Introduction
In this page you will find information regarding preparation of design files for patterning samples using the Heidelberg MLA150 Maskless Aligner.
The objective of this document is not teach anyone on how to produces those files, but to tell users what to avoid in order to have a smooth operation of the tool while converting these files into LIC files for patterning.
For more information and questions, please contact the tool trainer (Gustavo) or the fabrication group manager (Aaron).
General rules
These rules apply to all file formats supported by the HIMT MLA150 Menu software.
Naming:
- Do not use special characters (punctuation marks, umlauts etc.) or spaces in the design name.
- Do not use ‘main’ or any combination of “main” as name for a layer, cell or any structure. It does not make any difference if “main” is written in lowercase or uppercase letters or a combination of both.
Text: Is only supported for DXF format and ODB.
General size limits: Size limits can be given by hardware specific restrictions such as the available memory space of the RAM or CPU.
Areas: Use only polygons/closed polylines and circles to define exposure areas. Lines cannot be used to define an area; only the outlines will be exposed.
Combination of structures:
Some of the design formats that can be converted to LIC are layer based (DXF, GDSII). Also, structures can overlap within a design. In both cases, the treatment of overlapping structures must be defined, sometimes by the user, sometimes by the software. The following logical commands are available:
OR: Results in a true merging of the structures. Any structures that lie completely within bigger structures are covered by these. This will avoid double exposure of overlapping structures.
CUT: Subtracts a layer from the previous layer
XOR: Leads to a cutting only in the overlap region, while elsewhere both structures remain intact.
So OR and XOR are applied for overlapping structures within a layer / design. OR and CUT are used for merging layers.
In designs with multiple layers, different operations can be chosen for each layer combination. Operations are executed layer-by-layer from the top of the list to the bottom. There is no way of enclosing sets of operations in brackets, not even by merging of merged files. If several layers are merged, the result of each CUT operation depends on the order of operations.
Merged files still consist of the original files plus the merging information. If two merged files, or files with merged layers, are merged, the lists of operations are appended to each other in the given order, and the result is executed as one operations list. The operations are NOT executed file-wise before the file merging is done!
Lines:
- Lines have to be defined with a fixed positive line width.
- Lines with zero width are ignored, as well as any line width changes within a line (tapered lines).
Polygons/polylines: Vector coordinates that define closed lines
- Areas surrounded by a polyline are filled. Some data formats allow automatic closing of polylines. Always use this option if available, otherwise polylines are closed automatically.
- Do not cross polylines: crossed polylines create ambiguities that can lead to data errors.
- Do not create double vertices: double vertices are successive points with identical coordinates. These lead to data errors.
Arrays:
- Arrays can have positive or negative vectors.
- Non-orthogonal vectors are NOT supported.
Alignment marks
Design
Ideally, alignment marks are expected to extend through the whole field of view (FOV) of the high resolution camera (190 µm × 140 µm). However, for backside alignment (BSA) the field of view is the same as the low resolution camera (640 µm × 480 µm).
The software is optimized to look for the arms of a cross, which means that if you use any other type of marks, the alignment will have to be performed manually. Below is an example of an alignment mark that would work well for both top and back side alignment.
The narrower sections fit entirely in the FOV of the high resolution camera while still being detectable by the software. And the wider sections will cover the whole low resolution camera FOV and also be easily found by the software.
Smaller marks can also be used, however their reliability will depend on the capability of the software to find the edges of the arms of the cross.
Top side
For top side alignment the marks can be placed anywhere in the design — bear in mind that you have to know their coordinates in the layout precisely. The only restriction is that the overview camera cannot be used to search for them on the top region of 4" (or larger) wafers or masks, due to a mechanical limitation of the tool. But both low and high resolution cameras can be used normally.
Back side
For BSA the same type of marks can be used. However they must be placed such that they are aligned with the windows in the chuck of the sample holder. Use the diagram below to prepare you design (units in mm).
Also, note that when using BSA the tool does NOT mirror the coordinate system. This means that you must either prepare your design mirrored or mirror it when converting the file (prefered).
And note as well that the x-coordinate of the crosses will be mirrored when entered in the alignment setup:
GDSII
GDSII is a very common design format for layouts of integrated circuit boards. Numerous programs use it as standard output format or offer it as export option. Rules for this format apart from the general rules above are few:
- No Inclusions: Do not include other GDSII files or text libraries; they will be ignored.
- No nodes: Node statements in GDSII are ignored.
- No Text: Text statements are ignored. Make sure that any text (e.g., labels) are made out of polygons.
DXF
DXF is a common graphics design format, used by the well-known AutoCAD software. There is also other design software available that can export into DXF format, and software that can translate between different design formats (e.g., LinkCAD).
Regardless of the design software you use, make sure the resulting file format is 100% AutoCAD R12 compatible.
Create designs according to the software manufacturer’s instructions, taking into account the general plus the following additional rules.
Units: Use the metric system when designing (mm as basic unit recommended).
Entities:
- Polyline/LW Polyline
- Circle
- Solid
- Text: Only one text font is provided (Courier-like typewriter font). This will replace any other font used in the design creation. The only supported text attributes are: r otated, mirrored, scaled
Layers: Avoid placing structures in layer 0.
Not supported entities: Line, Arc
Blocks:
- When inserting blocks, the same scaling has to be used for x and y.
- External blocks are not supported
DXF translator tool: Although the DXF translator supports newer DXF releases, the manufacturer recommends using Release 12.
CIF
The Caltech Intermediate Format (CIF) is a design format for direct creation of designs in a text editor. For complex structures, it is recommendable to use a professional CAD software that can create e.g., AutoCAD R12 compatible DXF design files. If only small test structures are set up, they can be easily defined and varied in a CIF design file.
One rule has to be taken into account in addition to the general rules when setting up a CIF design:
- Maximum definition or reference depth is 50, meaning that not more than 50 subroutine calls may be nested into each other.
Gerber
In Gerber format, structures are described via apertures and a path across the design field. The opening and closing of the aperture along the path, and the selection of aperture types, create the final structure. Keep in mind that the effective line width also depends on system parameters and processing.
- Supported: RS 274-X
- Not supported: RS-274-D
- Do not call other Gerber files inside Gerber.
Bitmap
3D exposures require a file format where a depth for each pixel within the design area can be defined in the form of a greyscale value. This is given in the bitmap format, which offers 256 different intensity levels that can be addressed by the greyscale values 0–255 in an 8-bit greyscale format. This can also be used for 2D exposure of bitmap files, for which white (255) corresponds to exposed areas and black (0) to unexposed.
The bitmap format also has some specific rules that have to be taken into account in addition to the general ones when setting up a design.
- Greyscale Format: For greyscale exposures, use 1, 4, 8, and 24-bit greyscale format. For designs that require only 16 or less grey values, also 4-bit greyscale can be used. This has the advantage that a bigger area can be covered, as the file size is smaller (see next bullet point).
- File size: The official Microsoft possible image file size in BMP is 4 GB, but the software can ignore these limitations. In the above recommended setting for bitmap designs (8-bit greyscale), this corresponds to 4 billion pixels, or a size of e.g., 63000 × 63000 pixels. The physical size of such a design depends on the configuration of the system.