General

As shown in figure below, there are overlaps between Mechanism Simulation and Rate Prediction. The figure is explained in the Overview Page.

see_sp

As for Rate Prediction, there are two major types：

  • Ground test based (including IRPP, Bendel, PROFIT, J.Barak2006 models, etc.)

  • Device structure based (including NRED, NRED-MC models, etc.)

Ground Test Based

There are relatively mature models based on ground test：

  • For heavy ion：
    • RPP Mode
  • For proton：
    • Bendel Model with 2 parameters
    • PROFIT Model
    • J.Barak 2006 Model

Pros

  • The structure of the device under test doesn't need to be understood.
  • Calculation is simple and time-economic.

Cons

  • DICE or TMR hardened device are not supported.
  • Different sensitive volume sizes (for N and P transistors) are not well supported.
  • Ground tests are expensive.

Device Structure Based

The inner structure of a device can be described by the concept of "Sensitive Volume", which is an artificially defined geometry volume inside the device. Then we can start particle transport simulation, in Monte Carlo method or deterministic method, as shown in figure at the top of this page. The criteria of determining whether a SEE will happen is that the charge electric charge within the sensitive volume produced by the particle exceeds a certain value, which is usually called "Critical Charge" of the sensitive volume.

Pros

  • DICE and TMR hardened device are supported.
  • Different sensitive volume sizes (for N and P transistors) are well supported.
  • Ground test are not always necessary.

Cons

  • The structure of the device is expected to be understood.
  • Calculation is much complicated and time-consuming.

NRED vs. NRED-MC

The difference between NRED and NRED-MC is that the latter uses Monte Carlo method for energetic particle transport and the former uses deterministic method. Generally speaking, Monte Carlo method is expected to be more detailed and consumes more time.

Also for Mechanism Simulation

Because that this category of prediction methods deal with the inner structure of the device, although in a relatively simpler way, they can also be used as mechanism simulation tools.

PFTN's Solution

Based on years of experience of fully physical simulation, Cogenda has thorough understandings of SEE process, thus has a series of solutions for rate prediction. The dedicated rate prediction software of Cogenda is CRad.

As for ground test based methods, Cogenda implemented IRPP model for heavy ion, Bendel model, PROFIT, and J.Barak 2006 Model for proton.

As for device structure based methods, Cogenda developed NRED and NRED-MC.

close