IPKISS.flow

IPKISS.flow is a scripting environment that covers the complete photonic IC design flow up to measurement feedback for true component validation. The components rely on a single, centrally defined model for a smooth transition between the different design stages such as layout and circuit simulation This makes the design flow robust -"What you layout is what you calculate"-, will reduce design errors and will save considerable design time.

Raises the integrity of your design flow

  • fully parametric powerful Python scripting
  • from netlist to layout in the same component
  • circuit simulation & validation by measurement
  • customizable to internal design methodology

 

You will leverage on years of design experience by using the state of the art component library adaptable to different FAB technologies with specialized validated components such as AWGs, Mach-Zehnders, photonic crystals and grating couplers. Using the versatile Python scripting you will be able to centralize your expertise within a custom design flow. This streamlining of the design flow allows for more efficient collaboration between designers, faster design iterations and a more robust knowledge management of previous designs.

Constraining light scheme
ring resonator
ring resonator layout
MZI
Ring loaded Mach Zehnder interferrometer
AWG
AWG
AWG
coupler
heating circuit for tuning
Embedded thumbnail

Key features

Layout
  • Hierarchical component management
  • Photonic waveguide definitions: flexible cross sections, parametric bend algorithms
  • Semi-automatic routing
  • GDSII import/export
Circuit/ System simulation framework
  • CAPHE: Optical Network Solver
  • Layout integrated
  • Frequency simulation
  • S-matrix and custom models.
  • Equation based source definition
  • Multiphysics library (Temperature, free carrier effects)
  • Yield analysis
  • Optimization
  • Interface to Synopsys OptSim Circuit
  • Interface to VPI Componentmaker
Fabrication backend
  • Available PDKs: IMEC Passives, imec iSiPP50G, IHP, IME, VTT
  • Create and combine multiple custom fabrication process flows
  • Generate 2D and 3D Models for rendering and Physical simulation: Povray, HDF5, VTK
Python parametric design framework
  • Python: easy, industry standard scripting language
  • Define building blocks in one place: reduce copy/paste and translation between tools
  • Extract and exchange information between different representations (“views”) from a single definition: layout, 3D model, circuit connectivity, test procedure
  • Interface with third-party tools
  • Optimization, post-processing, visualizations ( numerous scientific Python libraries)
Physical simulation framework
  • CAMFR: Mode solver (2D Cartesian, 3D cylindrical)
  • Interface to CST Studio Suite
  • S-matrix extraction
  • Other solvers on demand

Stories 

Click on the links below to jump to the story

  1. Design and fabrication of a 5x20Gb/s WDM Ge Receiver

  2. The design of fabrication tolerant MZI filters

  3. The optimization of a splitter design combining CST Studio Suite and CAMFR


Design and fabrication of a 5x20Gb/s WDM Ge Receiver

IPKISS application example provided by Ghent University and imec (*).

The challenges are low insertion loss, low crosstalk, polarization insensitivity, compact footprint and low power consumption. Equally important are DfX factors such as design for manufacturability and robustness against temperature variations.

The dense WDM (DWDM) filter has 300GHz (2.34nm) channel spacing around 1540nm wavelength.
A 2-dimensional grating coupler decouples the two orthogonal polarization states of a single mode fiber into their own five-channel 300GHz DWDM filter bank (2nd order ring resonators). A single germanium (Ge) high-speed lateral PIN photodiode terminates each decoupled channel.

The device performed very close to the design specs

  • Open eye diagrams at 20Gb/s on all 5 channels.
  • High manufacturability and yield (wefer-scale measurements).

IPKISS was used to layout and combine 3 different circuits:

  • The optical circuit: The coupling factors of the dual ring resonators (2nd order filters) are optimized through Python programming. The obtained flat-top behavior compensates for local (10mm-1mm) fabrication variations and thus reduces crosstalk.
  • The electrical heating circuit: actively compensates for global wafer-scale variations and ambient temperature variations (very low power).
  • The high frequency contact circuit: creating 5 loops each including a single germanium (Ge) high-speed lateral PIN diode.

IPKISS generated the measurement routines for characterizing the static and dynamic response of this circuit, both optically and electro-optically, at wafer scale.

Click here to learn more.

*P. De Heyn, J. De Coster, P. Verheyen, G. Lepage, M. Pantouvaki, P. Absil, W. Bogaerts, D. Van Thourhout, J. Van Campenhout, Polarization-Insensitive 5x20Gb/s WDM Ge Receiver using Compact Si Ring Filters with Collective Thermal Tuning, Optical Fiber Communication Conference (OFC 2014), United States, p.Th4C.5 (2014)


The design of fabrication tolerant MZI filters.

IPKISS application example provided by Ghent University and imec (*).

Motivation

Si has a high thermo-optic coefficient (shifts of about 70-100 pm/K on resonance), and is extremely fabrication sensitive (i.e. 1 nm in resonance / nm wire width change and 1.4 nm / nm in thickness).

Proposed solution

Passive compensation by optimizing lengths and widths to reduce fabrication and thermal sensitivity. The width sensitivity is reduced by 10 times, the thermal sensitivity by 8 times and the footprint is increased by 2.5 times. CAPHE is used to create a thermal model of the waveguides and MZI as to model the fabrication variability and for parameter exploration.

Design Advantages.

  • Low-power: No temperature compensation needed

  • CMOS compatibility: All-silicon approach

  • No post fabrication required: low cost

IPKISS Circuit Simulation Advantages

  • Layout and circuit simulation fully coupled. Reducing design errors

  • Easy implementation of custom models for n; width and temperature dependent

  • Easy parameter exploration

  • Fast simulation (<1min) of the MZI over the whole wavelength range

* S. Dwivedi, H.  D’heer and W. Bogaerts, Ghent University - IMEC, Gent, Belgium, Group IV Photonics, Paris, ThP1 (2014)


The optimization of a splitter design combining CST Studio Suite and CAMFR

IPKISS.flow integrates with the CST Studio Suite to assure an automatic, consistent management of PDK, layout and coupled simulation data.

Incorporating a simulation model takes 4 easy steps from within IPKISS.flow:

  1. Export geometry + ports to CST Studio Suite
  2. Start the CST Simulation and S-parameter extraction
  3. Import S-parameters in IPKISS
  4. Circuit simulation in IPKISS

Example objective

We design a 1x2-splitter with a 50:50 percent splitting ratio and optimize it for minimal insertion loss.
In order to do this 3 steps are needed:

  1. Create a parameterized PCell for a splitter in such a way that its performance can be optimized.
  2. Optimize the splitter using a standard Python optimizer and CAMFR, a fast modal solver integrated in IPKISS.
  3. Run an accurate 3D-FIT/FDTD simulation  using the CST Studio Suite to extract an accurate s-matrix model for CAPHE circuit simulation.

Click here for more information on the photonics simulation capabilities of the CST Studio Suite

Documentation

IPKISS.flow 3.0

Click here to access the full documentation online

Demo icon
Get a
Trial Version

Starting from our extensive validated component library, we can assist you in the creation and validation of new model libraries