Reciprocity is a fundamental physical principle that applies to a variety of technological fields such as mechanics, acoustics, electronics, and photonics. For photonic systems especially, breaking reciprocity using nonreciprocal materials is a fundamental challenge and opportunity, which is both of scientific interest and of technological importance. This not only allows for the development of key photonic components such as optical isolators and circulators on chip, but also provides novel ways to transport and process data in photonic systems. Over the past several decades, developing integrated nonreciprocal photonic materials has been one of the most challenging and actively studied topics within the photonics research community. In this issue of MRS Bulletin, several representative research directions toward realizing integrated nonreciprocal photonic materials and devices are summarized. The six articles in this issue showcase cutting-edge progress in this field and exciting opportunities for the future.

Optical isolators, devices that only allow unidirectional light propagation, constitute an essential building block for photonic integrated circuits. For near-infrared communications wavelengths, most current isolator designs rely on the incorporation of magneto-optical (MO) materials to break time-reversal symmetry, such as iron garnets or magnetically substituted semiconductors. MO garnets form the backbone of traditional bulk isolators, but suffer from large lattice and thermal mismatch with common semiconductor substrates, which has significantly impeded their integration into on-chip optical isolators. Materials innovations over the past few years have overcome these barriers and enabled monolithic deposition of MO oxide thin films on silicon using techniques such as pulsed laser deposition and magnetron sputtering. On-chip optical isolator devices with polarization diversity in the telecommunication band have been demonstrated based on these materials. This article reviews the latest technological breakthroughs in MO oxide material growth as well as device design and integration strategies toward practical implementation of on-chip optical isolation.

Optical isolators and circulators are important elements in many photonic systems. These nonreciprocal devices are typically made of bulk optical components and are difficult to integrate with other elements of photonic integrated circuits. This article discusses the best performance for waveguide isolators and circulators achieved with heterogeneous bonding. By virtue of the bonding technology, the devices can make use of a large magneto-optical effect provided by a high-quality single-crystalline garnet grown in a separate process on a lattice-matched substrate. In a silicon-on-insulator waveguide, the low refractive index of the buried oxide layer contributes to the large penetration of the optical field into a magneto-optical garnet used as an upper-cladding layer. This enhances the magneto-optical phase shift and contributes greatly to reducing the device footprint and the optical loss. Several versions of silicon waveguide optical isolators and circulators, both based on the magneto-optical phase shift, are demonstrated with an optical isolation ratio of ≥30 dB in a wavelength band of 1550 nm. Furthermore, the isolation wavelength can be effectively tuned over several tens of nanometers.

This article discusses recent studies of on-chip integration of a plasmonic isolator on a Si substrate and a hybrid isolator on an InP substrate. The key characteristics of the plasmonic isolator are reviewed and future prospects are discussed. A method to enhance the magneto-optical figure of merit (FOM) and reduce the propagation loss of a surface plasmon in order to realize the proposed design of the plasmonic isolator is described. One hundred percent enhancement of the FOM and 20× reduction of propagation loss in the optimized ferromagnetic plasmonic structure are experimentally demonstrated.

Faraday rotators in optical isolators, typically composed of iron garnets, are photonic analogues of electrical diodes in that they do not allow reciprocal transmission of light. Isolators are especially important for blocking back-reflected light from reaching source lasers, as such feedback gives rise to unwanted noise and instabilities. In commonly implemented photonic integrated circuits (PICs), isolation is the only critical function that cannot yet be achieved by direct integration. While several techniques have been explored for integrating high-gyrotropy garnets into silicon-on-insulator PICs, this article focuses on sputter deposition, which is the most up-scalable process. High-gyrotropy Ce-doped yttrium iron garnet on nongarnet substrates can be made by sputter deposition with the use of garnet seed layers. Because these seed layers can compromise device performance, seed layer-free terbium iron garnet (TIG) has also recently been developed. Careful doping of TIG can produce Faraday rotations with opposite chiralities, which enable new device designs. Most optical isolator designs involve two-dimensional transverse magnetic-mode structures, such as interferometers or ring resonators, which employ nonreciprocal phase shift. One-dimensional Faraday rotation waveguides with quasi-phase matching have been shown to enable direct integration of isolators for all modes, including the transverse electric mode of lasers currently available for fully integrated PICs.

Lorentz reciprocity governs the symmetry with which electromagnetic signals travel in space and time. A reciprocal channel supports signal transport in two directions with the same transmission properties. Nonreciprocal devices do not obey this general symmetry, and therefore enable isolation and circulation, offering fundamental functionalities in modern GHz-to-THz photonic systems. While most nonreciprocal devices to date are based on magneto-optical phenomena, significant interest has been raised by approaches that avoid the use of magnetic materials, instead relying on artificial materials and circuits that mimic magnetically biased ferrites, enabling compact, light, integrated, and significantly cheaper nonreciprocal devices. Here, we review recent progress in and opportunities offered by artificial nonmagnetic materials that break reciprocity, revealing their potential for compact nonreciprocal devices and systems.

Synthetic photonic materials created by engineering the profile of refractive index or gain/loss distribution, such as negative-index metamaterials or parity-time-symmetric structures, can exhibit electric and magnetic properties that cannot be found in natural materials, allowing for photonic devices with unprecedented functionalities. In this article, we discuss two directions along this line—non-Hermitian photonics and topological photonics—and their applications in nonreciprocal light transport when nonlinearities are introduced. Both types of synthetic structures have been demonstrated in systems involving judicious arrangement of optical elements, such as optical waveguides and resonators. They can exhibit a transition between different phases by adjusting certain parameters, such as the distribution of refractive index, loss, or gain. The unique features of such synthetic structures help realize nonreciprocal optical devices with high contrast, low operation threshold, and broad bandwidth. They provide promising opportunities to realize nonreciprocal structures for wave transport.

The Materials Genome Initiative (MGI) seeks to accelerate the discovery, design, development, and deployment of new materials through the creation of a materials innovation infrastructure. This infrastructure is essentially a system for providing data and tools that encapsulate our existing knowledge about materials, and the means to create new knowledge. Given this approach, MGI is also deeply linked to the ongoing exponential growth in applications of machine learning and artificial intelligence (AI) to materials research. This article explores the connections between MGI, the consequent need for data publication, the implications for data-driven science, and the application of AI to materials design. Examples will demonstrate how materials research is transforming in remarkable ways, and that the MGI vision of accelerated materials discovery is within reach.