Access Official Technical PDF

Share Summary:LinkedInEmail

Photonic process control is essential to producing reliable photonic integrated circuits (PICs). As with electronic ICs, a photonic chip depends on a controlled fabrication process and components that behave as expected. But its optical paths and active devices create testing challenges: measurements may need access to both electrical and optical signals, and some platforms require cleaved chip facets before light can be coupled into waveguides.

Process control modules (PCMs) address part of this challenge by providing test structures for monitoring fabrication and characterizing photonic building blocks. Their measurements can help foundries track process variation, investigate problems and build evidence about component performance—information that also supports designers using a process design kit (PDK).

Why photonic process control needs dedicated test structures

In a generic integration model, a foundry provides a standardized process and a library of basic building blocks. Designers combine those blocks to create circuits for different applications. Because the foundry may not know the intended function of every customer design, it cannot rely on testing each customer circuit to assess the process. Instead, dedicated structures provide a consistent way to evaluate the building blocks and the technology behind them.

Testing can take place at several stages. In-process, wafer-level checks can identify process issues and help track variation between runs. After fabrication, wafer or bar-level measurements can characterize optical and electrical performance. Later testing can assess complete circuits or packaged modules. Each stage offers different access and measurement options, so test structures should be designed around the information required and the point in manufacturing when it is available.

Different tests answer different process questions

A useful testing program distinguishes between process measurements and functional characterization. Fabrication test structures can reveal low-level properties such as layer thickness, etch depth, line width or material characteristics. Process control structures, sometimes called parametric monitors, measure values such as sheet resistance, leakage current, continuity or isolation. These results can help diagnose process behavior, but they do not by themselves establish that an optical component meets its functional requirements.

Building block test structures address that next level. They measure the performance of devices such as photodetectors or semiconductor optical amplifiers (SOAs). Results can help characterize components and track wafer-to-wafer variation. Circuit test structures, including standard evaluation circuits, add a further level by checking integrated functionality.

The categories can overlap: one structure may provide both process information and device performance data. The practical objective is to make measurements accurate and useful, and to relate process and geometry data to the behavior of the building blocks.

Characterizing active photonic building blocks

Photodetectors

Photodetectors convert optical power into electrical current, making them a useful example of a building block that can be tested through electrical measurements while being illuminated. Relevant characterization includes dark current—the current measured without optical input—and responsivity, which describes the electrical response to incident optical power. These measurements provide different information: dark current indicates behavior in the absence of light, while responsivity describes the detector’s response when light is present.

SOA gain and absorption

SOAs provide optical amplification, but measuring their performance requires care. A test method based on amplified spontaneous emission (ASE) can assess gain using structures with different amplifier lengths. Since optical coupling can vary, a measurement approach must account for how coupling affects the observed signal; otherwise, coupling differences may be mistaken for changes in amplifier behavior.

Absorption characterization presents a related measurement problem. Test structures can combine an active section with an absorber, and methods may use polarization-resolved measurements or explicit polarization filtering. Separating polarization effects from the signal of interest is important when interpreting the measured transmission and estimating absorption.

Measurement access shapes the test strategy

Photonic circuits carry optical as well as electrical signals. Electrical probing can be automated at wafer scale, but it cannot preserve the spectral detail available from an optical spectrum analyzer. Optical testing provides that information, yet coupling light into and out of a chip can be demanding. For some platforms, access is through cleaved facets, which means more complete optical characterization may need to wait until later fabrication stages. Other platform designs can support vertical optical access, enabling different wafer-level testing arrangements.

These constraints affect test structure design, test timing and the choice of measurement equipment. They also explain why a single test method is unlikely to cover every fabrication stage or every device parameter.

Integrated wavelength measurement as a test function

Wavelength readout is another area where integrated photonic devices can support measurement tasks. An arrayed waveguide grating (AWG) can provide wavelength-dependent passbands for interrogating a fiber Bragg grating (FBG). In this arrangement, the optical response is used to determine wavelength changes. Characterizing the AWG passbands and analyzing the measurement resolution are important parts of evaluating the approach.

This example illustrates a broader point: test structures can assess individual building blocks, while integrated arrangements can demonstrate how those blocks contribute to a measurement function. The intended use determines which level of characterization is most relevant.

From measurements to process improvement

For foundries and system integrators, PCM data is most valuable when it supports clear decisions. Consistent measurements can help teams:

  • Check whether building blocks perform as intended.
  • Track repeatability and variation across wafers or fabrication runs.
  • Investigate process issues using measurements linked to specific structures or parameters.
  • Assess the effects of process changes and support ongoing process development.

Reliable characterization also gives designers a stronger basis for understanding the building blocks represented in a PDK. It does not replace testing of a finished device, but it can help separate process and component questions from application-specific circuit validation.

Conclusion

Photonic process control depends on matching each test structure to a specific question, from fabrication conditions to detector response, SOA behavior and wavelength measurement. Optical access, coupling variation and the stage of manufacturing all influence what can be measured and how results should be interpreted. A structured PCM strategy helps foundries monitor process quality and characterize building blocks, giving designers and technical decision-makers more useful information as photonic circuits move from fabrication toward system-level use.

Access Official Technical PDF

About The Author