Module classification approach for moving semiconductor...

Details Module classification approach for moving semiconductor... Module classification approach for moving semiconductor...

2002-12-18

2003-12-16

Tran, Khoi H. (Department: 3651)

Data processing: generic control systems or specific application

Specific application, apparatus or process

Article handling

Reexamination Certificate

active

06665584

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention generally relates to semiconductor wafer processing systems, and more particularly to methods and associated apparatus for transporting and processing semiconductor wafers.
2. Description of the Background Art
Semiconductor devices are fabricated using specialized wafer processing systems, which typically have several modules for performing various operations on a semiconductor wafer.
FIG. 1A
shows a schematic diagram of an exemplary wafer processing system
100
in the prior art. System
100
has several modules including modules
101
-
107
. System
100
further includes a computer
121
and a data acquisition and control system
122
for controlling various control elements
123
(e.g., valves, relays, robots, gates, sensors, heaters, motors, gas channels etc.) utilized in the modules of system
100
. A robot
120
in a transfer module
107
is employed to move wafers from one module to another. The movement and processing of wafers are performed in accordance with a list of steps, commonly referred to as a process recipe, which run on computer
121
.
The operation of system
100
is now described using process recipe
108
shown in
FIG. 1B
as an example. A wafer cassette containing several wafers is loaded in a cassette station module
101
. Robot
120
picks up a wafer from the wafer cassette and moves the wafer into aligner module
103
(recipe
108
, step
109
). In aligner
103
, the physical orientation of the wafer is adjusted prior to the wafer's subsequent movement to other modules. Thereafter, the wafer is transferred to a bake station module
104
(recipe
108
, step
110
), where the wafer is pre-heated prior to being placed in a CVD process module
105
. In CVD process module
105
, a film of processing material is deposited on the wafer (recipe
108
, step
111
). System
100
can also accommodate other types of process modules including physical vapor deposition, etching, evaporation, and electro-deposition modules to name a few. Because newly processed wafers can reach temperatures that are high enough to melt a wafer cassette, the wafer coming out of CVD process module
105
is first cooled in a cooling station module
102
(recipe
108
, step
112
), before it is returned to its wafer cassette. The just described steps are repeated for all wafers in cassette station
101
.
Recipe
201
shown in
FIG. 2
is similar to recipe
108
except for the use of a parallel step in step
204
. A parallel step identifies two or more modules that can be alternatively used. In step
204
, the wafer can be processed in either CVD process module
105
or CVD process module
106
whichever is available. As used in this disclosure, the term “module” includes a module identified in a regular step and any one of the modules identified in a parallel step.
Each step in a recipe invokes an associated control program for directing the operation of the listed module. Using recipe
108
as an example, a control program for directing an aligner to adjust the orientation of the wafer is invoked in step
109
. As another example, a control program for directing a process module to perform deposition steps on the wafer is invoked in step
111
. In wafer processing system
100
shown in
FIG. 1A
, such control programs run on computer
121
, and direct control elements
123
via data acquisition and control system
122
. To meet specific process requirements, each control program accepts parameters, such as temperature for the heating elements of bake station
104
or flow rates for the gas channels of CVD process module
105
. It is to be noted that control programs, in general, are well known.
In some situations, the processing of wafers in system
100
has to be abruptly terminated. For example, if the computer controlling system
100
encounters an irrecoverable error while running a recipe, all wafers currently in system
100
may have to be recalled back to their cassettes regardless of whether the wafers have been processed in a CVD process module. This allows a maintenance person to troubleshoot system
100
without risk of destroying the wafers. The removal of a wafer from a wafer processing system is also known as a wafer purge.
A wafer reload is the reverse of a wafer purge. During a reload, purged wafers are placed back to their original location prior to the purge to continue their processing. Wafer reload and wafer purge are examples of non-recipe tasks. Non-recipe tasks are run to execute maintenance functions, user requests, and other tasks that do not involve wafer processing in a process module.
Non-recipe tasks have been performed by following a fixed sequence of steps stored as static files. To purge a wafer from bake station
104
, for example, purge sequence
113
shown in
FIG. 1C
is invoked from a static file. In accordance with purge sequence
113
, a wafer to be purged from bake station
104
is first cooled in cooling station
102
before being placed in a cassette located in cassette station
101
. Purge sequences for other modules of system
100
are also stored as static files. Similarly, static files for wafer reload are available for each module.
The amount of static files that have to be maintained becomes unwieldy as the number of modules supported by system
100
is increased. For example, a patch to fix a common defect will have to be applied to each individual static file containing the purge sequence. Forgetting to apply the patch to even a single static file, which is likely to happen if there are many, can result in the loss of expensive wafers. Further, maintenance personnel will have to familiarize themselves with a large number of static files.
Another problem with static files is that they are inherently inflexible. It is difficult to optimize the sequence contained in static files because the static files are created in advance and are designed to accommodate a variety of situations.
SUMMARY
The present invention relates to a method and associated apparatus for directing the movement of wafers in a wafer processing system.
In one embodiment, each module of the wafer processing system is given a classification. Upon receipt of a request to move the wafer, a sequence enumerating the modules to be visited by the wafer before reaching its destination is created. The modules are added to the sequence based on their classification. The wafer is then worked on in each of the modules enumerated in the sequence.
By creating the sequence when needed, the present invention minimizes the number of static files that have to be maintained and stored in the wafer processing system. Further, creating the sequence at the time it is needed allows the sequence to take advantage of the history of the wafer and thereby eliminate unnecessary steps.


REFERENCES:
patent: 5024570 (1991-06-01), Kiriseko et al.
patent: 5751580 (1998-05-01), Chi
patent: 6122566 (2000-09-01), Nguyen
patent: 6201998 (2001-03-01), Lin et al.
patent: 6201999 (2001-03-01), Jevtic
patent: 6397111 (2002-05-01), Niwa
patent: 6560507 (2003-05-01), Malitsky et al.

Affiliated with

Also associated with

Rating

Module classification approach for moving semiconductor... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Module classification approach for moving semiconductor..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Module classification approach for moving semiconductor... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3122161