Browse Prior Art Database

Reduced product charging in a sputter etch process by soft switch off

IP.com Disclosure Number: IPCOM000015280D
Original Publication Date: 2001-Dec-19
Included in the Prior Art Database: 2003-Jun-20
Document File: 4 page(s) / 109K

Publishing Venue

IBM

Abstract

Modern thin film technology products like chips or GMR read/write heads are highly sensitive to charging defects. This results in ESD damages, electric shorts and low efficiency. There are many possibilities of charging during process. Especially, all plasma processes have to be regarded as critical steps. The plasma contains a lot of charge and if electrons or ions are transferred to the product surface in an uncontrolled way strong ESD damages are expected. Here in disclosed are new findings about the charging mechanism during a plasma process. The results can be used to modify sputter etch processes in that direction, that the charging risk is reduced. Switching off the RF-generator can be the most critical step of a sputter etch process in terms of charging effects. The experimental results show a clear dependence of wafer charging and RF power before switch off (see below). 0 200 400 600 800 1000 1200

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 4

Reduced product charging in a sputter etch process by soft switch off

      Modern thin film technology products like chips or GMR read/write heads are highly sensitive to charging defects. This results in ESD damages, electric shorts and low efficiency. There are many possibilities of charging during process. Especially, all plasma processes have to be regarded as critical steps. The plasma contains a lot of charge and if electrons or ions are transferred to the product surface in an uncontrolled way strong ESD damages are expected.

    Here in disclosed are new findings about the charging mechanism during a plasma process. The results can be used to modify sputter etch processes in that direction, that the charging risk is reduced.

    Switching off the RF-generator can be the most critical step of a sputter etch process in terms of charging effects. The experimental results show a clear dependence of wafer charging and RF power before switch off (see below).

0 200 400 600 800 1000 1200

800

700

500

400

200

100

Avg.

Peak Volt age betw

V]600

t he pal let [

4

DC Volt age of

300 UDC


. Surf ace/ subst rat e [

Peak Voltage


V]

    From this knowledge a new etch process can be derived. Instead of switching off the generator, a smooth ramp down of the generator is recommended. This ramp down can be fast (1 sec or faster) so it will not contribute significantly to the product cycle time. The advantage will be a more save process for our products and may result into significant yield improvement.

Experimental setup

    Sputter etch processes are investigated on a sputtering system which is also used for dry etching. This is a batch tool and runs pallets with multiple wafer positions.

The charge deposited on the wafer surface is determined by means of charging

1

2

etch pow er before sw itch off [W att]

[This page contains 1 picture or other non-text object]

Page 2 of 4

monitor wafers (called Charm2) [1]. These monitor wafers allow the measurement of the maximum positive/negative peak voltages between the wafer surface and the substrate. The Charm2 wafer was placed in pocket 1 of the pallet. The notch (flat side) of the wafers are directed towards the outside of the pallet. All other pockets have been filled with dummy wafers.

    The Charm2 wafers are populated with polarity sensitive sensors of wafer surface/substrate potentials, and sensors of UV dose. The sensors (EPROM cells) have to be calibrated and programmed by WCM, Inc. After processing the Charm2 wafers are sent to WCM, Inc for analysis.

Results and discussion

    Four different etch processes are carried out and are compared. We tested processes where the total amount of removed material is about the same. This is realized by keeping constant the product of etch time and etch power. The four processes differ in RF power after ignition, etch time and RF power duri...