Jump to content

Specific Process Knowledge/Lithography/EBeamLithography/eLINE: Difference between revisions

Thope (talk | contribs)
No edit summary
Mmat (talk | contribs)
 
(14 intermediate revisions by 3 users not shown)
Line 1: Line 1:
[[Image:eLINE-Plus.png|350x350px|right|frame|]]
=Raith eLINE Plus system=
<span style="background:#FF2800">THIS PAGE IS UNDER CONSTRUCTION</span>[[image:Under_construction.png|200px]]
[[Image:eLINE-Plus.png|300px|right|]]
 
 
=Process information for the Raith eLINE Plus system=
 


The system was installed in the cleanroom in May 2022. As we get more familiar with the tool this page will be populated with relevant process information.
The system was installed in the cleanroom in May 2022. As we get more familiar with the tool this page will be populated with relevant process information.
 
<br>
 
<br>
<br>
== System overview ==
== System overview ==
 
[[File:RaithHolders.jpg|600px|thumb|left|4 inch wafer holder and 100 mm Universal Sample Holder (USH)]]
 
<br clear=all>
[[File:RaithHolders.jpg|800px|thumb|right|4 inch wafer holder and 100 mm Universal Sample Holder (USH)]]
 


The system is a dual use SEM and EBL exposure tool. For SEM applications the most notable difference from our other SEM's is the automation functionality that allows users to link a design to a substrate and simply from the design define areas to image and the tool will then image those areas without further user input. Once an imaging routine has been setup several hundred images can be acquired per hour. For a brief introduction to this feature see more [https://youtu.be/YoZF_6FeVb4 in this video.]
The system is a dual use SEM and EBL exposure tool. For SEM applications the most notable difference from our other SEM's is the automation functionality that allows users to link a design to a substrate and simply from the design define areas to image and the tool will then image those areas without further user input. Once an imaging routine has been setup several hundred images can be acquired per hour. For a brief introduction to this feature see more [https://youtu.be/YoZF_6FeVb4 in this video.]
Line 22: Line 17:


Currently two holders are available, a 4" wafer holder and a 100 mm Universal Sample Holder (USH) for substrates up to about 75x75 mm. Chips can be clamped on either of the 8 chip clamps on the USH. The sample holders will rest on three ceramic balls on the stage, each ceramic ball fits into one of the kinematic mounts of the sample holder. The kinematic mounts are adjusted to keep the sample holder as level as possible. When handling the sample holders it is important not to touch the kinematic mounts.
Currently two holders are available, a 4" wafer holder and a 100 mm Universal Sample Holder (USH) for substrates up to about 75x75 mm. Chips can be clamped on either of the 8 chip clamps on the USH. The sample holders will rest on three ceramic balls on the stage, each ceramic ball fits into one of the kinematic mounts of the sample holder. The kinematic mounts are adjusted to keep the sample holder as level as possible. When handling the sample holders it is important not to touch the kinematic mounts.


==Dose information==  
==Dose information==  
Line 86: Line 80:
| style="background:lightgrey; color:black" | 180
| style="background:lightgrey; color:black" | 180
|}
|}


==Typical beam currents==
==Typical beam currents==
Line 170: Line 163:


==Writefields==
==Writefields==
Writefield dimension is a trade off between beam shot precision and field stitching. The maximum writefield size is 1000x1000 µm. The beam controller has a limit of 50k addressable positions along each axis and hence for a 1000x1000 µm writefield the minimum beam position grid (pitch) is 20 nm. For a 100x100 µm writefield the minimum beam pitch is 2 nm. Thus the precision is higher for smaller writing fields. Smaller writing fields will however fracture a design into more fields and create more field boundaries with higher potential for stitching errors.
[[Image:WF_imagescan.png|500x500px|right|thumb|WF alignment by image scan.]]
[[Image:WF_linescan.png|500x500px|right|thumb|WF alignment by line scan.]]
 
Writefield (WF) dimension is a trade off between beam shot precision and field stitching. The maximum writefield size is 1000x1000 µm. The beam controller has a limit of 50k addressable positions along each axis and hence for a 1000x1000 µm writefield the minimum beam position grid (pitch) is 20 nm. For a 100x100 µm writefield the minimum beam pitch is 2 nm. Thus the precision is higher for smaller writing fields. Smaller writing fields will however fracture a design into more fields and create more field boundaries with higher potential for stitching errors.
 
 
To minimize stitching errors it is important to perform WF alignment before writing. WF alignment should be done on either a particle or an existing feature close to the writing area. WF alignment can be done in three different ways:
 
 
:1. Manual alignment by image scan to a particle or feature
:2. Automatic alignment by image scan of a particle or feature
:3. Automatic alignment by line scan of feature (preferably a cross)




[[Image:WF_imagescan.png|500x500px|right|thumb|Illustration of WF alignment by image scan.]]
The concept is illustrated in the two drawings and relies on the high precision of the laser interferometer driven stage (about 1 nm precision). When using image scan for WF alignment the user must center the chosen feature or particle in the image field. When executing the WF alignment the stage will displace a bit less than half a WF in one direction. The beam will then deflect back by the same amount and make an image scan. If the WF is perfectly aligned the deflected image will be show the feature in the center of the image. In manual alignment mode the user will indicate any offset, in automatic mode the system will do image analysis and determine the offset. For WF alignment 3-8 different stage positions will be scanned in this way and at the end a correction is calculated and applied.




Prior to exposure the
WF alignment can also be done by line scanning, preferably over a cross from a previous lithography step. The procedure is the same as above except that at each position instead of making an image scan the system makes two line scans and detects each leg of the cross to calculate the center of the cross and in turn the offset of the cross. Again this is done for 3-8 different positions.