Specific Process Knowledge/Lithography/EBeamLithography/BEAMER: Difference between revisions
Line 4: | Line 4: | ||
The BEAMER software is used to convert files from GDSII format to v30-format. The BEAMER software is manufactured by GenISys (www.genisys-gmbh.de) and the software is installed on the ebprep computer. | The BEAMER software is used to convert files from GDSII format to v30-format. The BEAMER software is manufactured by GenISys (www.genisys-gmbh.de) and the software is installed on the ebprep computer. | ||
BEAMER is endowed with a software that corrects for proximity errors in the e-beam exposure. You can read more about this function in the BEAMER manual [[Specific_Process_Knowledge/Lithography/EBeamLithography/BEAMER|here]] and in the BEAMER presentation here [[media:BEAMERPresentation.pdf|BEAMERPresentation.pdf]]. | |||
The proximity error correction require a forward and a backward range parameter, alfa and beta, and a ratio of backscattered energy to the forward scattered energy, eta. As alfa depends on the electron acceleration voltage, which is constant at 100kV, alfa is in BEAMER fixed to 0.007. Help to find beta and eta can be found [http://nanolithography.gatech.edu/proximity.htm here]. | |||
Alternatively, a [http://en.wikipedia.org/wiki/Proximity_effect_%28electron_beam_lithography%29 point-spread function] can be used in BEAMER to calculate the optimised dose-variation. | |||
Revision as of 09:30, 8 September 2022
Feedback to this page: click here
Introduction
The BEAMER software is used to convert files from GDSII format to v30-format. The BEAMER software is manufactured by GenISys (www.genisys-gmbh.de) and the software is installed on the ebprep computer.
BEAMER is endowed with a software that corrects for proximity errors in the e-beam exposure. You can read more about this function in the BEAMER manual here and in the BEAMER presentation here BEAMERPresentation.pdf.
The proximity error correction require a forward and a backward range parameter, alfa and beta, and a ratio of backscattered energy to the forward scattered energy, eta. As alfa depends on the electron acceleration voltage, which is constant at 100kV, alfa is in BEAMER fixed to 0.007. Help to find beta and eta can be found here.
Alternatively, a point-spread function can be used in BEAMER to calculate the optimised dose-variation.
The main functions of BEAMER are:
Module Function Import Imports a GDS, CIF or v30-file Export Exports to GDS, CIF or v30-format Grid Modifies the base unit of the imported file Fracture Optimizes the fracturing of the layout Heal Removes overlaps in the layout and joins abutting polygons Bias Adds or subtracts geometrical bias to the pattern Merge Merges two layouts Minus Subtracts two layouts PEC Proximity Error Correction Split Splits the output in two
To use the modules, they are dragged to the process flow area.
The inputs and outputs of each module are illustrated by white tags. Two modules are coupled to each other by positioning the input of e.g. the Export module on top of the output of the Import module. The input/output tags turns black when they are connected.
By clicking on active modules, the layout of the file becomes visible in the layout view area.
Simple conversion from GDSII to v30
Import a GDSII-file by dragging the 'Import' module to the process flow area. Select the file. In the 'Import Layout' window, type the layer number to import or type * if you wish to import all layers. | Export the GDSII file to v30 by dragging an 'Export' module on top of the 'Import' icon. An 'Export JEOL' window appears. | In the tab 'General' of this export window, choose JBX9500FS (100kV). In the tab 'Advanced', type the correct field size (maximum is 1000 mm in both X and Y). The fields will be distributed automatically by choosing 'Fixed'. In case you have small patterns, is it recommended to tick 'Center to Field'. |
If your pattern is larger than one field size, it will be stitched by several fields. In the tab 'Multipass', you can choose to smooth the stitching errors by overlapping the fields to be stitched or by multipass method.
There are 3 types of overlap methods; for all 3 methods you define an overlap width:
a) standard: the overlap will be exposed with half dose both of field 1 and field2;
b) interleaving: the overlap will be finger-jointed by field 1 and field 2;
c) interleaving with extra field: the overlap will be finger-jointed by field 1, field 2 and an extra field 3, as illustrated to the right.
The multipass method smooths the field stitching errors by writing the pattern with half dose in different areas of the main deflector; by this method, pattern located in the corner of a writing field will be written partly by the corner of a writing field and partly by the central part of the writing field.
Dose variations
Dose variation defined by datatype or layertype
When designing your pattern, you can define a dose variation either by defining different areas in different GDS datatype or in different layertype.
In both cases you import the GDS-file as usual into BEAMER, importing all layers. The layers should be visible when you view the GDSfile:
If you defined different doses by different GDS layers, you should tick 'Datatype to Class' under 'Advanced' when you convert to v30.
If you defined different doses by different layer type, you should tick 'Layer to Class' under 'Advanced' when you convert to v30.
View the v30-file and make sure all variations have been converted correctly.
In the jdf-file, manually create a shot modulation (similar to the jdi-file created by PEC) that fits to the number of layers in your v30-file. This shot modulation should be defined in the end of the jdf-file (see the sdf- and jdf-file manual). In the example below, the three layers (green, blue and red in the BEAMER viewer to the right) are defined to have shot modulation from -20 % to + 20 % of the base-dose:
JOB/W 'TEST',4 ; 4 inch wafer PATH DTU5M ARRAY (0,1,0)/(10000,3,10000) ASSIGN P(1)-> ((*,*),SHOT1) AEND PEND LAYER 1 P(1) 'fredrikpnov1507prox.v30' SPPRM 4.0,,,,1.0,1 STDCUR 0.22 ;nA SHOT1: MODULAT (( 0, -20) , ( 1, 0) , ( 2, 20)) END
Dose variation defined by an array in jdf file
Another even faster way to define dose modulation (if you have less than 10-15 different doses) is to repeat the pattern (v30-file) in an array with different doses in the jdf file, just type the dose modulation you wish to apply to every chip in the array:
JOB/W 'TEST',4 ; 4 inch wafer PATH DTU5M ARRAY (-10000,2,10000)/(10000,1,10000) ASSIGN P(1)-> ((1,1),SHOT1) ASSIGN P(1)-> ((1,2),SHOT2) ASSIGN P(1)-> ((2,1),SHOT3) ASSIGN P(1)-> ((2,2),SHOT4) AEND PEND LAYER 1 P(1) 'tjulahej.v30' SPPRM 4.0,,,,1.0,1 STDCUR 0.22 ;nA SHOT1: MODULAT (( 0, -20)) SHOT2: MODULAT (( 0, -10)) SHOT3: MODULAT (( 0, 0)) SHOT4: MODULAT (( 0, 10)) END
Proximity Error Correction (PEC)
BEAMER has a built-in proximity error correction simulation program; import the GDS-file as usual and connect the 'PEC' module to its output. In the 'Proximity Error Correction' window you can either type Beta and Eta manually or import a point spread function (PSF).
Add a PEC-module on top of the import module. A Proximity Error Correction parameter window opens. Here, you can either type in values of alpha, beta and eta or load a point spread function. | After converting to v30-format, the dose variation is illustrated in your pattern | BEAMER has at this point created a jdi-file in the same folder as the v30-file. This jdi-file is the shot modulation that should be inserted in the jdf-file. The jdi-file opens with SuperEdi. |
Bulk and sleeve
If you have structures with both bulky and fine parts that you wish to pattern either with different doses or different currents, you can use BEAMER to divide the structures in different layers or different v30-files.
An illustration of how the 'bulk & sleeves' technique this is done by using the Bias and Minus modules; the Bias function subtracts in this case 1 micron from the pattern, defining layer 0 (red layer). Layer 1 (green) is the original pattern minus layer 0. | In the flow to the left, the two layers are exported separately in two different v30-files. In the flow to the right, the two layers are combined in one single v30-file |