QCORE Seminar Series · November 19, 2025
Featuring Parvinder Gill
📄 Access the full slide deck (PDF)
Why We Hosted This Talk
One of the most persistent challenges in quantum technology isn’t just building better qubits—it’s building systems that can scale.
As quantum platforms evolve, the bottleneck increasingly shifts toward packaging, interconnects, and integration: how do we reliably connect multiple chips, route optical signals with extremely low loss, and assemble quantum photonic systems in a way that is manufacturable and cost-effective?
In this QCORE seminar, Parvinder Gill introduced the audience to a crucial frontier in quantum engineering: scalable quantum integration through advanced silicon photonic packaging, including the use of photonic wire bonds and 3D printed optical structures.
Her talk explored why photonic packaging has become a limiting factor in the next generation of quantum systems, and how new fabrication techniques—especially two-photon polymerization—are enabling entirely new forms of chip-to-chip and fiber-to-chip connectivity.
About the Speaker
Dr. Parvinder Gill
Parvinder Gill is a faculty member in the Department of Electrical and Computer Engineering at Montana State University, hired as part of the Headwaters Tech Hub project. Her research focuses on silicon photonic packaging, scalable optical interconnects, and emerging integration technologies that support advanced photonic systems—including those needed for quantum photonics.
Her work addresses a key problem in next-generation computing systems: how to build photonic architectures that are not only high-performance, but also manufacturable at scale.
Opening the Talk: Packaging Is the Problem We Still Haven’t Solved
Parvinder begins by stating her goal clearly: she wants to talk about the challenges in silicon photonic packaging, and why those challenges also apply directly to quantum photonic packaging.
She notes that quantum computing research often focuses heavily on qubits and architectures—but the practical reality is that even the best quantum system cannot reach real-world usefulness without solving the physical problem of integration.
That integration challenge is where her research lives.
Integrated Photonics: Miniaturizing Optical Systems Like Electronics
Before diving into packaging, Parvinder provides a brief introduction to integrated photonics for audience members who may be less familiar with it.
Integrated photonics is essentially the miniaturization of optical components—including lasers, detectors, modulators, and couplers—onto a single substrate.
Instead of using electrons for processing and transmission (as in electronic integrated circuits), integrated photonics uses photons.
She explains that photonic chips contain structures such as grating couplers, waveguides, splitters, and channels.
Optical signals propagate through the chip as guided modes, forming a photonic circuit that can manipulate light in highly controlled ways.
Photonic Circuits vs. Electronic Circuits
Parvinder offers a helpful analogy by comparing photonic and electronic circuitry. In electronic circuits:
- data transmission uses electrical signals
- copper wire is the basic building block
- key components include resistors, capacitors, inductors, and transistors
In photonic circuits:
- data transmission uses optical signals
- the waveguide is the basic building block
- key components include resonators, wavelength filters, phase shifters, and fiber couplers
The takeaway is that integrated photonics is not simply “optics made smaller”—it is a complete platform for optical processing and transmission, built with its own fundamental design components.
Photonic Wire Bonds: A New Equivalent to Electrical Wire Bonds
Parvinder then introduces one of the key integration technologies featured in her talk: photonic wire bonds.
She explains that in electronic packaging, wire bonds are a standard solution for connecting multiple chips.
In photonics, a similar concept exists: photonic wire bonds can connect the waveguide of one photonic chip to the waveguide of another.
These are typically made from polymer waveguides, printed or fabricated in a way that forms a physical optical bridge between chips.
This concept becomes critical for quantum photonics, because quantum systems often require, multiple chips, multiple materials, different foundries, and different device functions. All of those must still be connected with extremely low loss.
Why Photonics Matters: The Limits of Electronics
Parvinder explains why integrated photonics is increasingly important.
Electronic circuits face limitations including:
- limited bandwidth
- susceptibility to interference
- higher losses
- higher power consumption due to resistive heating
In contrast, photonic integrated circuits offer:
- higher bandwidth
- immunity to electromagnetic interference
- lower loss transmission
She briefly notes that integrated photonic circuits already support a wide range of applications, and that commercial foundries are producing photonic integrated circuit (PIC) platforms at scale.
Where Quantum Computing Fits Into the Story
Parvinder then transitions into quantum technology.
She makes a clear point: she does not work directly on quantum computing architectures, but her work is highly relevant because packaging is one of the major constraints limiting quantum commercialization.
She references multiple quantum qubit platforms, including, superconducting qubits, trapped ion qubits, neutral atom qubits, and photonic qubits.
All of these platforms have made major progress, but none can reach full commercialization without solving scaling challenges such as limited coherence time, limited number of qubits, high error rates, cryogenic constraints, and miniaturization requirements.
But even beyond those physics challenges, she emphasizes the engineering reality:
To commercialize quantum systems, we must solve scalable, low-cost fabrication and—critically—assembly and packaging.
This is where integrated photonics and quantum packaging overlap directly.
A Typical Quantum Photonic Architecture: Multi-Chip Integration Is Inevitable
Parvinder describes a typical quantum photonic computing architecture: a system that includes:
- a quantum photonic chip
- an electronic control chip
- hybrid bonding between multiple electronic chips
- optical fiber attachments
She explains that a single substrate cannot easily support every component needed in a full quantum system. Different components often require different materials and foundries.
That means scalable quantum photonics requires multi-chip integration. Once multiple chips are required, the packaging problem becomes unavoidable.
A scalable quantum package must be able to integrate components such as:
- photonic integrated circuits
- DFB lasers
- SNSPD detectors
- other specialized optical subsystems
The key problem becomes interconnection:
- fiber-to-chip connectivity
- chip-to-chip connectivity
And for quantum applications, losses must be extremely low—often requiring sub-dB loss—which is difficult to achieve experimentally.
Additionally, quantum systems often operate at cryogenic temperatures, so the interconnects must remain stable and functional under extreme thermal conditions.
The Fiber-to-Chip Problem: A Major Packaging Bottleneck
Parvinder identifies one of the biggest challenges in silicon photonics packaging: connecting a standard optical fiber to an on-chip waveguide.
She explains the mismatch:
- single-mode fiber mode field diameter ≈ 9 microns
- silicon waveguide dimensions can be as small as 500 × 200 nanometers
This creates a massive mode mismatch. If direct butt coupling is attempted, insertion losses can reach 10–12 dB per facet, which is far too high for quantum systems.
Even worse, fiber alignment is extremely sensitive. Misalignment of just a few microns introduces significant loss.
She notes that achieving proper alignment requires six degrees of freedom and placement accuracy better than one micron.
Co-Packaged Optics (CPO): Industry’s Push Toward Scalable Integration
To solve fiber-to-chip connectivity at scale, industry is moving toward co-packaged optics (CPO).
The idea of CPO is to integrate photonic and electronic components within the same package to reduce distance, reduce loss, and improve performance.
She explains that CPO often involves multi-chip modules that use existing electronics packaging infrastructure, including interposers, pick-and-place machines, and standardized electronic packaging workflows
However, this introduces new requirements: fiber ribbon connectors must often be detachable because during electronic packaging steps like flip-chip bonding and reflow, high temperatures are used, and fiber assemblies cannot survive those conditions.
So the packaging process needs connectors that can be attached and detached during assembly.
Industry Solutions: Corning, Intel, IBM and Others
Parvinder mentions that several companies are actively developing CPO solutions, including: Corning, Intel, TeraMount, and IBM.
Some approaches use polymer waveguides to achieve mode conversion, while others use metamaterial mode converters fabricated in silicon.
But she notes a persistent issue: many of these connectors are permanently attached and not detachable.
That limits flexibility during manufacturing.
A Detachable Fiber Ribbon Micro-Connector (Postdoc Work with IBM)
Parvinder then shares work from her postdoctoral research, done in collaboration with IBM, aimed at building a CPO-compatible, cost-effective detachable fiber ribbon micro-connector.
She explains the goal:
Design a connector made from silicon material that is compatible with reflow processing, while remaining detachable and precise.
She describes the system architecture:
- a substrate holds a nanophotonic die
- V-grooves are embedded to guide fiber placement
- fiber ribbons are aligned using these grooves
- a coupling mechanism allows attachment and detachment
She notes that some details cannot be disclosed due to IP restrictions, but the design supports highly precise mating and demating.
She presents results indicating the connector could significantly enhance fiber-to-chip connectivity at low cost.
3D Printed Optics: Two-Photon Polymerization as a Packaging Tool
After addressing detachable connectors, Parvinder transitions to the second major theme of the talk: 3D printed optics for mode conversion and interconnect fabrication.
She introduces a fabrication technique called two-photon polymerization, used in specialized 3D printers such as: Nanoscribe, NanoOne, Vanguard, and Microlight. Montana State University has a NanoOne system in the Spectrum Lab.
This fabrication approach enables researchers to directly print optical structures that act as waveguides, couplers, and bridges between photonic chips.
She cites published work on photonic wire bonds that connect waveguides across chips with very low loss, and highlights research on multicore fibers for high-density optical connections.
Printing Optical Structures Directly on Fiber Tips
Parvinder then presents one of the most innovative sections of the talk: printing photonic structures directly onto the tip of an optical fiber.
In this approach one core is used for signal coupling, another core is used for decoupling, and optical structures are fabricated directly on the fiber surface
A novel waveguide cross-section shows:
- circular core
- air cladding
- supported by a nano-fin
The purpose of this geometry is to reduce substrate coupling and isolate the waveguide from the surrounding material.
She shares simulation results related to taper performance and bending loss, followed by fabrication images showing structures printed on fiber tips.
Ring Resonators on a Dual-Core Fiber Tip
Parvinder then discusses a specific device demonstration: a ring resonator printed on a dual-core fiber tip.
She describes the use of a total internal reflection mirror coupler and shows results from ring resonator measurements.
The quality factor and effective index were estimated for two different coupling gaps: 0.5 microns and 0.6 microns.
The device is presented as a functional demonstration of what 3D printed optics can enable: not just couplers, but integrated photonic devices fabricated directly onto fiber platforms. It was used for temperature sensing, showing a practical sensing application enabled by the technique.
Key Takeaways: Packaging Will Determine the Future of Quantum Photonics
Parvinder closes the seminar with a broader conclusion:
Quantum computing will depend on scalable fabrication, robust packaging approaches, and seamless chip-to-chip interconnection.
She emphasizes that the future will require:
- ultra-low-loss interconnects (sub-dB)
- photonic wire bonds
- 3D printed optical components
- scalable integration of quantum dot single photon sources
- cryogenic-compatible optical interconnects
In other words, the path to scalable quantum photonics will be shaped as much by packaging and integration engineering as by qubit physics.
Acknowledgments
Parvinder acknowledges collaborators and institutions including:
- Hebrew University of Jerusalem
- IBM
- C2MI
- University of Sherbrooke (Quebec, Canada)
She thanks the audience for their attention.
Q&A Highlights
Q1 — Have these photonic wire bond or 3D printed structures been tested at cryogenic temperatures?
Parvinder explains that she has not personally tested them at cryogenic temperatures yet, but notes that other researchers have performed reliability testing at cryogenic conditions and results appear promising. She adds that cryogenic testing is something she plans to pursue at MSU.
Q2 — Can the structure be polarization-maintaining?
Yes, Parvinder says it can be designed for polarization-maintaining behavior. This may require polarization-sensitive experiments and potentially modifying the waveguide cross-section—for example, making it slightly elliptical instead of perfectly circular. She notes that fabrication precision becomes the main challenge.
Q3 — How do you solve mode mismatch between fiber and chip without a grating coupler?
Parvinder explains that grating couplers are not required in this approach. Instead, the mode matching can be achieved using an adiabatic taper, potentially combined with an adiabatic S-bend. This allows the optical mode from the fiber to transition smoothly into the waveguide without major loss.
Q4 — Are grating couplers still an option?
Yes, she notes that grating couplers can still be designed if needed, but fiber-tip freeform optics and adiabatic taper approaches offer an alternative path.
| Ask Parvinder Have a question about silicon photonic packaging, fiber-to-chip coupling, photonic wire bonds, or 3D printed optics for quantum integration? Send it our way—Parvinder may answer it in an upcoming QCORE feature. ➜ Submit a question |
