Intelligent Connectivity

Quantum technologies are advancing quickly and could dramatically transform computing and sensing. However, most quantum systems operate in isolation because they can’t yet be connected into powerful networks. The main challenge to this connectivity is the lack of reliable quantum interconnects, the links that let quantum devices share information (think of the wires, fiber optics and wireless connections in classical computing networks). To overcome this obstacle, we need new switching and routing protocols capable of handling quantum information. Moreover, a quantum internet is vulnerable to noise and signal loss, so it needs real-time error correction to keep information intact.

“A promising solution is to explore classical–decisive quantum systems,” says Liang Feng, Professor in Materials Science and Engineering and in Electrical and Systems Engineering. “At the physical layer, both quantum and classical information would converge on the same electromagnetic guiding mechanism. Such a hybrid system could leverage well-developed classical interconnect hardware, circumventing the need for entirely new quantum-specific infrastructure, while simultaneously providing flexibility to incorporate quantum functionalities.”

Hybrid Networks

As reported in a study published in Science in August, Feng and his team showed that quantum information can be combined with advanced photonic technology, allowing both quantum and classical data to run together on the same chip with high precision. This breakthrough points toward a quantum internet that works smoothly alongside existing communication networks.

---

---

“Our demonstration, incorporating state-of-the-art, high-speed integrated photonic components and synergizing various multiplexing techniques, enables quantum networking using the widely used IP architecture,” says Feng. “As quantum memories become available, they can be readily supported by our system, facilitating the establishment of a practical and resilient quantum internet for long-haul, large-scale distributed quantum information systems in the future.”

The overarching research focus of the Feng lab centers on integrated photonics — integrating photonic components on a chip — and quantum optics — the creation of superposition and entanglement by photons — to design advanced optical systems and explore new ways that light can interact with matter. “This will both deepen our understanding of fundamental quantum physics and facilitate technological breakthroughs for photonic applications, such as photonic communication and photonic computing,” says Feng.

Accelerated Networks

In addition to the group’s Science paper, Feng and his colleagues reported groundbreaking work in several other prestigious journals this year. In a study published in Nature Photonics in January, the researchers developed a hybrid silicon–photonic switch that is very compact, excels in switching time and has the potential to boost on-chip data bandwidth density by approximately three orders of magnitude.

With the rise of data-heavy applications and rapid communications, there’s a growing demand for efficient optical interconnects in data centers, telecommunications and high-performance computing. Optical switches are key, but current designs face trade-offs in speed, size, scalability and bandwidth, making it hard to achieve both fast switching and high data capacity.

By manipulating light at the nanoscale with unprecedented efficiency, the Feng lab’s new switch speeds up the process of getting data on and off the information superhighway of fiberoptic cables that encircles the globe. “Our results highlight a high bandwidth density, optimizing on-chip real estate,” says Feng. “This has the potential to accelerate everything from streaming movies to training AI.”

Nonlinear Networks

In a study published in Nature Photonics in April, Feng’s team unveiled a “field-programmable photonic nonlinearity,” demonstrating that by controlling how light interacts with materials, they can create highly tunable on-chip nonlinear computation functions, which are essential for advanced functions used in signal processing, quantum computing and light-based logic. Their photonic chip also enables programmable nonlinear responses with very low power use, moving the field beyond the limits of traditional, linear photonics.

To demonstrate the utility of this approach, the researchers implemented a simple photonic neural network using their chip. In a task of speech command recognition, they showed that their nonlinear network significantly outperformed an optimized linear on-chip network, which achieved only 86.7% test accuracy, by attaining 91.7% accuracy. Importantly, the simple, flexible design could make scalable optical processors possible, potentially rivaling their counterparts in early-stage electronic neural networks. Using light to control light–matter interactions without changing the material may also inspire more versatile photonic hardware. “Our work serves as a pioneer in the exploration of photonic paradigms specifically tailored for computing with light, accounting for the intrinsic disparities between photons and electrons,” says Feng.

Secure Networks

In a study published in Physical Review X in February, Feng’s team made significant headway toward establishing secure communication networks by generating high-dimensional quantum states with high energy efficiency. Their resulting stable microlaser transmitter shows promise for compact, high-capacity quantum networks. “The ability to encode, manipulate and transmit quantum information using high-dimensional states holds the potential to revolutionize quantum information technologies,” says Feng.

Quantum key distribution (QKD) promises ultimately secure networks, and Feng’s team is leveraging the power of qudits to protect information. While traditional qubits, the commonly used 2D quantum information carriers, are limited to one bit per photon, qudits can carry more information and better resist noise. “We demonstrated high-dimensional QKD using photonic microlaser chips, creating secure keys in real time,” says Feng. “This approach moves us toward compact, energy-efficient quantum networks and could enable long-distance communication from ground to satellite or drone, laying the foundation for next-generation global quantum systems.”

Reflecting on his group’s numerous breakthroughs this year, Feng credits the talented students in Penn Engineering. “Over the past decade, we have relentlessly pursued innovation and have finally made this leap from fundamental theory to device application and system integration,” says Feng. “But we aren’t limiting ourselves to specific devices and systems. Our goal is to uncover entirely new frontiers of science.”

---

Story by Janelle Weaver / Photos by Kevin Monko

---

Inside Engineering

---

Pulling Water From Thin Air

Penn Engineers are developing new materials and devices that can harvest moisture from air without heavy energy inputs.

The first of two advances in this space was led by Daeyeon Lee, Russell Pearce and Elizabeth Crimian Heuer Professor in Chemical and Biomolecular Engineering (CBE), and Amish Patel, Professor in CBE, whose collaboration discovered a new class of nanoporous materials that combine hydrophilic (water-loving) pores with hydrophobic (water-repelling) polymers. These materials require no added energy to passively harvest water through capillary condensation, where water vapor condenses inside tiny pores at moderate humidity and is then released as droplets onto the surface, pictured below.

Because these materials are made from common polymers and fabricated in scalable ways, this work could enable passive water collection in dry conditions, or new approaches to cooling electronics or buildings.

A second water-harvesting breakthrough, led by the lab of Shu Yang, Joseph Bordogna Professor and Chair of Materials Science and Engineering, has yielded a nature-inspired system using “raspberry” microbeads mounted on disks patterned like sunflower seed arrangements, shown here. The raspberry-bead structure (with hygroscopic salt confined inside the hollow nanoparticles, which are embedded in hydrogels) offers both strong vapor capture (because of the salt, and nano/microstructures) and large water retention (thanks to the hydrogel). To release the captured water, the lab uses solar heat, arranging the beads in sunflower-spiral patterns so each bead gets maximum sunlight, minimizing self-shading and improving efficiency. Under real-world conditions, their device produces ~2.39 liters per square meter per day using only ambient humidity and solar heat.

---

From Fungus to Anti-Cancer Compound

Work from the lab of Xue (Sherry) Gao, Presidential Penn Compact Associate Professor in Chemical and Biomolecular Engineering, has turned a notoriously toxic fungus, Aspergillus flavus, into a promising new cancer-treatment lead.

Looking for RiPPs (ribosomally synthesized and post-translationally modified peptides), compounds frequently identified in bacteria but rarely in fungi, the team discovered asperigimycins, an entirely new class of RiPPs.

When enhanced with a lipid and tested against human leukemia cells, one performed on par with FDA-approved drugs while also showing low toxicity toward other cell types. Key to the potential treatment’s selectivity is its dependence on SLC46A3, a gene that facilitates the transport of small molecules, and whose expression varies across cell types. The researchers found that asperigimycins likely work by blocking microtubule formation, disrupting the division of leukemia cells.

---

Departmental Highlights

---