Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 163 of the 166 references it cites were found in a registry or an index; 30 of the 197 sources its claims cite were checked for support.

Roadmapping the next generation of silicon photonics

Open access
Shekhar, Sudip, Bogaerts, Wim, Chrostowski, Lukas, Bowers, John E., Hochberg, Michael, Soref, Richard, Shastri, Bhavin J. · 2024 · Nature Communications

Abstract

Silicon photonics has developed into a mainstream technology driven by advances in optical communications. The current generation has led to a proliferation of integrated photonic devices from thousands to millions-mainly in the form of communication transceivers for data centers. Products in many exciting applications, such as sensing and computing, are around the corner. What will it take to increase the proliferation of silicon photonics from millions to billions of units shipped? What will the next generation of silicon photonics look like? What are the common threads in the integration and fabrication bottlenecks that silicon photonic applications face, and which emerging technologies can solve them? This perspective article is an attempt to answer such questions. We chart the generational trends in silicon photonics technology, drawing parallels from the generational definitions of CMOS technology. We identify the crucial challenges that must be solved to make giant strides in CMOS-foundry-compatible devices, circuits, integration, and packaging. We identify challenges critical to the next generation of systems and applications-in communication, signal processing, and sensing. By identifying and summarizing such challenges and opportunities, we aim to stimulate further research on devices, circuits, and systems for the silicon photonics ecosystem.

Overview

CoChat Summary

## TL;DR

This perspective paper roadmaps the next generation of silicon photonics, arguing that the field is moving from medium-scale integration into large-scale integration and will need major advances in modulators, lasers, phase shifters, packaging, and photonics-electronics co-integration. It identifies application-specific bottlenecks across communications, sensing, computing, and other systems, and frames the next decade as one of ecosystem maturation and technology diversification.

## Key findings

  • The paper defines silicon photonics generations from SSI to MSI to LSI and notes that silicon photonics is now entering LSI, targeted at 500-to-10,000 components per chip; it also says VLSI (>10,000 components) prototypes have already been demonstrated.
  • For communications, the paper says IMDD transceivers need modulators with lower efficiency figure-of-merit and “-3 dB E/O BW” toward 100 GHz, while lasers need higher wall-plug efficiency; it states that the WPE of most C/L/O band lasers is only about 10%.
  • For phase shifters, the paper argues that thermo-optic heaters are available in all foundries but consume considerable power and cause thermal crosstalk, while NOEMS phase shifters are described as compact, <1 V drive, low-power, and with negligible IL.
  • The summary forecasts that within the next decade, foundries will likely support integrated lasers with WPE exceeding 20%, multi-wavelength and tunable lasers, and coupling losses improving so that typical fiber-to-PIC IL drops to <0.5 dB.

## Methods

  • This is a perspective/roadmap article, not an experimental study; it synthesizes prior literature and industry practice to chart trends and bottlenecks in silicon photonics.
  • The paper organizes the discussion into technology, systems, and applications perspectives, and uses generational framing analogous to CMOS process evolution.
  • Its analysis is mainly qualitative and comparative, weighing integration schemes, device mechanisms, packaging options, and application requirements against PPA, bandwidth, loss, power, and manufacturability constraints.

## Limitations

  • The paper itself states that it is “an attempt” to answer roadmap questions and to stimulate further research, so it is not presenting new primary experimental data.
  • It explicitly says it limits the application discussion to PIC/EIC technology factors and excludes economic, regulatory, market, chemistry, biomarkers, quantum advantage, and other non-technical factors.
  • Evident from the text: many claims are forward-looking and conditional (“we expect,” “likely,” “promising”), so the roadmap’s conclusions are not validated by a single empirical dataset.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41467-024-44750-0

Cites

163 of 163 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Timeline for the number of components on a silicon photonic integrated circuit (PIC) over generations of small-scale, medium-scale, large-scale, and very-large-scale integration (SSI, MSI, LSI, VLSI, respectively). A component is a unit cell that is combined with other unit cells to build a circuit, such as a waveguide, directional coupler, heater, grating coupler, etc. Heterogeneous silicon photonics lags hybrid by approximately two years. For comparison, data for InP-based integrated photonics is also shown. In general, the higher the number of high-speed modulators, the more challenging the scaling. The figure is adapted from refs. 1,2.
Fig. 1 Timeline for the number of components on a silicon photonic integrated circuit (PIC) over generations of small-scale, medium-scale, large-scale, and very-large-scale integration (SSI, MSI, LSI, VLSI, respectively). A component is a unit cell that is combined with other unit cells to build a circuit, such as a waveguide, directional coupler, heater, grating coupler, etc. Heterogeneous silicon photonics lags hybrid by approximately two years. For comparison, data for InP-based integrated photonics is also shown. In general, the higher the number of high-speed modulators, the more challenging the scaling. The figure is adapted from refs. 1,2.CC BY · PubMed Central
Fig. 2 Illustrative renditions of LSI silicon photonic systems capturing current and future technologies. a WDM Transceiver: A semiconductor mode-locked laser (SMLL) provides multi-wavelength continuous-wave (CW) light to an array of compact, WDM-capable modulators and filters. Reflection control circuits limit back reflections into the laser. High-speed photodetectors (PDs) carry out the O/E conversion. b The electrical current is then amplified by transimpedance amplifiers (TIAs) and limiting amplifiers. Analog-to-digital converters (ADCs) are used to digitize the signal for further digital signal processing (DSP). Monitoring PDs are used for control and stabilization of wavelength, phase shift, and temperature. Digital-to-analog converters (DACs) and drivers are used for E/O modulation of the digital signal. Dynamic random-access memory (DRAM) provides large memory access. Micro-controllers (μC) may be used to offload some of the digital processing as well. c LIDAR: A tunable laser provides frequency chirped light to a network of phase shifters, circulators/duplexers and coherent frontend for homodyne/heterodyne frequency-modulated CW (FMCW) ranging and detection. Beam steering is done using optical phase arrays (OPAs) or focal-plane arrays (FPAs). Delay line interferometers aid in calibrating the received beat frequencies and support chirp linearization by directly controlling the tunable laser or a modulator and various forms of error correction through DSP.
Fig. 2 Illustrative renditions of LSI silicon photonic systems capturing current and future technologies. a WDM Transceiver: A semiconductor mode-locked laser (SMLL) provides multi-wavelength continuous-wave (CW) light to an array of compact, WDM-capable modulators and filters. Reflection control circuits limit back reflections into the laser. High-speed photodetectors (PDs) carry out the O/E conversion. b The electrical current is then amplified by transimpedance amplifiers (TIAs) and limiting amplifiers. Analog-to-digital converters (ADCs) are used to digitize the signal for further digital signal processing (DSP). Monitoring PDs are used for control and stabilization of wavelength, phase shift, and temperature. Digital-to-analog converters (DACs) and drivers are used for E/O modulation of the digital signal. Dynamic random-access memory (DRAM) provides large memory access. Micro-controllers (μC) may be used to offload some of the digital processing as well. c LIDAR: A tunable laser provides frequency chirped light to a network of phase shifters, circulators/duplexers and coherent frontend for homodyne/heterodyne frequency-modulated CW (FMCW) ranging and detection. Beam steering is done using optical phase arrays (OPAs) or focal-plane arrays (FPAs). Delay line interferometers aid in calibrating the received beat frequencies and support chirp linearization by directly controlling the tunable laser or a modulator and various forms of error correction through DSP.CC BY · PubMed Central
Fig. 3 Techniques to attach a laser to a silicon PIC. a Conventional laser-isolator (ISO)-fiber-PIC with free-space optics (FSO). b Hybrid 2.5D with FSO. c Hybrid 2.5D with photonic wire bonding (PWB). d Hybrid 3D (flip chip or transfer printing). e Heterogenous (Direct bonding or transfer printing). f Monolithic (Hetero-epitaxy).
Fig. 3 Techniques to attach a laser to a silicon PIC. a Conventional laser-isolator (ISO)-fiber-PIC with free-space optics (FSO). b Hybrid 2.5D with FSO. c Hybrid 2.5D with photonic wire bonding (PWB). d Hybrid 3D (flip chip or transfer printing). e Heterogenous (Direct bonding or transfer printing). f Monolithic (Hetero-epitaxy).CC BY · PubMed Central
Fig. 4 Comparing different techniques to attach a PIC to an electronic IC (EIC). a Electrical wire bonding (EWB) side-by-side. b 2.5D flip-chipped side-by-side or stacked. c Hybrid 3D TSV (Through-Silicon Via). d Heterogenous 3D with TOV (Through-Oxide Via). e Monolithic electronic photonic IC (EPIC).
Fig. 4 Comparing different techniques to attach a PIC to an electronic IC (EIC). a Electrical wire bonding (EWB) side-by-side. b 2.5D flip-chipped side-by-side or stacked. c Hybrid 3D TSV (Through-Silicon Via). d Heterogenous 3D with TOV (Through-Oxide Via). e Monolithic electronic photonic IC (EPIC).CC BY · PubMed Central

Verification

Claims the paper makes

24 Valid · 57 Not valid · 87 Not checked · 142 Own results and interpretations

75 more claims · 9 with their cited sources checked

Create a free account to see all 310 claims and to verify the 16 references still open.

See all claims — free account
  • It may state two separate claims as one. 136
  • Probably not a substantial claim. 165
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Show all 166 references

Research works printed but not linked: 2

  1. Dobbelaere, P. D. Silicon photonics transceivers for hyperscale datacenters: deployment and roadmap. In 2016 European Conference on Optical Communication (2016).
  2. Eschenbaum, C. et al. Thermally stable Silicon-Organic Hybrid (SOH) Mach-Zehnder Modulator for 140 GBd PAM4 transmission with sub-1 V drive signals. In 2022 European Conference on Optical Communication (ECOC), 1–4 (2022).

Cited by

See all 878

Metadata

Authors
Shekhar, S., Bogaerts, W., Chrostowski, L., Bowers, J. E., Hochberg, M., Soref, R., & Shastri, B. J.
Affiliations
  • University of British Columbia (CA)
  • Ghent University (BE)
  • University of California, Santa Barbara (US)
  • University of Massachusetts Boston (US)
  • Queen's University (CA)
Corresponding author
  • Shekhar, Sudip
  • Shastri, Bhavin J.
Venue
Nature Communications
Published
2024-01-25
Volume
15
Issue
1
Pages
751-751
ISSN
2041-1723 (ISSN-L)
OpenAlex
W4391232091
Cited by
878
References
166
Citations by year
2026: 328 · 2025: 392 · 2024: 153
FWCI
19.74
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Ghent University Academic Bibliography (Ghent University) (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • Ghent University Academic Bibliography (Ghent University) (submittedVersion, cc-by)
  • eScholarship (California Digital Library) (submittedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Figshare (submittedVersion, cc-by)
  • arXiv (Cornell University)
Funders
  • Defense Advanced Research Projects Agency
  • Natural Sciences and Engineering Research Council of Canada
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Photonic and Optical Devices
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Semiconductor Lasers and Optical Devices
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Photonic Communication Systems
Keywords
silicon photonics, photonic integration, photonic packaging, data center interconnects, optical signal processing, photonic sensing, photonic circuits, technology roadmap

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

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: Roadmapping the next generation of silicon photonics. https://app.cochat.ai/papers/10.1038/s41467-024-44750-0
The paper
Shekhar, S., Bogaerts, W., Chrostowski, L., Bowers, J. E., Hochberg, M., Soref, R., & Shastri, B. J. (2024). Roadmapping the next generation of silicon photonics. Nature Communications. https://doi.org/10.1038/s41467-024-44750-0

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.