Few questions regarding clarification of STIW.

[Alex Liao]

I know that Shallow trench Isolation width is STIW. It is the distance between devices for isolation.
First, from the following picture it seems that STIW is the distance between the device(surrounded by pplus) and its substrate(nplus). There is only one device in the figure. What about if there are many devices? STIW inserted in between?
figure 1:

Second, from the 2nd following picture, it is to investigate the STI y-axis effect (Not only X-axis effect which is SA/SB). The SA-Y and SB-Y can be seen from it.
figure 2:

It occurs to me that the STI should be right up beside the SAy. That is, all STIs should be on the outer border of green rectangular(active area). If the upper and lower active is only for this one device, both of them should be substrate area. Therefore, STI is outside substrate which contradicts to the conclusion from figure 1.

I guess people from fab. area can clarify this question which conclusion is right. Is the STI outside the substrate or the other way? By the way, if you can give me a crossover plot demonstrating this, that would be nice. Thanks!

 

[erikl]

STI is everywhere active area and no diffusion. See e.g. , especially pp. 25-26 (31-32 of the PDF) and 28 (34).

 

[Alex Liao]

Hi erikl,

STI is everywhere active area and no diffusion.

What is exactly STI is everywhere active area and no diffusion? STI can be located between active area and no-diffusion-area or be located on active area and/or on no-diffusion area? From picture of that manual, it seems STI exists between any combination of (diffusion area, n+, p+).

 

[erikl]

For my understanding active areas are all regions where the field oxide (thick oxide) is removed. In these areas you have doped (p,p+,n,n+) regions and non-doped regions. STI is wherever there is no doping in active areas.

I can't express it more clearly, sorry!

 

[Alex Liao]

Hi Erikl,

This is quite clear now. Wherever there are non-doped regions left, do a STI there. Thanks.