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

From LabAdviser
Tigre (talk | contribs)
Tigre (talk | contribs)
Line 60: Line 60:
Below, an explanation of the template 'simple.sdf'. Note that semicolon outcomments the text:
Below, an explanation of the template 'simple.sdf'. Note that semicolon outcomments the text:


<pre>
___________________________________________________________________________
___________________________________________________________________________
MAGAZIN    'SIMPLE1'       The magazine name is SIMPLE1; max. 20 capital letters
MAGAZIN    'SIMPLE1'         The magazine name is SIMPLE1; max. 20 capital letters


#6                                         Cassette from slot no. 6 is used
#6                           Cassette from slot no. 6 is used
%4A                                     Wafer of 4" in position A is exposed
%4A                           Wafer of 4" in position A is exposed
JDF    'simple',1                 Layer block no. 1 of the jdf-file 'simple.jdf' is exposed     
JDF    'simple',1           Layer block no. 1 of the jdf-file 'simple.jdf' is exposed     
ACC 100                               Acceleration voltage of 100keV is used
ACC 100                       Acceleration voltage of 100keV is used
CALPRM '0.2na_ap5'         The condition file 0.2na_ap5 is used  
CALPRM '0.2na_ap5'           The condition file 0.2na_ap5 is used  
DEFMODE 2                         Both deflectors are used (default)
DEFMODE 2                     Both deflectors are used (default)
RESIST 240                         A dose of 240 µC/cm2 is used  
RESIST 240                   A dose of 240 µC/cm2 is used  
SHOT A,8                             The shot step between individual beam shots is 4 nm
SHOT A,8                     The shot step between individual beam shots is 4 nm
OFFSET(0,0)                       An offset of 0 µm is applied in both X and Y
OFFSET(0,0)                   An offset of 0 µm is applied in both X and Y
      
      
END                                       After exposure, the stage is left empty, i.e. the cassette is unloaded
END                           After exposure, the stage is left empty, i.e. the cassette is unloaded


___________________________________________________________________________
___________________________________________________________________________
</pre>


Note that the unit in the SHOT command is in 0.5nm, i.e. 'SHOT A,16' gives a shot pitch of 8nm.
Note that the unit in the SHOT command is in 0.5nm, i.e. 'SHOT A,16' gives a shot pitch of 8nm.

Revision as of 09:45, 18 June 2015

Introduction

An e-beam exposure requires a v30-file, which contains information on the pattern to write, a jobdeck file (jdf-file), and a schedule file (sdf-file). The sdf- and jdf-file contain information about size and position of substrate, dose, current, and shot step. The sdf-file, jdf-file and v30-file are compiled to a magazine-file (mgn) on the 9500 computer.


When a magazine-file is prepared, the actual shot time can be estimated by the e-beam computer (9500).



Preparing a GDS file

If the layout is asymmetric, the conversion might result in an offset of the final layout onto the wafer. It is therefore recommended to symmetrise the layout, e.g. by inserting small structures in the corners of a rectangle that covers the layout.


How to convert to GDS in L-edit:

1 In L-edit, click File/Export Mask Data/GDSII

2 Choose Default GDSII unit: 1 database unit = 0.001 microns


Even though you do not export hidden layers or un-selected cells, unwanted layers might appear in the final GDSII file. Therefore, after conversion til GDSII, check your layout in a different layout viewer, e.g. CleWin (freeware), or BEAMER.

Converting from GDS to v30 (BEAMER)

The gds-file is converted to a v30-file by using the programme BEAMER. Please read the special BEAMER manual for a more advanced description of the BEAMER program.


  1. Open BEAMER.
  2. Drag the ‘Import’-module to the top-right window of Layout Beamer, open the gds-file to convert
  3. If the pattern has abutting polygons and even overlapping polygons which lead to double exposure, incorrect simulations etc., this can be fixed by running a ‘Heal’ program; drag the ‘HEAL’-module on top of the gds-file
  4. Drag the ‘Export’-module on top of the gds-file, export to the location ‘C:\1 gds files\name\date’. Export in JEOL52-format and change field-size to 1000µm x 1000 µm (under 'Advanced’). Click RUN on the Export-icon to execute the whole procedure


The gds-file is now converted to v30-format. In order to check the layout pattern double-click on the v30-module and and a VIEW LAYOUT window will open.

Preparing sdf and jdf files

Sdf- and jdf-files should not be created or edited in a regular notepad or wordpad program. Use only SuperEdi, which is freeware and available on the computers in the e-beam room. When you save a sdf- or jdf-file in SuperEdi, make sure to set the line endings to UNIX.

Both the sdf- and jdf-file should carry names using no capital letters.


You can find templates of sdf- and jdf-files in the folder home/eb0/jeoleb/job/templates on the 9500 computer.

Simple sdf and jdf files

sdf-file

The sdf-file contains a series of commands and parameters that specifies jdf filename, type

of cassette, shot pitch and condition file. It also specifies whether the pattern to be exposed requires alignment.

Below, an explanation of the template 'simple.sdf'. Note that semicolon outcomments the text:

___________________________________________________________________________
MAGAZIN    'SIMPLE1'          The magazine name is SIMPLE1; max. 20 capital letters

#6                            Cassette from slot no. 6 is used
%4A                           Wafer of 4" in position A is exposed
JDF     'simple',1            Layer block no. 1 of the jdf-file 'simple.jdf' is exposed     
ACC 100                       Acceleration voltage of 100keV is used
CALPRM '0.2na_ap5'            The condition file 0.2na_ap5 is used 
DEFMODE 2                     Both deflectors are used (default)
RESIST 240                    A dose of 240 µC/cm2 is used 
SHOT A,8                      The shot step between individual beam shots is 4 nm
OFFSET(0,0)                   An offset of 0 µm is applied in both X and Y
     
END                           After exposure, the stage is left empty, i.e. the cassette is unloaded

___________________________________________________________________________

Note that the unit in the SHOT command is in 0.5nm, i.e. 'SHOT A,16' gives a shot pitch of 8nm. If the next user of the machine has loaded a cassette in slot #10 and you wish to load that cassette after exposure, you can finalize the sdf-file with 'END 10'.

jdf-file

The jdf-file contains a series of commands and parameters that specifies what pattern (v30-file) is to be written where on the sample. It also specifies what calibrations are to be performed during writing.


The jdf-file 'simple.jdf' could look like this:

___________________________________________________________________________
JOB/W  'SIMPLE',4                             4" wafer, jobname is 'SIMPLE', max. 20 capital letters

PATH DRF5M                                    The DRF5M calibration is used (see section 3.3)
    ARRAY   (0,3,2000)/(0,3,3000)             see note 1
        ASSIGN  P(1) -> (*,1)                 Pattern 1 is assigned to row 1-3 coloumn 1
        ASSIGN  P(2) -> (*,2-3)               Pattern 2 is assigned to row 1-3 coloumn 2-3  
        SKIP (1,1)                            Nothing is assigned to chip (1,1)
    AEND                                      End-command to 'ARRAY'
PEND                                          End-command to 'PATH'
 
LAYER   1                                     Start of layer block 1
P(1)  'template1.v30'                         Pattern 1 defined in 'template1.v30'
P(2)  'template2.v30'                         Pattern 2 defined in 'template2.v30'
SPPRM 4.0,,,,1.0,1                            Beam parameters (default - see section 6.2)
STDCUR  0.22 ;nA                              Beam current (in nA) used for writing, see note 2
        
END                                           End of jdf-file
___________________________________________________________________________

Note 1: 9 chips are written, the upper left chip has center in (X,Y) = (0,0). The pitch between the chips is 2 mm in x-direction and 3 mm in y-direction.

Note 2: You should always add 10% to the current defined in 'STDCUR'; this ensures you work well below a scan speed of 100 MHz and thus gives room for a fluctuating current.

___________________________________________________________________________

Examples

20 coloums of chips on a wafer; the first 10 coloums of the array is exposed with one shot pitch, the rest of the chips with another.
An array of 20 chips is written on one 4" wafer. The first 10 coloums of the array is defined in layer block no. 1 of the jdf, coloumn 11-20 is defined in layer block no. 2 of the jdf. The two layer blocks uses different beam shot pitch and base dose. The two layer blocks must be called in two different sequences in the sdf-file.

SDF file


MAGAZIN 'LAYERS'

#1
%4A
JDF     'layers',1
ACC 100
CALPRM '0.2na_ap5'
DEFMODE 2 
OFFSET(0,0)

#1
%4A
JDF     'layers',2
ACC 100
CALPRM '0.2na_ap5'
DEFMODE 2   
OFFSET(0,0)
END 1

JDF FILE

JOB/W  'TWOLAYERS',4 
           
PATH DRF5M
ARRAY   (-10000,20,2000)/( 10000,20,2000) 
  ASSIGN  P(1) -> (1-10,*)
  ASSIGN  P(2) -> (11-20,*)
AEND
PEND
 
LAYER   1

P( 1 )  'test2.v30'
SPPRM 4.0,,,,1.0,1
SHOT A,20
RESIST 220
STDCUR  0.22 ;nA

 
LAYER   2
P( 2 )  'test2.v30'
SPPRM 4.0,,,,1.0,1
SHOT A,40
RESIST 250
STDCUR  0.22 ;nA

END