Pith. sign in

REVIEW 2 major objections 6 minor 53 references

Monolithic Barium Titanate Nanophotonics and Electro-optics

T0 review · 2 major / 6 minor · reviewed 2026-07-12 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read An anisotropic etch yields low-loss monolithic barium titanate photonic crystals and high-Q cavities with material-limited electro-optic bandwidth near 11 GHz.

desk verdict Solid experimental platform paper: first real low-loss monolithic BTOI 1-D PhCs and 230k FP cavities, with reff~154 pm/V; RF bandwidth claim is only partially isolated from contacts. read the letter →

arxiv 2607.03690 v1 pith:E2C6J6VE submitted 2026-07-04 physics.optics

Sarah Berman , Sina Dereshgi , David Barton This is my paper · ORCID
classification physics.optics
keywords bariumtitanate-on-insulatormonolithicnanophotonicsphotoniccrystalsFabry–PérotcavitiesPockelseffectelectro-opticmodulationhigh-Qresonatorsferroelectricdomainpoling
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Barium titanate on insulator has a very large Pockels coefficient, so it can make compact, low-voltage electro-optic modulators, but only if the optical mode sits inside the barium titanate itself and the film can be etched into sub-wavelength features without destroying optical quality. This paper shows that an optimized dry etch produces 75-degree sidewalls and low roughness, enough to support single-mode waveguides, one-dimensional photonic crystals, and Fabry–Pérot cavities entirely in the barium titanate film. Racetrack resonators reach intrinsic quality factors near one million and propagation losses of about 0.5 dB/cm; photonic crystals open bandgaps with more than 40 dB contrast; and photonic-crystal-mirror cavities reach loaded quality factors up to 230k. After poling, the effective electro-optic coefficient is about 154 pm/V. By modulating at a photonic-crystal band edge rather than a high-Q resonance, the authors measure a 3 dB electro-optic bandwidth of 11 GHz that tracks the known frequency roll-off of the material coefficient rather than cavity lifetime. The same platform also produces asymmetric sidebands, frequency-comb-like spectra under strong drive, and the usual sideband-resolved and unresolved resonant modulation regimes.

What carries the argument

The anisotropic (≈75°) dry etch of thin-film barium titanate that simultaneously preserves high index contrast for photonic-crystal unit cells and low sidewall scattering, enabling both deep bandgaps and high-Q cavities in a single monolithic platform.

What would settle it

A re-measurement of the same band-edge modulator with improved 50-ohm electrode geometry that drives S11 well below –10 dB across 1–25 GHz; if the 3 dB electro-optic bandwidth then moves substantially higher than 11 GHz, the material-limited interpretation is false.

Watch

Extended reading notes

Core claim

A carefully optimized anisotropic dry etch of commercial barium titanate-on-insulator films produces low-loss monolithic nanostructures—one-dimensional photonic crystals with >40 dB bandgap contrast and photonic-crystal Fabry–Pérot cavities with loaded Q up to 230k—while racetracks reach intrinsic Q near 1 million and ~0.5 dB/cm loss. After domain poling, the effective Pockels coefficient is ~154 pm/V, and microwave modulation at the photonic-crystal band edge yields a material-limited 3 dB bandwidth of 11 GHz (6 dB at 21 GHz).

Load-bearing premise

The claim that the measured 11 GHz roll-off is set by the intrinsic frequency dependence of barium titanate’s electro-optic coefficient, rather than by electrode reflections, contact parasitics, or optical filtering at the band edge.

Editorial extensions

If this is right

  • Monolithic BTOI photonic crystals and Fabry–Pérot cavities can shrink modulator capacitance and switching energy relative to hybrid silicon or silicon-nitride overcladding designs.
  • Band-edge and cavity Q can be designed independently, giving a route to slow-light or bandwidth-engineered modulators whose speed is not forced by photon lifetime.
  • The same etch enables visible-wavelength electro-optic devices and dispersion-engineered cavities for nonlinear and quantum photonics on a CMOS-compatible film.
  • Asymmetric sideband generation and multi-line comb spectra at a sharp band edge become available as native functions of the platform without extra filtering optics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because domain writing energy scales with the volume of material that is poled, the same nanostructures that raise light–matter interaction should also lower the energy cost of non-volatile ferroelectric phase shifters or reconfigurable photonic gates.
  • If contact parasitics can be eliminated, the residual material roll-off near 10 GHz becomes the next target for materials engineering (stoichiometry or strain) rather than device geometry.
  • The demonstrated feature sizes and Q values are already in the range needed for Purcell-enhanced coupling to visible solid-state emitters once the platform is extended below 800 nm.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The manuscript reports an optimized anisotropic dry-etch process for monolithic barium titanate-on-insulator (BTOI) that enables low-loss 1-D photonic crystals and photonic-crystal Fabry–Pérot cavities. Racetrack resonators reach intrinsic Q near 1 million and propagation loss of ~0.45 dB/cm; PhCs show >40 dB bandgap contrast; FP cavities reach loaded Q up to 230k. After SHG-verified poling, DC tuning yields reff ≈ 154 pm/V (racetracks) and ~145 pm/V (FPs). Microwave modulation at a PhC band edge gives a 3 dB EO bandwidth of 11 GHz (6 dB at 21 GHz), interpreted as material-limited by the frequency dependence of r42, together with demonstrations of sideband-resolved/unresolved resonator modulation, quasi-single-sideband generation, and high-power asymmetric comb-like spectra.

Significance. Monolithic, high-contrast, low-loss nanostructures have been a missing capability for BTOI relative to LNOI. Demonstrating Qi ~1M, >40 dB PhC contrast, and Q_L = 230k FP cavities with a usable reff ~150 pm/V is a concrete materials-and-process advance that strengthens BTOI as a CMOS-compatible EO platform. The optical metrics are supported by spectra, length-extrapolated loss with error bars, SEM sidewall angle, and SHG poling maps; reff is extracted via a standard formula with stated parameters. These results open a path to compact, low-energy modulators, dispersion-engineered devices, and visible-wavelength EO components, even if the absolute bandwidth claim remains partially RF-packaging limited.

major comments (2)
  1. The central claim that the PhC-band-edge S21 roll-off (Fig. 4b) is set by BTO’s intrinsic r42 frequency dependence rather than cavity lifetime is only partially isolated from electrode parasitics. The paper reports large S11 reflections, contact ringing, and LOESS smoothing of S21 (Fig. 4b and Methods). A de-embedded or better-matched electrode measurement (or a clear upper bound on contact-limited bandwidth independent of the optical device) is needed before the 11 GHz / 21 GHz numbers can be presented as material-limited without qualification.
  2. Eq. (1) for reff depends on a simulated electro-optic overlap Γ_eo = 0.51 (Supplementary Fig. 3) and on n_eff = 1.9. The manuscript should state the uncertainty on Γ_eo (mesh, index, electrode geometry) and show that the extracted 154 pm/V remains consistent under reasonable variation of these inputs; otherwise the comparison to literature values is less robust than claimed.
minor comments (6)
  1. Fig. 4b caption and Methods should explicitly state the LOESS parameters and whether any contact de-embedding was applied, so the smoothed curve can be reproduced.
  2. Inconsistent notation for lattice constant (Λ vs a) appears in the Fig. 2 caption and body text; standardize.
  3. Fig. 1d y-axis includes negative loss values; clarify that these are fit residuals or adjust the plot range.
  4. The abstract and introduction cite bulk r42 > 1200 pm/V; a brief sentence contrasting bulk vs thin-film effective coefficients would help non-specialist readers.
  5. Methods: state the number of devices/resonances averaged for the Qi ~1M and loss figures, and whether the 230k FP Q is a single peak or representative.
  6. Typographical inconsistencies (e.g., “Fabry–Pérot” vs “Fabry-Perot”, missing spaces around units) should be cleaned in a revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: experimental device paper whose headline numbers (Q, loss, reff, S21 bandwidth) are extracted from measured spectra and RF response via standard formulas, not forced by construction from fitted theory.

full rationale

This is a fabrication-and-characterization paper. Propagation loss is obtained by linear extrapolation of racetrack Qi versus straight-section fraction (Fig. 1d, Methods Eq. 2). Loaded/intrinsic Q factors are Lorentzian fits to measured transmission. Photonic-crystal bandgaps and FP resonances are direct transmission spectra. The effective Pockels coefficient is computed from the measured DC tuning slope dλ/dV via the standard resonant-modulator formula (Eq. 1) that also uses a simulated mode-overlap Γeo and measured FSR/Lprobe; the numerical value is therefore an extraction, not a prediction that re-uses the same data under a different name. Microwave bandwidth is a measured S21 at the PhC band edge, merely compared for consistency with literature frequency-dependent r42. No uniqueness theorem, self-definitional identity, or ansatz-smuggled-via-self-citation underpins any central claim. Minor use of simulation (Γeo, bandstructure) and external literature for interpretation is ordinary and does not close a circular loop. Score 0 is therefore appropriate.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

Experimental platform paper. Load-bearing inputs are commercial a–b oriented BTOI film properties, standard Pockels/resonator formulas, simulated EO overlap, and process choices. No new physical entities are postulated; free parameters are process and analysis constants that affect quantitative reff and bandwidth interpretation more than the existence of high-Q PhC devices.

free parameters (5)
  • Γ_eo (electro-optic overlap factor) = 0.51
    Taken as 0.51 from simulation (Supplementary Fig. 3) and inserted into the reff extraction; changes reff linearly.
  • n_eff used in reff formula = 1.9
    Set to 1.9 in Eq. (1); affects absolute reff scale.
  • Etch process setpoints (Ar/CHF3 mix, ICP conditions) = etch rate ~12 nm/min; 75° sidewalls
    Optimized for 75° sidewalls and low roughness; exact flows/powers not fully tabulated but determine whether the PhC contrast and loss claims hold.
  • Poling protocol (voltage, gap, time) = 200 V, 6 µm, 10 min
    200 V across 6 µm for 10 min chosen to saturate SHG; defines the domain state underlying reff.
  • S21 LOESS smoothing / contact de-embedding choices = 3 dB ≈ 11 GHz; 6 dB ≈ 21 GHz
    Bandwidth numbers (11 GHz / 21 GHz) are read from a smoothed curve with acknowledged contact ringing.
assumptions (5)
  • domain assumption Linear Pockels response Δn ∝ r_eff E holds in the stated bias window without significant domain reorientation or higher-order EO effects.
    Used for DC tuning slopes and reff extraction; paper notes depoling risk above ±20 V and restricts FP tuning to +2…+10 V.
  • domain assumption Standard resonator loss and reff formulas (Eqs. 1–2) correctly convert measured Q, FSR, and dλ/dV into α and reff given n_g, Γ_eo, and electrode geometry.
    Central quantitative claims for loss and reff rest on these textbook conversions.
  • domain assumption Commercial a–b oriented BTOI films provide access to a large r42 component when poled and driven in the y–z plane.
    Motivation and poling strategy follow bulk/tensor literature (e.g. Chelladurai et al.).
  • domain assumption PhC band-edge transmission slope converts index modulation into measurable intensity modulation whose RF bandwidth tracks material EO response when cavity photon lifetime is not limiting.
    Justifies using the band edge as a material-bandwidth probe rather than a high-Q resonance.
  • standard math Eigenmode/FDTD/COMSOL simulations with stated geometries adequately predict single-mode TE guidance and bandgap locations for design.
    Design section; experimental bandgaps track period sweeps as expected.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Monolithic Barium Titanate Nanophotonics and Electro-optics." pith.science (2026). https://pith.science/paper/E2C6J6VE

@misc{pith2026260703690,
  author       = {Pith},
  title        = {Pith review of: Monolithic Barium Titanate Nanophotonics and Electro-optics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E2C6J6VE}},
  note         = {Machine review of arXiv:2607.03690}
}
read the original abstract

Barium titanate-on-insulator (BTOI) is a compelling material for high-speed integrated photonic modulators due to its large Pockels coefficient (r42 > 1200 pm/V in bulk), which allows for the miniaturization of modulators while maintaining strong electro-optic performance. Sub-wavelength nanostructures monolithically etched into BTOI are particularly exciting, as they offer a path toward subwavelength light-matter interaction and reduced modulator energy consumption. Here, we design, fabricate, and characterize monolithic one-dimensional nanophotonic crystals (PhCs) and high-Q (230k) photonic-crystal Fabry-Perot (FP) cavities in BTOI. We develop and optimize a nanofabrication process that yields anisotropic (75-degree sidewalls) and deep etching that features low optical loss, with racetrack resonators achieving intrinsic quality factors near 1 million and propagation losses of about 0.5 dB/cm. Our photonic crystals exhibit bandgap contrasts greater than 40 dB, and FP cavities reach loaded quality factors up to 230k. We verify ferroelectric domain alignment via second-harmonic generation microscopy and extract an effective Pockels coefficient of 154 pm/V. By probing the microwave response at the PhC band edge, where modulation bandwidth is set by the material's electro-optic response rather than cavity photon lifetime, we measure a 3-dB electro-optic bandwidth of 11 GHz and a 6-dB bandwidth of 21 GHz, consistent with the frequency-dependent roll-off of BTO's r42 coefficient near 10 GHz. Finally, we show a variety of modulation effects in resonators and at photonic crystal band edges, including sideband-resolved modulation, resonant bandwidth-limited modulation, and photonic-crystal based single sideband modulation and frequency comb generation.

Figures

Figures reproduced from arXiv: 2607.03690 by the authors.

Figure 1
Figure 1. a) SEM images of fabricated structures demonstrating smooth, redeposition-free sidewalls [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. a) SEM image and schematic of a BTO photonic crystal with deliniated design parameters. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. a) SEM of Fabry–Pérot cavities with deposited contacts; inlay describes individual FP mirror [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: a) Modulation setup used for microwave and DC characterization. b) Recorded scattering [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 1 linked inside Pith

  1. [1]

    Zhang, Y., Shu, H., Guo, Y., Zhou, P., Wang, L.,et al.Integrated photonics enabling ultra-wideband fibre–wireless communication.Nature(2026)

    2026

  2. [2]

    & Capmany, J

    Marpaung, D., Yao, J. & Capmany, J. Integrated microwave photonics.Nature Photonics13, 80–90 (2019)

    2019

  3. [3]

    Vandoorne, K.et al.Experimental demonstration of reservoir computing on a silicon photonics chip.Nature Communications5,3541 (2014)

    2014

  4. [4]

    Aharonovich, I., Crozier, K. B. & Neshev, D. Programmable integrated quantum photonics. Nature Photonics20,254–265 (2026)

    2026

  5. [5]

    & Chrostowski, L

    Bogaerts, W. & Chrostowski, L. Silicon Photonics Circuit Design: Methods, Tools and Challenges.Laser & Photonics Reviews12,1700237 (2018)

    2018

  6. [6]

    & Pérez-López, D

    Torrijos-Morán, L. & Pérez-López, D. Industry insight: photonics to scale AI data centers.npj Nanophotonics3,8 (2026)

    2026

  7. [7]

    Watanabe, K., Wu, H.-Y., Xavier, J., Joshi, L. T. & Vollmer, F. Single Virus Detection on Silicon Photonic Crystal Random Cavities.Small18,2107597 (2022)

    2022

  8. [8]

    H., Lee, C.-M., Buyukkaya, M

    Rahaman, M. H., Lee, C.-M., Buyukkaya, M. A., Zhao, Y. & Waks, E. Hybrid Si-GaAs photonic crystal cavity for lasing and bistability.Optics Express31,37574–37582 (2023)

    2023

Show all 53 references
  1. [9]

    Debnath, K.et al.Photonic crystal waveguides on silicon rich nitride platform.Optics Express 25,3214–3221 (2017)

    2017

  2. [10]

    Li, M.et al.Lithium niobate photonic-crystal electro-optic modulator.Nature Communications 11,4123 (2020)

    2020

  3. [11]

    Wang, C.et al.Second harmonic generation in nano-structured thin-film lithium niobate waveguides.Optics Express25,6963–6973 (2017). 10

    2017

  4. [12]

    Liu, X.et al.Ultralow-Threshold Lithium Niobate Photonic Crystal Nanocavity Laser.Nano Letters25,6454–6460 (2025)

    2025

  5. [13]

    Krauss, T. F. Why do we need slow light?Nature Photonics2,448–450 (2008)

    2008

  6. [14]

    Slow light in photonic crystals.Nature Photonics2,465–473 (2008)

    Baba, T. Slow light in photonic crystals.Nature Photonics2,465–473 (2008)

    2008

  7. [15]

    D., Johnson, S

    Joannopoulos, J. D., Johnson, S. G., Winn, J. N. & Meade, R. D.Photonic Crystals: Molding the Flow of Light2nd (Princeton University Press, Princeton, 2008)

    2008

  8. [16]

    Miao, W.-C.et al.Compact Polarization-Multiplexed Optical Transmitters through the Monolithic Integration of Metasurface and Photonic Crystal Surface-Emitting Lasers.ACS Photonics12,4432–4439 (2025)

    2025

  9. [17]

    Muis, D.et al.Broadband localization of light at the termination of a topological photonic waveguide.Science Advances11,eadr9569 (2025)

    2025

  10. [18]

    Yu, M.et al.Integrated femtosecond pulse generator on thin-film lithium niobate.Nature612, 252–258 (2022)

    2022

  11. [19]

    Du, Z.et al.Silicon nitride chirped spiral Bragg grating with large group delay.APL Photonics 5(2020)

    2020

  12. [20]

    & Liu, F

    Wang, Y., Mu, Y., Li, Y., Gao, F. & Liu, F. Photonic-Crystal-Enhanced Single-Photon Sources Based on Semiconductor Quantum Dots.Chip(2026)

    2026

  13. [21]

    Ohta, R.et al.Strong coupling between a photonic crystal nanobeam cavity and a single quantum dot.Applied Physics Letters98(2011)

    2011

  14. [22]

    W.et al.Purcell-enhanced emissions from diamond color centers in slow light photonic crystal waveguides.Nano Letters25,12125–12131 (2025)

    Ding, S. W.et al.Purcell-enhanced emissions from diamond color centers in slow light photonic crystal waveguides.Nano Letters25,12125–12131 (2025)

    2025

  15. [23]

    & Tang, H

    Yang, L., Li, C., Xie, J. & Tang, H. X. High-Q and compact Fabry–Pérot microresonators on thin-film lithium niobate.Nanophotonics14,4675–4681 (2025)

    2025

  16. [24]

    Hwang, H., Go, S., Kim, G.,et al.Electro-optic frequency comb generation in lithium niobate photonic crystal Fabry–Pérot micro-resonator.npj Nanophotonics3,15 (2026)

    2026

  17. [25]

    Chelladurai, D.et al.Barium titanate and lithium niobate permittivity and Pockels coefficients from megahertz to sub-terahertz frequencies.Nature Materials24,868–875 (2025)

    2025

  18. [26]

    I., Yim, J

    Kim, G. I., Yim, J. & Bahl, G. Low loss monolithic barium titanate on insulator integrated photonics with intrinsic quality factor>1 million.arXiv preprint arXiv:2507.17150(2025)

    2025 arXiv

  19. [27]

    Dong, Z.et al.Monolithic Barium Titanate Modulators on Silicon-on-Insulator Substrates.ACS Photonics10,4367–4376 (2023)

    2023

  20. [28]

    B., Stenger, V

    Posadas, A. B., Stenger, V. E., DeFouw, J. D., Warner, J. H. & Demkov, A. A.Electro-Optic Barium Titanate Modulators on Silicon Photonics Platformin2023 IEEE Silicon Photonics Conference (SiPhotonics)(2023), 1–2

    2023

  21. [29]

    Eltes, F.et al.A BaTiO 3-Based Electro-Optic Pockels Modulator Monolithically Integrated on an Advanced Silicon Photonics Platform.Journal of Lightwave Technology37,1456–1462 (2019)

    2019

  22. [30]

    E.et al.Ultra-Low-Power Tuning in Hybrid Barium Titanate–Silicon Nitride Electro-optic Devices on Silicon.ACS Photonics6,2677–2684 (2019)

    Ortmann, J. E.et al.Ultra-Low-Power Tuning in Hybrid Barium Titanate–Silicon Nitride Electro-optic Devices on Silicon.ACS Photonics6,2677–2684 (2019)

    2019

  23. [31]

    Prountzou, E.et al.Electro-optic Modulation in Polycrystalline Barium Titanate Metasurfaces Enhanced by Poling.ACS photonics13,2928–2936 (2026)

    2026

  24. [32]

    Zhu, D.et al.Integrated photonics on thin-film lithium niobate.Advances in Optics and Photonics13,242–352 (2021)

    2021

  25. [33]

    Boes, A.et al.Lithium niobate photonics: Unlocking the electromagnetic spectrum.Science 379,eabj4396 (2023)

    2023

  26. [34]

    R., Khan, S

    Liu, K., Ye, C. R., Khan, S. & Sorger, V. J. Review and perspective on ultrafast wavelength-size electro-optic modulators.Laser & Photonics Reviews9,172–194 (2015)

    2015

  27. [35]

    T., Yi, F., Ho, S.-T

    Lin, P. T., Yi, F., Ho, S.-T. & Wessels, B. W. Two-Dimensional Ferroelectric Photonic Crystal Waveguides: Simulation, Fabrication, and Optical Characterization.Journal of Lightwave Technology27,4330–4337 (2009). 11

    2009

  28. [36]

    Applied Physics Letters86,121111 (2005)

    Melchiorri, M.et al.Propagation losses of silicon nitride waveguides in the near-infrared range. Applied Physics Letters86,121111 (2005)

    2005

  29. [37]

    Laser & Photonics Reviews19,2402086 (2025)

    Raju, A.et al.High-Q Monolithic Ring Resonators in Low-Loss Barium Titanate on Silicon. Laser & Photonics Reviews19,2402086 (2025)

    2025

  30. [38]

    D.et al.A silicon-organic hybrid platform for quantum microwave-to-optical transduction.Quantum Science & Technology5,034004 (2020)

    Witmer, J. D.et al.A silicon-organic hybrid platform for quantum microwave-to-optical transduction.Quantum Science & Technology5,034004 (2020)

    2020

  31. [39]

    & Jiao, Y

    Zhang, S.-M., Cai, L.-T., Jiang, Y.-P. & Jiao, Y. High extinction ratio bandgap of photonic crystals in LNOI wafer.Optical Materials64,203–207 (2017)

    2017

  32. [40]

    Liu, H.et al.Ultra-compact lithium niobate photonic chip for high-capacity and energy-efficient wavelength-division-multiplexing transmitters.Light: Advanced Manufacturing4,133–142 (2023)

    2023

  33. [41]

    Prencipe, A., Baghban, M. A. & Gallo, K. Tunable Ultranarrowband Grating Filters in Thin-Film Lithium Niobate.ACS Photonics8,2923–2930 (2021)

    2021

  34. [42]

    Girouard, P.et al.Enhancement of the pockels effect in photonic crystal modulators through slow light.Optics Letters41,5531–5534 (2016)

    2016

  35. [43]

    Chen, Y.et al.Mono-drive single-sideband modulation via optical delay lines on thin-film lithium niobate.Optica12,666–673 (2025)

    2025

  36. [44]

    Pohl, D.et al.100-GBd waveguide Bragg grating modulator in thin-film lithium niobate.IEEE Photonics Technology Letters33,85–88 (2020)

    2020

  37. [45]

    & Danner, A

    Lin, H.-L., Yin, Y., Roos, L., Dogheche, E. & Danner, A. J. Thermo-optic coefficient of single-crystal thin-film barium titanate on insulator.Optical Materials Express15,2975–2983 (2025)

    2025

  38. [46]

    J.et al.Photonics for artificial intelligence and neuromorphic computing.Nature Photonics15,102–114 (2021)

    Shastri, B. J.et al.Photonics for artificial intelligence and neuromorphic computing.Nature Photonics15,102–114 (2021)

    2021

  39. [47]

    Jha, A.et al.Nanophotonic cavity based synapse for scalable photonic neural networks.IEEE journal of selected topics in quantum electronics28,1–8 (2022)

    2022

  40. [48]

    & Karabchevsky, A

    Katiyi, A. & Karabchevsky, A. Quantum Photonics on a Chip.APL Quantum2,020901 (2025)

    2025

  41. [49]

    C., Srinivasan, K

    Yu, S.-P., Jung, H., Briles, T. C., Srinivasan, K. & Papp, S. B. Photonic-crystal-reflector nanoresonators for Kerr-frequency combs.ACS Photonics6,2083–2089 (2019)

  42. [50]

    Catalá-Lahoz, C.et al.High-speed non-volatile barium titanate field-programmable photonic gate array.Nature Photonics(2026)

    2026

  43. [51]

    Geler-Kremer, J.et al.A ferroelectric multilevel non-volatile photonic phase shifter.Nature Photonics16,491–497 (2022)

    2022

  44. [52]

    E.et al.Effect of Stoichiometry on the Structure and Polarization of BaTiO3

    Cavanagh, A. E.et al.Effect of Stoichiometry on the Structure and Polarization of BaTiO3. Advanced Optical Materials13,2501037 (2025)

    2025

  45. [53]

    Deng, C.et al.Self-buffered epitaxy of barium titanate on oxide insulators enables high-performance electro-optic modulators.Light: Science & Applications15,21 (2026). 12

    2026

Pith tools

Reviewed July 12, 2026 · model on record in the stance chip above.