CoChat

Challenges hindering memristive neuromorphic hardware from going mainstream

Open access
Adam, Gina C., Khiat, Ali, Prodromakis, Themis · 2018 · Nature Communications

The DOI matches its Crossref record; 17 of the 17 references it cites were found in a registry or an index.

Abstract

Memristive devices have elicited intense research in the past decade thanks to their inherent low voltage operation, multi-bit storage and cost-effective manufacturability. Nonetheless, several outstanding performance and manufacturability challenges have prevented the widespread industry adoption of redox-based memristive matrices. Here, we discuss these challenges in terms of key metrics and propose a roadmap towards realizing competitive memristive-based neuromorphic processing systems.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-018-07565-4

Cites

17 of 17 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Matrix-level metrics and manufacturing choices impacting them. a–c Variability metric. The variability is a measure of the spread of device performance (in this example, the two extreme resistance states RON and ROFF) in a memristive matrix as defined based on the standard deviation and the means of the resistance distributions (σ/μ). The variability of the resistance states RON and ROFF across a matrix is heavily influenced by a the choice of active material and of the material stack (e.g., single material HfOx vs. bilayer HfOx + TaOx)5, 12; b the device scaling as determined by the smallest feature dimension (also known as critical dimension or CD);6 and c the presence of a series selector/cell which has its own variability profile8. The variability results presented in a–c are extracted from different studies so they have different orders of magnitude depending on the manufacturing process used. d–f Latency metric. The latency is a measure of the delay in accessing the desired device, delay caused by the charging and discharging of the wires. d Impact of the wire downscaling on latency and read margin, which is a measure of the capability to discriminate between the two extreme states (RON and ROFF) of the memristive device13. e Practical matrix size limited by latency vs. the density (number of devices in a μm2) allowed by the critical dimension of the manufacturing process. f The impact of the device / selector non-linearity on latency14. g–i Density metric discussed from the perspective of the most common device designs—crosspoint, plug-via and vertical. g The availability of materials suitable for each device design, given aspects such as uniformity, conformal deposition, etc. h The state-of-the-art scalability for each design (crosspoint: 2 nm CD/12 nm pitch9, plug-via ~30 nm/100 nm15, vertical structure has yet to be optimized for scalability12). i State-of-the-art stackability for each design and its approximate cost per matrix layer (represented by the relative size of the bubble)
Fig. 1 Matrix-level metrics and manufacturing choices impacting them. a–c Variability metric. The variability is a measure of the spread of device performance (in this example, the two extreme resistance states RON and ROFF) in a memristive matrix as defined based on the standard deviation and the means of the resistance distributions (σ/μ). The variability of the resistance states RON and ROFF across a matrix is heavily influenced by a the choice of active material and of the material stack (e.g., single material HfOx vs. bilayer HfOx + TaOx)5, 12; b the device scaling as determined by the smallest feature dimension (also known as critical dimension or CD);6 and c the presence of a series selector/cell which has its own variability profile8. The variability results presented in a–c are extracted from different studies so they have different orders of magnitude depending on the manufacturing process used. d–f Latency metric. The latency is a measure of the delay in accessing the desired device, delay caused by the charging and discharging of the wires. d Impact of the wire downscaling on latency and read margin, which is a measure of the capability to discriminate between the two extreme states (RON and ROFF) of the memristive device13. e Practical matrix size limited by latency vs. the density (number of devices in a μm2) allowed by the critical dimension of the manufacturing process. f The impact of the device / selector non-linearity on latency14. g–i Density metric discussed from the perspective of the most common device designs—crosspoint, plug-via and vertical. g The availability of materials suitable for each device design, given aspects such as uniformity, conformal deposition, etc. h The state-of-the-art scalability for each design (crosspoint: 2 nm CD/12 nm pitch9, plug-via ~30 nm/100 nm15, vertical structure has yet to be optimized for scalability12). i State-of-the-art stackability for each design and its approximate cost per matrix layer (represented by the relative size of the bubble)CC BY · PubMed Central
Fig. 2 Roadmap for manufacturing challenges and possible approaches to accelerate progress. Understanding the underlying factors behind variability can be enabled by data-driven research through lab-to-fab designs and very large-scale integration of memristive matrices with traditional digital access circuitry. Benchmarking (performance metrics, standardized device/matrix sizes, methods of testing, etc) will ensure comparable results between groups. Ultimately, once variability and latency issues are tackled, the technology development will benefit from advanced nano-prototyping techniques, such as extreme ultraviolet lithography, for cost-effective scalability and stackability
Fig. 2 Roadmap for manufacturing challenges and possible approaches to accelerate progress. Understanding the underlying factors behind variability can be enabled by data-driven research through lab-to-fab designs and very large-scale integration of memristive matrices with traditional digital access circuitry. Benchmarking (performance metrics, standardized device/matrix sizes, methods of testing, etc) will ensure comparable results between groups. Ultimately, once variability and latency issues are tackled, the technology development will benefit from advanced nano-prototyping techniques, such as extreme ultraviolet lithography, for cost-effective scalability and stackabilityCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 17 references

Cited by

See all 136

Metadata

Authors
Adam, G. C., Khiat, A., & Prodromakis, T.
Affiliations
  • George Washington University (US)
  • University of Southampton (GB)
Corresponding author
Prodromakis, Themis
Venue
Nature Communications
Published
2018-12-10
Volume
9
Issue
1
Pages
5267-5267
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2903012893
Cited by
136
References
16
Citations by year
2026: 13 · 2025: 27 · 2024: 14 · 2023: 25 · 2022: 14 · 2021: 15 · 2020: 18 · 2019: 10
FWCI
4.10
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • ePrints Soton (University of Southampton) (acceptedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Royal Society
  • Engineering and Physical Sciences Research Council (EP/R024642/1, EP/R024642)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Ferroelectric and Negative Capacitance Devices
  • Physical Sciences › Mathematics › Theoretical Computer Science › DNA and Biological Computing
Keywords
memristors, neuromorphic hardware, multi-bit storage, low-voltage operation, manufacturability, technology roadmap, industrial adoption

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Challenges hindering memristive neuromorphic hardware from going mainstream. https://app.cochat.ai/papers/10.1038/s41467-018-07565-4
The paper
Adam, G. C., Khiat, A., & Prodromakis, T. (2018). Challenges hindering memristive neuromorphic hardware from going mainstream. Nature Communications. https://doi.org/10.1038/s41467-018-07565-4

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.