Foundry-Ready Active Design with PhotonForge

Prashanta Kharel, PhD
Published 26 May 2025

In this case study, we demonstrate loading a foundry PDK, performing RF and optical simulations, conducting time-domain analyses, and generating fabrication-ready layouts — all within a single interface. The example features an ultra-high-speed electro-optic modulator in thin-film lithium niobate (TFLN).

 

Background

Electro-optic Mach-Zehnder modulators (EO-MZMs) are essential in photonic integrated circuits (PICs), particularly for high-speed optical communication. These devices modulate light by changing the refractive index of a waveguide in response to an applied electric field—a process that requires careful coordination of electrical and optical properties.

Designing such devices traditionally involves fragmented workflows: separate tools for layout, electrostatic modeling, optical simulation, and data analysis. These disconnected steps can lead to inefficiencies, longer development cycles, and integration errors.

Explore the complete example:  Electro-Optic MZM Example in PhotonForge Documentation

 

Challenges

To design and optimize EO-MZMs, engineers need to:

  • Accurately simulate the electro-optic effect, including voltage-induced refractive index changes.

  • Understand the interaction between RF electrodes and optical waveguides.

  • Quantify performance metrics like extinction ratio, insertion loss, and phase modulation efficiency.

  • Iterate rapidly on device geometry, electrode configuration, and material stack-up—all while ensuring fabrication readiness.

 

Traditional limitations:

  • Separate solvers for electrical and optical domains make co-simulation cumbersome.

  • Manual data exchange between tools increases the chance of error.

  • High computational costs due to 3D field simulations.

  • Limited layout-driven simulation capabilities inhibit rapid prototyping.

 

Solution with PhotonForge

PhotonForge unifies layout, electrostatic, and full-wave optical simulation within one cohesive platform. Using the Tidy3D engine, PhotonForge allows for:

  • Electro-optic co-simulation using voltage-dependent permittivity mapping

  • GPU-accelerated FDTD simulation for rapid optical modeling

  • Scriptable Python interface for parameter sweeps and design automation

  • Drag-and-drop layout tools for fast, fabrication-aligned design creation

  • Direct GDS import/export for seamless foundry integration

  • 3D field visualization of electric and optical field distributions

 

Design details

The EO-MZM design in this case study includes:

  • Y-splitter: An input waveguide splits the optical signal into two arms.
    ../_images/examples_EO-MZM_44_0.svg

  • Phase modulator arms: Each arm contains a straight waveguide surrounded by metal electrodes. A voltage applied to these electrodes induces an electric field that modifies the local refractive index.
    ../_images/examples_EO-MZM_39_0.svg

  • Whole device: Integrate the edge couplers with the MZM and create a complete device.
    ../_images/examples_EO-MZM_56_0.svg

 

Simulation process

  1. Layout construction:
    The entire modulator geometry, including optical waveguides, cladding layers, and electrode geometries, is created in PhotonForge using its layout editor or scriptable Python interface.
    ../_images/examples_EO-MZM_59_0.svg

  2. RF photonic solver:
    PhotonForge calculates the electric field distribution generated by the applied voltage. From this field, the change in refractive index is computed using the linear electro-optic (Pockels) effect for the selected material (e.g., LiNbO₃, silicon, etc.).
    ../_images/examples_EO-MZM_30_1.png

  3. Permittivity mapping:
    The induced index change is transformed into a spatially varying permittivity map, which is then used to perturb the optical domain for accurate co-simulation.../_images/examples_EO-MZM_24_1.png

  4. FDTD Simulation:
    The perturbed structure is simulated using Tidy3D’s GPU-accelerated FDTD solver to analyze how the modulated index distribution affects light propagation.../_images/examples_EO-MZM_50_0.png

  5. Result analysis:
    The output fields are analyzed to determine: insertion loss, extinction ratio (ER), phase shift efficiency, and output intensity modulation vs. input voltage.

 

Results

Using PhotonForge, the simulated EO-MZM achieved:

  • High extinction ratio: ~20–30 dB depending on phase mismatch

  • Insertion loss: Maintained below 1 dB with optimized tapers

  • π-phase shift voltage-length product (VπL): Tunable via electrode spacing and material selection

A full 3D model's complete co-simulation runtime is under 10 minutesThe in-built visualization tools provide 3D overlays of both the optical and electric fields, enabling intuitive debugging and optimization. Cross-sectional slices and field animations provide insights into wave interference dynamics and loss mechanisms.

 

Advantages Over Traditional Tools

Feature Traditional Workflow PhotonForge
Electro-optic co-simulation Manual, disconnected tools Fully integrated
Layout-to-simulation Manual translation Automated, layout-driven
GPU acceleration Often CPU-bound Native GPU support
Design automation Limited Python scripting
3D visualization Basic or missing Advanced GUI with overlays
Foundry-ready PDK support Often custom-built Built-in and customizable

 

Conclusion

This case study illustrates how PhotonForge transforms the workflow for designing and simulating electro-optic modulators. With a unified environment that supports full electro-optic co-simulation, high-performance GPU-based FDTD solvers, and fabrication-ready layouts, engineers can now:

  • Rapidly iterate on device concepts

  • Confidently evaluate performance

  • Streamline handoff to fabrication

Whether you're developing cutting-edge optical interconnects or optimizing modulators for quantum applications, PhotonForge allows you to move faster, with accuracy and confidence.

 

Explore the example

Electro-Optic MZM Example in PhotonForge Documentation

 

You Might Also Like
Simulating a Quantum Photonic Chip Using PhotonForge

Simulating a Quantum Photonic Chip Using PhotonForge

02 Jun 2025 By Prashanta Kharel, PhD