Question relating to implant wafer processing

Thread Starter

yt23

Joined Jul 10, 2013
1
Can you restart the process again after the implant process is interrupted? How do you monitor the dose profile after resuming the process?
 

WBahn

Joined Mar 31, 2012
29,978
I don't see any fundamental reason that you couldn't resume the implantation, but I don't know that it would be a good idea or that anyone actually does it (unless it is a planned part of the process). You would have to consider the uncertainties introduced by the stop and start and see if they were tolerable.

You would have to know the implant current and energy profiles before and after the interuption, just like you need to know them over the entire implantation step presently.

If you are talking about a situation in which the implantation was not interupted as part fo the planned processing, for instance if it was interupted due to an equipment failure, then my guess is that they would just scrap the wafer and start over. The cost to complete the fab and subsequent testing is too high to justify the risk at that point.
 

nsaspook

Joined Aug 27, 2009
13,081
Sure you can restart the process (it happens all the time for some minor fault that is usually auto-corrected by software) and keep the uniformity in spec if the implant scan rate is fairly slow. Usually the machine will keep the current dose in a separate dose processor or similar device that's very robust and resistant to power failure or glitches. If the wafer is a special or something valuable it's possible to get a good idea of a unknown dose value if you have some implant profile data from a good implant using a measurement device like the Therma-wave.
http://www.entrepix.com/thermawave/therma-probe/entrepix-therma-probe-tp500xp.php

Typical CMOS implants: http://users.ece.gatech.edu/pallen/Academic/ECE_4430/Summer_2004/DSMCMOSTech(2-UP).pdf
If the implant is something other than a plain-jane junction bulk layer then delays in restarting the dosing can have critical time factors due to substrate annealing, layer damage/recombination/migration and a host of factors that usually get tricker to judge as the line structures gets smaller.

http://www.siliconfareast.com/implant-annealing.htm
 
Last edited:
Top