About
Martin M. Fejer is a Professor of Applied Physics at Stanford University. His research interests include photonics, with a focus on lasers and accelerators, nonlinear optical materials and devices, guided wave optics, microstructured ferroelectrics and semiconductors, photorefractive phenomena, and optical characterization of materials and material synthesis processes. His work encompasses the development and study of advanced optical materials and devices, contributing to the fields of nanoscience and quantum engineering.
Research topics
- Computer Science
- Physics
- Telecommunications
- Engineering physics
- Materials science
- Optics
- Optoelectronics
- Nanotechnology
- Theoretical physics
- Astrophysics
- Electrical engineering
- Quantum mechanics
- Engineering
Selected publications
Nonlinear Nanophotonics in Thin-Film Lithium Niobate: How Many Octaves, How Few Photons?
2025-06-23
article1st authorCorrespondingNonlinear frequency conversion in lithium niobate waveguides has long been exploited as a means for extending the operating wavelength range and pulse durations/spectral bandwidth of available laser sources. Other applications have included classical and quantum optical signal processing, and generation of nonclassical states of light. Over the past half decade, the emergence of thin-film lithium niobate (TFLN) as a platform for nonlinear nanophotonics has greatly expanded the design space for these devices: The tight optical confinement in TFLN is enabling in several regards: orders of magnitude reductions in the required pump powers, dispersion engineering for ultra-broadband (even multi-octave) operation, and compact devices along with tight bends for dense multi-function integration. The combination of these degrees of freedom has led to a resurgence of interest in guided-wave nonlinear interactions.
Physical Review Materials · 2025-10-31
articleThe mechanical loss angle of amorphous ${\mathrm{TiO}}_{2}$-doped ${\mathrm{GeO}}_{2}$ can be lower than ${10}^{\ensuremath{-}4}$, making it a candidate for Laser Interferometer Gravitational-wave Observatory (LIGO) mirror coatings. Amorphous oxides have complex atomic structures that are influenced by various factors, including doping concentration, preparation, and thermal history, resulting in different mass densities and physical properties. Modeling at the atomistic level enables capturing these effects by generating atomic structure models according to experimental conditions. In order to obtain reliable and physical amorphous models at an affordable cost, we develop classical and machine-learning potentials (MLP) to speed up simulations. First-principles calculations are used to train and validate MLP as well as to validate structure models. To better reproduce properties such as elastic modulus, radial distribution function (RDF), and the variations in mass density of doped amorphous oxides, density functional theory (DFT) calculations are used to optimize the final models. We find that the mass densities of amorphous systems are correlated with the total void volume. The experimental mass density matches the models with the most symmetric potential energy wells under volume change. The elastic response of the metal-oxygen network is also studied. The 27% ${\mathrm{TiO}}_{2}$ doped ${\mathrm{GeO}}_{2}$ system shows the least number of large atom-atom distance changes, while for 44% ${\mathrm{TiO}}_{2}$ doped ${\mathrm{GeO}}_{2}$, a majority of Ti-O distances are significantly changed. In response to strains, the metal-oxygen network at low mass densities prefers to adjust bond angles, while at high mass densities, the adjustment is mainly done by changing atom-atom distance.
2025-01-01
articleThe absorption coefficient of vapor-deposited amorphous silicon is reduced orders of magnitude at near-infrared wavelengths when hydrogenated despite low H incorporation. Raman and ESR measurements suggest H catalyzes structural relaxation of weak Si–Si bonds.
Low-power integrated optical parametric amplification via second-harmonic resonance
Research Square · 2025-09-22
preprintOpen accessSenior authorNavigating the classical-to-quantum transition in ultrafast nonlinear nanophotonics
2025-03-21
articleRecent advances in nonlinear nanophotonics have enabled optical frequency conversion at energy scales remarkably close to the few-photon regime of quantum optics. A key technique in these efforts is dispersion engineering, allowing dynamical confinement of light into ultrashort pulses with high peak intensities. In this talk, we explore quantum effects expected to emerge in next-generation devices, with a focus on understanding how these phenomena interplay with the multimode physics of femtosecond pulses. Drawing inspiration from recent experiments in thin-film lithium-niobate nanophotonics, we present several numerical studies of broadband frequency conversion in chi(2)-nonlinear waveguides. Our analysis provides a detailed understanding of the dynamics and modal structure of quantum noise and entanglement, e.g., in parametric and supercontinuum generation. This talk also surveys some recent results we obtained for harnessing exotic quantum non-Gaussian states of light in ultrafast nonlinear nanophotonics. Proper engineering of these quantum effects may generate entirely new functionalities for nonlinear optics at and beyond the classical-quantum transition.
Low thermal noise mirror coatings utilising titanium dioxide and germanium dioxide mixtures
ArXiv.org · 2025-02-11
preprintOpen accessUpgrades to ground-based gravitational-wave observatories will require mirror coatings with reduced thermal noise, enabling improved detector sensitivity and extended astrophysical reach. Recent studies have shown that optical coatings utilising amorphous materials that exhibit a larger fraction of corner-sharing between adjacent structural units of metal-centered polyhedra are a promising route for reducing mechanical dissipation and thus thermal noise at room temperature. We report on multilayer optical coatings that are fabricated using germanium dioxide mixed with titanium dioxide (TiO$_2$:GeO$_2$) for the high index layers, and silicon dioxide (SiO$_2$) for the low index material. Single layers of TiO$_2$:GeO$_2$ are characterised to optimise the mixture proportion and based on that highly reflective multilayer stacks were deposited. Exceptional optical absorption at 1064 nm below 1 part-per-million (ppm) is observed in the multilayer stacks after heat treatment. The annealing process also induces the formation of blisters which leads to increased optical scattering. However, there is indication that blisters can be suppressed by decreasing the water partial pressure in the deposition chamber. Direct thermal noise measurements provide experimental verification of a significant 25\% reduction of thermal noise over the mirrors currently employed, which combined with sub-ppm levels of optical absorption show the potential of TiO$_2$:GeO$_2$ to improve the sensitivity of gravitational-wave observatories.
Roughness-Limited Performance in Ultra-Low-Loss Lithium Niobate Cavities
ArXiv.org · 2025-05-03
preprintOpen accessAchieving low optical loss is critical for scaling complex photonic systems. Thin-film lithium niobate (TFLN) offers strong electro-optic and nonlinear properties in a compact platform, making it ideal for quantum and nonlinear optics. While $Q$ factors above $10^7$ have been achieved, they remain below the intrinsic material limit. We present a systematic study of scattering losses due to roughness in TFLN racetrack cavities, isolating contributions from sidewall and interface roughness. Quality factors up to $27 \times 10^6$ are demonstrated in waveguides with widths of $2.2λ$ ($\sim3.5\,μ$m), where interface roughness dominates, and up to $1.2 \times 10^7$ in narrower waveguides $0.8λ$ wide ($\sim1.2\,μ$m), where sidewall roughness is the primary limitation. Our modeling framework, based on 3D wave simulations informed by AFM-measured roughness, is material-independent and broadly applicable across integrated photonic platforms.
Optical computing: Revisit an old question with new hardware (TFLN) and software (cEP) perspectives
2025-01-01
articleSenior authorIn this talk we will discuss how the two game changers, TFLN as a new hardware platform and cEP as a new training algorithm, may change a prospect of optical computing and machine learning.
Physical Review Materials · 2025-10-10 · 1 citations
articleHydrogenated $a$-Si films grown by magnetron sputtering have greatly reduced subgap optical absorption, down to 122 ppm (9.0 ${\mathrm{cm}}^{\ensuremath{-}1}$) at 1064 nm, 6 ppm (0.4 ${\mathrm{cm}}^{\ensuremath{-}1}$) at 1550 nm, and 2 ppm (0.1 ${\mathrm{cm}}^{\ensuremath{-}1}$) at 2000 nm for quarter-wavelength-thick films, and they reduced room-temperature mechanical loss below $2\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}5}$. The dangling bond density is reduced to ${10}^{17}\phantom{\rule{4pt}{0ex}}{\mathrm{cm}}^{\ensuremath{-}3}$ and the refraction index remains above that of crystalline silicon for all wavelengths. The above wavelengths are well below the Urbach edge of $a$-Si, and thus absorption is associated with deep-trap states. Although reductions in absorption and dissipation increase with increasing hydrogen content in the working gas, the hydrogen concentration in the films remains low (below 1 at.%), and neither absorption nor mechanical loss correlate with dangling bond density. Raman-derived bond angle deviation measurements of $a$-Si show a reduction in structural disorder as the films are further processed by vacuum annealing and post-hydrogenation, while $a$-Si:H films show a lower disorder in the as-deposited state without further reductions after thermal processing. These results suggest that the presence of hydrogen during film growth, acting as a catalyst, facilitates local rearrangement of atoms and removes strained Si--Si bonds that are more responsible for sub-band-gap absorption and dissipative mechanisms than dangling bonds.
ArXiv.org · 2025-03-27 · 1 citations
preprintOpen accessThe mechanical loss angle of amorphous TiO$_2$-doped GeO$_2$ can be lower than 10$^{-4}$, making it a candidate for Laser Interferometer Gravitational-wave Observatory (LIGO) mirror coatings. Amorphous oxides have complex atomic structures that are influenced by various factors, including doping concentration, preparation, and thermal history, resulting in different mass densities and physical properties. Modeling at atomistic level enables capturing these effects by generating atomic structure models according to experimental conditions. In order to obtain reliable and physical amorphous models at an affordable cost, we develop classical and machine-learning potentials (MLP) to speed up simulations. First-principles calculations are used to train and validate MLP as well as validating structure models. To better reproduce properties such as elastic modulus, radial distribution function (RDF) and the variations in mass density of doped amorphous oxides, density functional theory (DFT) calculations are used to optimize the final models. We find that the mass densities of amorphous systems are correlated with the total void volume. The experimental mass density matches the models with the most symmetric potential energy wells under volume change. The elastic response of the metal-oxygen network is also studied. The 27\% TiO$_2$ doped GeO$_2$ system shows the least number of large atom-atom distance changes, while for 44\% TiO$_2$ doped GeO$_2$, a majority of Ti-O distances are significantly changed. In response to strains, the metal-oxygen network at low mass densities prefers to adjust bond angles, while at high mass densities, the adjustment is mainly done by changing atom-atom distance.
Recent grants
High Throughput Structure Determination for Low Thermal Noise Coatings
NSF · $270k · 2020–2023
Collaborative Research: LSC Center for Coatings Research
NSF · $339k · 2017–2020
Collaborative Research: Stanford-Florida Program in Support of LIGO on Coatings and Core Optics
NSF · $1.4M · 2020–2023
Collaborative Research: LSC Center for Coatings Research
NSF · $254k · 2020–2023
NSF · $250k · 2016–2019
Frequent coauthors
- 282 shared
Carsten Langrock
- 132 shared
Robert L. Byer
Stanford University
- 114 shared
R. K. Route
- 83 shared
Sheila Rowan
- 81 shared
Alan E. Willner
University of Southern California
- 79 shared
Marc Jankowski
Stanford University
- 78 shared
R. Bassiri
University of Glasgow
- 77 shared
J. Hough
Education
- 1985
Ph.D., Physics
Stanford University
- 1981
M.S., Physics
Stanford University
- 1977
B.S., Physics
University of California, Berkeley
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Martin M. Fejer
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup