The Nanoscale World

Z based vs. sidewall based trench depth analysis

rated by 0 users
Answered (Verified) This post has 1 verified answer | 2 Replies | 2 Followers

Top 50 Contributor
21 Posts
Points 220
Robert Melzer posted on Thu, Nov 10 2011 6:51 AM

Hi,

could you please explain, which algorithms are behind Z based and sidewall based trench depth analysis?

Does Z based means histogramm based? We do have challenging images with etch residuals left within the trench, where we do see a dependency between trench depth reported and algorithm used.

Thank you in advance.

Regards, Robert

  • | Post Points: 12

Answered (Verified) Verified Answer

Top 10 Contributor
99 Posts
Points 958
Bruker Employee
Answered (Verified) SeanHand replied on Thu, Nov 10 2011 12:43 PM

Hi Robert -

     Trench depth is an analytical, not statistical, Nanoscope analysis that measures the depth of trenches from the top surface of a trench to either the bottom of the trench or a change in sidewall angle.  This analysis was designed around measuring RC2 DRAM trenches to the top of the trench to the top of the void at which the slope of the sidewall changes from near vertical to near 45 degrees or less.  Changing the analysis from Z based to sidewall based will by design output different results.

     For the application that you describe above, I would suggest a statistical analysis such as Depth, LDA or Layers.  The scumming on the bottom of your trenches will likely smear out the lower of the histogram peaks but you will be able to adjust the peak selection criteria identify the bottom peak as you need.

 

Cheers,

Sean

All Replies

Top 10 Contributor
99 Posts
Points 958
Bruker Employee
Answered (Verified) SeanHand replied on Thu, Nov 10 2011 12:43 PM

Hi Robert -

     Trench depth is an analytical, not statistical, Nanoscope analysis that measures the depth of trenches from the top surface of a trench to either the bottom of the trench or a change in sidewall angle.  This analysis was designed around measuring RC2 DRAM trenches to the top of the trench to the top of the void at which the slope of the sidewall changes from near vertical to near 45 degrees or less.  Changing the analysis from Z based to sidewall based will by design output different results.

     For the application that you describe above, I would suggest a statistical analysis such as Depth, LDA or Layers.  The scumming on the bottom of your trenches will likely smear out the lower of the histogram peaks but you will be able to adjust the peak selection criteria identify the bottom peak as you need.

 

Cheers,

Sean

Top 50 Contributor
21 Posts
Points 220

Thank you Sean for your quick reply.  That information helps a lot.

Btw, we do not have Layers analysis available.

Best Robert

  • | Post Points: 10
Page 1 of 1 (3 items) | RSS
Copyright (c) 2011 Bruker Instruments