Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Ion Implant Dose Monitor With Infrared Transmission Measurement

IP.com Disclosure Number: IPCOM000120884D
Original Publication Date: 1991-Jun-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 5 page(s) / 196K

Publishing Venue

IBM

Related People

Draina, JM: AUTHOR [+2]

Abstract

An ion implant dose monitor using infrared transmission measurement is presented. This monitor is nondestructive and accurate. It is based on absorption of infrared light by free carriers (electrons or holes) inside implanted and annealed wafer.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 45% of the total text.

Ion Implant Dose Monitor With Infrared Transmission Measurement

      An ion implant dose monitor using infrared transmission
measurement is presented.  This monitor is nondestructive and
accurate.  It is based on absorption of infrared light by free
carriers (electrons or holes) inside implanted and annealed wafer.

      Ion implant dose can be determined by measuring sheet
resistance on a furnace annealed test wafer.  This method has dose
resolution on 1% over a wide dose range from 1E12 to 1E16 ions/square
cm.  The drawback of this method is the lengthy time needed for
annealing and measurement.

      An ion implant dose monitor (Therma Wave) has generated
attention.  It measures a visible laser reflectance from implant
surface which is periodically illuminated by a second visible laser.
The reflectance changes with implant damage which depends on implant
dose, implant energy, and implant ion.  The advantage of this monitor
is its direct measurement on an unannealed test wafer.  The
disadvantage is its low signal sensitivity in some dose range, and
poor measurement repeatability.

      A new implant dose monitor which measures infrared transmission
through an annealed test wafer is disclosed. It is repeatable,
nondestructive to wafer, insensitive to implant ion and implant
energy, and has dose resolution within 10% over a dose range from
1E12 to 1E16 ions/sq. cm.

      The monitor set-up is shown in Fig. 1.  A C02 laser is
vertically polarized and has a power of 3.8 watts with only 0.5% long
term fluctuation.  The Gaussian beam size is 2.6 mm in diameter at
the exit of the laser.  Two CaF2 crystals attenuate the power to 45
mW.  A ZnSe beam splitter splits 20% of power into a pyroelectric
detector A.  The rest passes through a wafer and enters a
pyroelectric detector B. The power ratio of B to A is recorded in a
ratiometer.  In order to eliminate reflection from wafer surfaces,
which could cause interference effect, the wafer with two polished
sides is set at Brewster angle (75 degrees).  A vacuum chuck and
translational stage supports the wafer for measurement. The
inaccuracy of the transmission reading is 0.1%.

      Seven test wafers were used.  The p-type wafer resistivity
varied from 11 to 25 ohm-cm.  Six of them were boron implanted at
30KeV with doses at 9E13, 1E14, 9E14, 1E15, 9E15, and 1E16 ions/sq.
cm, respec tively.  All seven wafers were then furnace annealed.  The
transmission results are shown in Fig. 2 and in Table 1.  It
decreases from 90% to 14% with increased dose.  The dose resolution
was 5% at 1E14, and 3% at 1E15 ions/sq. cm.

      At 1E16 dose, the sensitivity was very small, and the
resolution was very large and meaningless.  It is either because
boron ions are lost from highly dosed wafer surface during annealing
process, and oxide capping before annealing may prevent this loss; or
because implant dose reaches solubility of wafer, and no more free
carriers coul...