cross-point resistive memory applications (2011)
10.1088/0022-3727/48/19/195105 / J. Phys. D: Appl. Phys. / Threshold current reduction for the metal-insulator transition in NbO2-x-selector devices: The effect of ReRAM integration (2015)
10.1016/j.mee.2013.02.084 / Microelectron. Eng. / Threshold-switching characteristics of a nanothin-NbO2-layer-based Pt/NbO2/Pt stack for use in cross-point-type resistive memories (2013)
10.1002/aelm.201600169 / Adv. Electron. Mater. / Multidimensional simulation of threshold switching in NbO2 based on an electric field triggered thermal runaway model (2016)
{'year': '2012', 'article-title': 'Ultrathin (<10\u2009nm) Nb2O5/NbO2 hybrid memory with both memory and selector characteristics for high density 3D vertically stackable rram applications', 'key': '2023061719132671500_c23'} / Ultrathin (<10 nm) Nb2O5/NbO2 hybrid memory with both memory and selector characteristics for high density 3D vertically stackable rram applications (2012)
10.1109/LED.2014.2344105 / IEEE Electron Device Lett. / Reduced threshold current in NbO2 selector by engineering device structure (2014)

Dates

Type	When
Created	7 years, 11 months ago (Sept. 7, 2017, 3:11 p.m.)
Deposited	2 years, 2 months ago (June 17, 2023, 3:13 p.m.)
Indexed	49 minutes ago (Sept. 6, 2025, 2:41 p.m.)
Issued	8 years ago (Sept. 4, 2017)
Published	8 years ago (Sept. 4, 2017)
Published Online	7 years, 11 months ago (Sept. 7, 2017)
Published Print	8 years ago (Sept. 4, 2017)

Funders 1

National Science Foundation 10.13039/100000001
Region: Americas
gov (National government)
Labels4
1. U.S. National Science Foundation
2. NSF
3. US NSF
4. USA NSF
Awards2
1. NSF-CCF-1552687
2. NSF-ECCS-1701565

BibTeX

@article{Gao_2017, title={NbOx based oscillation neuron for neuromorphic computing}, volume={111}, ISSN={1077-3118}, url={http://dx.doi.org/10.1063/1.4991917}, DOI={10.1063/1.4991917}, number={10}, journal={Applied Physics Letters}, publisher={AIP Publishing}, author={Gao, Ligang and Chen, Pai-Yu and Yu, Shimeng}, year={2017}, month=sep }

JSON

{
  "indexed": {
    "date-parts": [
      [
        2025,
        9,
        6
      ]
    ],
    "date-time": "2025-09-06T18:41:19Z",
    "timestamp": 1757184079939,
    "version": "3.41.2"
  },
  "reference-count": 24,
  "publisher": "AIP Publishing",
  "issue": "10",
  "funder": [
    {
      "DOI": "10.13039/100000001",
      "name": "National Science Foundation",
      "doi-asserted-by": "publisher",
      "award": [
        "NSF-CCF-1552687",
        "NSF-ECCS-1701565"
      ],
      "id": [
        {
          "id": "10.13039/100000001",
          "id-type": "DOI",
          "asserted-by": "publisher"
        }
      ]
    }
  ],
  "content-domain": {
    "domain": [
      "pubs.aip.org"
    ],
    "crossmark-restriction": true
  },
  "published-print": {
    "date-parts": [
      [
        2017,
        9,
        4
      ]
    ]
  },
  "abstract": "<jats:p>In a neuromorphic computing system, the complex CMOS neuron circuits have been the bottleneck for efficient implementation of weighted sum operation. The phenomenon of metal-insulator-transition (MIT) in strongly correlated oxides, such as NbO2, has shown the oscillation behavior in recent experiments. In this work, we propose using a MIT device to function as a compact oscillation neuron, achieving the same functionality as the CMOS neuron but occupying a much smaller area. Pt/NbOx/Pt devices are fabricated, exhibiting the threshold switching I-V hysteresis. When the NbOx device is connected with an external resistor (i.e., the synapse), the neuron membrane voltage starts a self-oscillation. We experimentally demonstrate that the oscillation frequency is proportional to the conductance of the synapse, showing its feasibility for integrating the weighted sum current. The switching speed measurement indicates that the oscillation frequency could achieve\u2009&amp;gt;33\u2009MHz if parasitic capacitance can be eliminated.</jats:p>",
  "DOI": "10.1063/1.4991917",
  "type": "journal-article",
  "created": {
    "date-parts": [
      [
        2017,
        9,
        7
      ]
    ],
    "date-time": "2017-09-07T19:11:27Z",
    "timestamp": 1504811487000
  },
  "update-policy": "https://doi.org/10.1063/aip-crossmark-policy-page",
  "source": "Crossref",
  "is-referenced-by-count": 137,
  "title": "NbOx based oscillation neuron for neuromorphic computing",
  "prefix": "10.1063",
  "volume": "111",
  "author": [
    {
      "given": "Ligang",
      "family": "Gao",
      "sequence": "first",
      "affiliation": [
        {
          "name": "School of Electrical, Computer, and Energy Engineering, Arizona State University , Tempe, Arizona 85287, USA"
        }
      ]
    },
    {
      "ORCID": "https://orcid.org/0000-0002-9146-2192",
      "authenticated-orcid": false,
      "given": "Pai-Yu",
      "family": "Chen",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "School of Electrical, Computer, and Energy Engineering, Arizona State University , Tempe, Arizona 85287, USA"
        }
      ]
    },
    {
      "given": "Shimeng",
      "family": "Yu",
      "sequence": "additional",
      "affiliation": [
        {
          "name": "School of Electrical, Computer, and Energy Engineering, Arizona State University , Tempe, Arizona 85287, USA"
        }
      ]
    }
  ],
  "member": "317",
  "published-online": {
    "date-parts": [
      [
        2017,
        9,
        7
      ]
    ]
  },
  "reference": [
    {
      "year": "2015",
      "article-title": "Device and system level design consideration for analog-non-volatile-memory based neuromorphic architectures",
      "key": "2023061719132671500_c1"
    },
    {
      "year": "2015",
      "article-title": "Large-scale neural networks implemented with non-volatile memory as the synaptic weight element: comparative performance analysis (accuracy, speed, and power)",
      "key": "2023061719132671500_c2"
    },
    {
      "key": "2023061719132671500_c3",
      "doi-asserted-by": "publisher",
      "first-page": "382001",
      "DOI": "10.1088/0957-4484/24/38/382001",
      "article-title": "Synaptic electronics: Materials, devices and applications",
      "volume": "24",
      "year": "2013",
      "journal-title": "Nanotechnology"
    },
    {
      "issue": "4",
      "key": "2023061719132671500_c4",
      "doi-asserted-by": "publisher",
      "first-page": "1297",
      "DOI": "10.1021/nl904092h",
      "article-title": "Nanoscale memristor device as synapse in neuromorphic systems",
      "volume": "10",
      "year": "2010",
      "journal-title": "Nano Lett."
    },
    {
      "issue": "8",
      "key": "2023061719132671500_c5",
      "doi-asserted-by": "publisher",
      "first-page": "2494",
      "DOI": "10.1109/TED.2015.2440102",
      "article-title": "HfO2-based OxRAM devices as synapses for convolutional neural networks",
      "volume": "62",
      "year": "2015",
      "journal-title": "IEEE Trans. Electron Devices"
    },
    {
      "issue": "4",
      "key": "2023061719132671500_c6",
      "doi-asserted-by": "publisher",
      "first-page": "1508",
      "DOI": "10.1109/TED.2016.2526647",
      "article-title": "Neuromorphic learning and recognition with one-transistor-one-resistor synapses and bistable metal oxide RRAM",
      "volume": "63",
      "year": "2016",
      "journal-title": "IEEE Trans. Electron Devices"
    },
    {
      "issue": "45",
      "key": "2023061719132671500_c7",
      "doi-asserted-by": "publisher",
      "first-page": "455204",
      "DOI": "10.1088/0957-4484/26/45/455204",
      "article-title": "Fully parallel write/read in resistive synaptic array for accelerating on-chip learning",
      "volume": "26",
      "year": "2015",
      "journal-title": "Nanotechnology"
    },
    {
      "key": "2023061719132671500_c8",
      "doi-asserted-by": "publisher",
      "first-page": "61",
      "DOI": "10.1038/nature14441",
      "article-title": "Training and operation of an integrated neuromorphic network based on metal-oxide memristors",
      "volume": "521",
      "year": "2015",
      "journal-title": "Nature"
    },
    {
      "year": "2015",
      "article-title": "Technology-design co-optimization of resistive cross-point array for accelerating learning algorithms on chip",
      "key": "2023061719132671500_c9"
    },
    {
      "issue": "2",
      "key": "2023061719132671500_c10",
      "doi-asserted-by": "publisher",
      "first-page": "194",
      "DOI": "10.1109/JETCAS.2015.2426495",
      "article-title": "Parallel architecture with resistive crosspoint array for dictionary learning acceleration",
      "volume": "5",
      "year": "2015",
      "journal-title": "IEEE J. Emerging Sel. Top. Circuits Syst."
    },
    {
      "issue": "2",
      "key": "2023061719132671500_c11",
      "doi-asserted-by": "publisher",
      "first-page": "220",
      "DOI": "10.1109/LED.2012.2229457",
      "article-title": "Voltage-triggered ultrafast phase transition in vanadium dioxide switches",
      "volume": "34",
      "year": "2013",
      "journal-title": "IEEE Electron Device Lett."
    },
    {
      "key": "2023061719132671500_c12",
      "doi-asserted-by": "publisher",
      "first-page": "215202",
      "DOI": "10.1088/0957-4484/23/21/215202",
      "article-title": "Sub-100 fJ and sub-nanosecond thermally driven threshold switching in niobium oxide crosspoint nanodevices",
      "volume": "23",
      "year": "2012",
      "journal-title": "Nanotechnology"
    },
    {
      "year": "2015",
      "article-title": "High density neuromorphic system with Mo/Pr0.7Ca0.3MnO3 synapse and NbO2 IMT oscillator neuron",
      "key": "2023061719132671500_c13"
    },
    {
      "key": "2023061719132671500_c14",
      "doi-asserted-by": "publisher",
      "first-page": "212902",
      "DOI": "10.1063/1.4921745",
      "article-title": "High-endurance megahertz electrical self-oscillation in Ti/NbOx bilayer structures",
      "volume": "106",
      "year": "2015",
      "journal-title": "Appl. Phys. Lett."
    },
    {
      "key": "2023061719132671500_c15",
      "doi-asserted-by": "publisher",
      "first-page": "114",
      "DOI": "10.1038/nmat3510",
      "article-title": "A scalable neuristor built with Mott memristors",
      "volume": "12",
      "year": "2013",
      "journal-title": "Nat. Mater."
    },
    {
      "key": "2023061719132671500_c16",
      "doi-asserted-by": "publisher",
      "first-page": "103708",
      "DOI": "10.1063/1.2930959",
      "article-title": "High-efficiency voltage oscillation in VO2 planer-type junctions with infinite negative differential resistance",
      "volume": "103",
      "year": "2008",
      "journal-title": "J. Appl. Phys."
    },
    {
      "year": "2016",
      "article-title": "Ultra low power coupled oscillator arrays for computer vision applications",
      "key": "2023061719132671500_c17"
    },
    {
      "year": "2016",
      "article-title": "Compact oscillation neuron exploiting metal-insulator-transition for neuromorphic computing",
      "key": "2023061719132671500_c18"
    },
    {
      "issue": "47",
      "key": "2023061719132671500_c19",
      "doi-asserted-by": "publisher",
      "first-page": "475702",
      "DOI": "10.1088/0957-4484/22/47/475702",
      "article-title": "Diode-less bilayer oxide (WOx\u2013NbOx) device for cross-point resistive memory applications",
      "volume": "22",
      "year": "2011",
      "journal-title": "Nanotechnology"
    },
    {
      "issue": "19",
      "key": "2023061719132671500_c20",
      "doi-asserted-by": "publisher",
      "first-page": "195105",
      "DOI": "10.1088/0022-3727/48/19/195105",
      "article-title": "Threshold current reduction for the metal-insulator transition in NbO2-x-selector devices: The effect of ReRAM integration",
      "volume": "48",
      "year": "2015",
      "journal-title": "J. Phys. D: Appl. Phys."
    },
    {
      "key": "2023061719132671500_c21",
      "doi-asserted-by": "publisher",
      "first-page": "33",
      "DOI": "10.1016/j.mee.2013.02.084",
      "article-title": "Threshold-switching characteristics of a nanothin-NbO2-layer-based Pt/NbO2/Pt stack for use in cross-point-type resistive memories",
      "volume": "107",
      "year": "2013",
      "journal-title": "Microelectron. Eng."
    },
    {
      "issue": "7",
      "key": "2023061719132671500_c22",
      "doi-asserted-by": "publisher",
      "first-page": "1600169",
      "DOI": "10.1002/aelm.201600169",
      "article-title": "Multidimensional simulation of threshold switching in NbO2 based on an electric field triggered thermal runaway model",
      "volume": "2",
      "year": "2016",
      "journal-title": "Adv. Electron. Mater."
    },
    {
      "year": "2012",
      "article-title": "Ultrathin (&lt;10\u2009nm) Nb2O5/NbO2 hybrid memory with both memory and selector characteristics for high density 3D vertically stackable rram applications",
      "key": "2023061719132671500_c23"
    },
    {
      "issue": "10",
      "key": "2023061719132671500_c24",
      "doi-asserted-by": "publisher",
      "first-page": "1055",
      "DOI": "10.1109/LED.2014.2344105",
      "article-title": "Reduced threshold current in NbO2 selector by engineering device structure",
      "volume": "35",
      "year": "2014",
      "journal-title": "IEEE Electron Device Lett."
    }
  ],
  "container-title": "Applied Physics Letters",
  "original-title": [],
  "language": "en",
  "link": [
    {
      "URL": "http://aip.scitation.org/doi/am-pdf/10.1063/1.4991917",
      "content-type": "application/pdf",
      "content-version": "am",
      "intended-application": "syndication"
    },
    {
      "URL": "https://pubs.aip.org/aip/apl/article-pdf/doi/10.1063/1.4991917/14501983/103503_1_online.pdf",
      "content-type": "application/pdf",
      "content-version": "vor",
      "intended-application": "syndication"
    },
    {
      "URL": "https://pubs.aip.org/aip/apl/article-pdf/doi/10.1063/1.4991917/14501983/103503_1_online.pdf",
      "content-type": "unspecified",
      "content-version": "vor",
      "intended-application": "similarity-checking"
    }
  ],
  "deposited": {
    "date-parts": [
      [
        2023,
        6,
        17
      ]
    ],
    "date-time": "2023-06-17T19:13:36Z",
    "timestamp": 1687029216000
  },
  "score": 1,
  "resource": {
    "primary": {
      "URL": "https://pubs.aip.org/apl/article/111/10/103503/33657/NbOx-based-oscillation-neuron-for-neuromorphic"
    }
  },
  "subtitle": [],
  "short-title": [],
  "issued": {
    "date-parts": [
      [
        2017,
        9,
        4
      ]
    ]
  },
  "references-count": 24,
  "journal-issue": {
    "issue": "10",
    "published-print": {
      "date-parts": [
        [
          2017,
          9,
          4
        ]
      ]
    }
  },
  "URL": "http://dx.doi.org/10.1063/1.4991917",
  "relation": {},
  "ISSN": [
    "0003-6951",
    "1077-3118"
  ],
  "subject": [],
  "published-other": {
    "date-parts": [
      [
        2017,
        9,
        4
      ]
    ]
  },
  "published": {
    "date-parts": [
      [
        2017,
        9,
        4
      ]
    ]
  },
  "article-number": "103503"
}