GROUP. SUB-GROUP ~Cleaning, surface cleaning, ultraviolet light, UV, ozone,. 09 ... Dr. John R. Vig ...... John Crane Lapping Vehicle 3M, Crane Packing Co.,.
1VCLEANING OF SURFACES ,U
VIG
ELECTRONICS TECHNOLOGY AND DEVICES LABORATORY
MAY
DTIC
1986
ELECTE
DISTRIBUTION STATEMENT Approved for public release; distribution is unlimited.
US ARMY LABORATORY COMMAND FORT MONMOUTH, NEW JERSEY 07703-5302
D
NOTICES
Disclaimers
The citation of trade names and names of manufacturers in this report is not to be construed as official Government indorsement or approval of commercial products or services referenced herein.
Destruction
Notice
Destroy this report when it is no longer needed. return it to the originator.
Do not
h/0%#9169c gr
UNCLASSIFIED WOCRRTY allAS~FcATIO
Or. rHIS PAGE
REPORT DOCUMENTATION PAGE lb RESTRICTIVE MARKINGS
I&. REPORT SECURITY CLASSIFICATION
F Unclassified
______________________
l.a SECURITY CLAs5iFIC.ATlON AUTHORITY
I DiSrRIBUTIONIAVAILABLiTY OF REPORT
'b DkCLASSIPHLArIONIDOONGRAOING SCHEOULE
Approved for public release; distribution is unlimited,
4 PERFORMING ORGANIZATION REPORT NUMBER(S)
5 MONITORING ORGANIZATION REPORT NUMBER(S)
SLCET-TR-86-6
I6b
6a NAME OF PERFORMING ORGANIZATION
OFFICE SYMBOL
7a3NAME OF MONITORING ORGANIZATION
US Army Laboratory Cmd (LABC0M~ (if applicable) Electronics Tech & Devices Lab1 SLCET-EQ
(sm) 7b ADDRESS (City, State. and ZIP Code)
State, and ZIP Code) 6c ADDRESS (City,
Electronics Technology and Devices Laboratory
ATTN.- SLCET-EQ 07703-5302
Fort Monmouth, NJ
18b
8.NAME OF FUNDING ISPONSORING
ORGANIZATION
_____________________
_______
OFFICE SYMBOL
(i applicable)
8c ADDRESS(City, State, and ZIP Code)
9 PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER
10 SOURCE OF FUNDING NUMBERS TASK ~PROJECT PROGRAM NO ELEMENT NO
INO
1 11162705__
_______________________________ 11 TITLE (include Security Clasubfcation)
AH94
I
jWORK UNIT ACCESSION NO
09
IDAOD20731
UV/OZONE CLEANING OF SURFACES (U) 'Z "SONAL AUTHOR(S)
John R. Vig
I13b
13a TYPE OF REPORT
Tchnical Progrs
114DATE OF REPORT (Year, Month, Day)
TIME COVERED
S15PA GE COUNT
N/A To _ 198 1ayi 4 16 SUPPLEMENTARY NOTATION This report is a reprint of a chapter prepared for Treatise on Clean Surf-ace. Technology, K.L. Mittal, editor; to be published by Plenum Publishig Corp. GROUP
17FIELD
~I ii
18 SUBJECT TERMS (Continue on reverse
CODES
17COSATI
09
FROM
SUB-GROUP
01 11
and identify by block number)
)mer
ilABSTRACT (Contrinue on reverse ifnecessary 3nd identify by block nu
I
if necessary
~Cleaning, surface cleaning, ultraviolet light, UV, ozone, contamination control, surface contamination, organic conjtamination, '.cQtaminI1nation.
"itThe UV/ozone method, which is reviewed in this repo-Tt, is anefective method of removing a vat jety of contaminants from surfaces. It is a simlple-to-use dry process which is inexpensive to set up and operate. It can rapidly produce clean surfaces, in air 3r in a
vacuum systemn, at ambient temperatures.
Placing properly precleaned surfaces wilhin a
few millimeters of an ozonIe-producing UV Source can produce clean surfaces in less than one mlinute. The technique can produce near-atomically clean surfaces, as evidenced by Auqer electron spectroscopy (AES), electron spectroscopy for chemlical analysis, (ESCA), alnd iIon scaitterinq spectroscopy/sevcondary ion mass spectroscoi'y (IM/SlMS) studies.TolpiC, (lisctssed include tile variables of the lpro(vs%, the typIe% of surfaces wh i h have
been cleaned successfully, Lhe contamlinants that Lan be remloved, the construction of a UV/ozone cleaning facility, the mlechanism of the process, UV/ozone cleaning in vaccum systems, rate-enhancement techniques, safety considerations, effects of UV/ozone other 20 D S'R,JU T'O%, AVA:LABIL.TY OF ABSTRACT
RI'CASFE).LM1E
0 SAME AS RPT
'2?a NANIE 0; ilVONSIBLE INDivIDUAL
Dr. John R. Vig DO FORM 1473. 84 MAR
21 ABSTRACT SECURITY CLASSIFICATION Q OTIC USERS
Unclassified
I22b 7ELEPmONE (include Area Code) I
(201) 544-4275
83 APii ed~tion may be used until exhawsted All other editiofls are obsolete
I
22c OFFICE SYMBOL
SLCET-EQ
SECURITY CLASSIFICATION OF TWIS PAGE
UNCLASS IF IED
CONTENTS Page
S3.
1.
SUMMARY ......................................
2.
INTRODUCTION ............................................
....
1
THE VARIABLES OF UV/OZONE CLEANING
.....................
3
3.1 The W"-avelengths Emitted by the UV Sources......
3
3.2 Distance Between The Sample and UV Source ...........
10
3.3 The Contaminants ....................................
11
3.4 The Precleaning .....................................
14
3.5 The Substrate .......................................
15
3.6 Rate Enhancement Techniques .........................
17
4.
THE MECHANISM OF UV/OZONE CLEANING ......................
18
5.
UV/OZONE CLEANING IN VACUUM SYSTEMS .....................
21
6.
S'kFETY CONSIDERATIONS .........................
7.
UV/OZONE CLEANING FACILITY CONSTRUCTION .................
8.
APPLICATIONS ........................................
9.
EFFECTS OTHER THAN CLEANING ............................
30
9.1 Oxidation ...........................................
30
..........
22
24 ....
9.2 UV-enhanced Outgassing ...................... 9.3 Other Surface/Interface Effects .......................
.......
30 32
10. SUMMARY AND CONCLUSIONS ................................. 11. REFERENCES ................................................
32 34
-Accesioii For 'NTIS CRA& DTIC TAB Unannounced Justification•
i[D [3
Dfstrib-utloo I
Avm3iabthty Codes Dist
iii
26
31
9.4 Etching .............................................
I
0
Avaai a.,djor Special
FIGURES
Page 1 2 3 4
5 6 7 8
Apparatus for UV/ozone cleaning experiments. Absorption spectrum of oxygen. Absorption spectrum of ozone. Auger spectra of evaporated aluminum film on silicon substrate: (a) before UV/ozone cleaning; (b) after UV/ozone cleaning. Schematic 6rawing of a UV/ozone cleaner that uses a silent-discharge ozone generator. Photoresist stripping rate vs. substrate temperature for three types of photoresists. Simplified schematic representation of OV/ozone cleaning process. Effect of UV/ozone cleaning on gold-to-gold thermocompression bonding.
3 6 6
9 18 19 20 29
TABLES 1 2
Principal wavelengths of low-pressure Hg discharge lamps. Effects of the principal wavelengths generated by low-pressure Hg discharge lamps.
3
Exposure
4
UV/ozone exposure vs.
types vs.
cleaning
times.
oxide thickness
iv
4 5 10
on aluminum.
31
1. SUMMARY
The
been cleaned
the mechanism rate-
systems,
vacuum
effects of UV/ozone
and applications.
other than cleaning, 2.
in
have
can be removed,
that
facility,
cleaning
cleaning
include
which
surfaces
of
safety considerations,
enhancement techniques,
clean
electron
Auger
by
the contaminants
UV/ozone
few
a
produce
can
Topics discussed
types
the
of an UV/ozone
process,
the
process,
successfully,
the construction of
the
of
variables
ambient
within
surfaces
evidenced
ISS/SIMS studies.
and
ESCA,
spectroscopy,
process
The technique can produce near-
as
surfaces,
clean
atomically
of
at
system,
source
UV
variety
can rapidly produce
It
vacuum
ozone-producing
an
in
dry
simple-to-use
precleaned
less than one minute.
in
surfaces
a
in
properly
Placing of
millimeters
or
air
in
surfaces,
temperatures.
the
a
reviewed a
removing
of
inexpensive to set up and operate.
which is clean
is
It
surfaces.
from
contaminants
method
effective
an
is
report,*
this
is
which
method,
surface-cleaning
(UV)/ozone
INTRODUCTION (UV)
The capability of ultraviolet has been known
molecules
for a
long
1972,
The
polymer
evacuated
*This invitation cleaning Technology.
Kunz(I)
depolymerize
to
capability
and
Bolon
were
films
and
was
from for Dr.
K.
enclosed
the Mittal
Mittal
forthcoming is
in
with
originally L.
of
is
only during
a
for
UV
light
photoresist quartz
a
tube
The
oxygen.
in
prepared
the Treatise Editor.
on
polymers. that
samples
on Clear
the
had
was were
to
response
chapter
Treatise
be published by Plenum Publishing Corp.
SI1
it
that
reported
variety
backfilled
then
report
a
but
time,
of surfaces has been explored.
the past decade that UV cleaning In
light to decompose organic
an
UV/ozone Surface
The Treatise
is
to
irradiated with UV light from a medium pressure mercury lamp that ozone.
generated films
were successfully depolymerized
major
products
of
Thus,
1974,
In
15
from
hydrocarbons
adsorbed hours
of
system at
to
exposure
10-4 torr
the
)
of
02
cleaned
Sowell et
radiation."
of
UV
under
noted
air after
was obtained In
dir.
of
in
surfaces,
in
vacuum
a
the partial
and H2 0
CO 2
increased. air,
In
by adsorbed hydrocarbons of
exposure
the
to
UV
that storing clean surfaces the
maintained
radiation
cleaning
vacuum chamber walls.
hours
also
al.
hindered.
wavelengths
During cleaning,
that
"sevferal
by
placed
surfaces were produced
clean gold
gold surfaces which had been contaminated be
was
UV
gold
and
The UV also desorbed gases from the could
was
and
oxygen
radiation
while
decreased,
and were found
filter
Pyrex
described
after about two hours of UV exposure. pressure
substrates
the
(AES)
glass surface
UV
of oxygen,
a
that
glass
A clean
a vacuum system.
in
be
ini the depolymeriz~tion. 2
al.(
et
and
to
or when a nitrogen atmosphere
recognized
Sowell
water
found
depolymerization
nm played a role
than 300
shorter
and
Kunz
and
Bolon
the
oxygen,
of
instead
used
The
Only inorganic residues such
When
the UV light and the films,
between was
found.
were
chlorine
and
than one hour.
depolymerization,
to
residues.
to be free of carbondceous tin
less
by Auger electron spectroscopy
were examined as
in
were
depolymerization
Subsequent
dioxide.
carbon
polymer
several-thousand-angstroms-thick
The
surface
cleanliness
indefinitely. Duting series
of
variables
the
under of
1976,
steadily.
clean
proper of
at
aimed
by UV
producing use
Vig
1974-1976,
producing
of cleaning
capability Since
period
experiments for
conditions that,
the
al.(
by
UV
light were defined,
clean
UV/ozone
surfaces
UV/ozone
3
-
5
)
determining
surfaces
conditions,
the
et
in
less
cleaning
described the
optimum
irradiation. ind
it
was
cleaning
has
than one method
has
UV/ozone cleaners are now available commercially.
2
a The
shown the
minute. grown
3.
OF UV/OZONE CLEANING
THE VARIABLES
The Wavelengths Emitted by the UV Sources
3.1
and LeBus(
5
mercury
discharge
short-wavelength tube.
the
from
UV
1.6
contained
experiments.
The
intensities
equal
nearly
-,nout
boxes
Both
Figure
contained low-pressure 6 aluminum stand with Alzak( )
produced
lignt,
Vig
procedure,
cleaning
both
and an
and
lamps
these
throughout
room)
aluminum,
lamps two
The
reflectors.
of
made
were
1 cm
UV
the
of
) constructed the two UV cleaning boxes shown in
1. Both
of
variables
the
study
To
mW/cm 2
were
boxes
sample a clean
(in
air
room
a
for
completely
by air circulation.
enclosed to reduce recontamination
LOW PRESSURE HI /LOW PRESSURE Hg
,
OZONE GENERATOR
UARTZ TUBE 'ALZAK REFLECTOR, -..-- SAMPLE'---
•ALZAK REFLECTOR "-"AL
101-
-I
AL STAND '
S.___..__j
TRANSFORER
UV BOX 2
UV BOX I
1. Apparatus for UV/ozone cleaning experiments.
Figure
Since only
the
light which
is
the wavelengths
producing photochemical changes, sources discharge
are
tubes
and 253.7 nm. upon
the
important generate
variables. two
can
absorbed The
wavelengths
of
envelopes.
The
emissions
types of envelopes are summarized
in
3
interest:
mercury 184.9
nm
are emitted depends
through
Table 1.
in
emitted by the UV
low-pressure
Whether or not these wa,,elengths
lamp
be effective
the
three
main
of low-pressure Hg
Table 1. Principal wavelengths discharge lamps.
LAMP ENVELOPE WAVELENGTH
QUARTZ
HIGH SILICA GLASS
GLASS
184.9 nm
T
0
0
253.7 nm
T
T
0
>300.0 nm
T
T
T
T = Transparent, 0 = Opaque
absorbed ozone.(
7
)
wavelength
is
by
oxygen,
thus
leading
to
The
253.7
is
not
to
ozone
absorbed
does
it
therefore,
by most
nm
radiation
not
contribute
by
ozone
destruction
of
ozone
wavelengths
are
the
An
molecules
organic
absorption
destroyed.
important
nm
184.9
The
present,
intermediate
destruction
UV
the
in
product,
both
generation
of
and
by
but
ozone.(
7
the both
when
Therefore,
and
formed
being
the formation is
which
is The
)
for
responmible
of
oxygen,
by
generation,
oxygen,
atomic
is
processes,
box.
is
absorbed
continually
is
ozone
the
(8,9)
principally
is
it
because
and a
of
very
strong oxidizing agent. The tube of the "hairpin-bent"
fused
UV lamp(
10
quartz
tubing.
both the 253.7 nm and the
184.9
in
)
box The
1 consisted fused
nm wavelengths.
4
of 91 cm of
quartz The
transmits
lamp emitted
about
0.1
mW/cm 2
radiation
nm
184.9
of
measured
1
at
from
cm
the tube. The absorption that
of
spectrum
of
Figure
3.
in
ozone
oxygen
is
shown
wavelengths generated by low-pressure mercury discharge summarized Table 2.
2
and
principal
the
of
effects
The
Figure
in
lamps are
An Table 2. Effects of the principal wavelengths generated by
low-pressure Hg discharge lamps.
EFFECTS
WAVELENGTH
* Absorbed by 02 and organic molecules e Creates atomic oxygen and ozone
184.9 nm
* Breaks contaminant molecule bonds
* Absorbed by organic molecules and 03;
253.7 nm
not absorbed by 02 * Destroys ozone * Breaks contaminant molecule bonds
The lamp in box 2 had two straight and parallel 46 cm long high-silica glass tubes made of Corning UV Glass No. 9823, whuich nm
transmits
at
253.7
generated
no
measurable was
generator(II) not
emit
discharge
k
UV when
built
light.
but
a
ozone,
box
into Ozone
high-voltage
at
not
ac
was was
5
184.9
nm.
This
ozone
produced applied
type
Siemens
separate 2.
this
Since
generator
by across
lamp ozone did
a
"silent"
a
discharge
10
S
a-
II 0
175
180
185
190
wavelength (nm)
Figure 2. Absorption spectrum of oxygen. CL
60-
140.. -100
o
60
".
23
•o
•
wavelength (nm)
Figure
3. Absorption
spectrum of ozone.
gap
formed
wrapped
by
in
two
concentric
aluminum
foil
glass
tubes,
electrodes.
each
of
which
The ozone-generating
was tubes
were parallel to the UV tubes and were spaced approximately apart.
6 cm
UV box 1 was used to expose samples, simultaneously, to the 253.7 nm and 184.9 nm wavelengths and to the ozone generated by the 184.9 nm wavelength. UV box 2 permitted the options of exposing samples to 253.7 nm plus ozone, 253.7 nm only, or ozone only. Vig tests,
et
and
results
al.
used
angle
measurements,
Auger
electron spectroscopy (AES) cleaning experiments. Most of the
of
conducted
on
could
evaluated
be
contact
polished
quartz
by
the
wafers,
the
"steam
test,"
wettability
to
evaluate the experiments were cleanliness of which a
bighly
sensitive measurements and the
wettability test.( 5 , 1 2 , 1 5 ) Contact angle steam test can detect fractional monolayers of hydrophobic surface contamination. Also
tested
that emitted produced no
was
a
"black-light"
wavelengths noticeable
long-wavelength
above 300 nm only. This cleaning, even after 24
UV
source
UV source hours of
irradiation. In
be
the studies of Vig et al.,
cleaned
consistently
by
it
UV
was found that samples could
irradiation
only
if
gross
contamination was first removed from the surfaces. precleaning procedure consisted Df the following steps:
Their
scrubbing
with
a
while
the
(2)
"immersed in ethyl alcohol, degreasing ultrasonically
in
solvent
(3)
trichlorotrifluoroethane, boiling in fresh ethyl
alcohol,
(1)
swab
ultrasonically, 7
a
sample
then
such
was as
agitating
(4) (5)
rinsing in running ultrapure (18RM cm) water, spinning dry immediately after the running-water rinse.
Subsequent to this precleaning procedure, the steam test and contact angle measurements invariably indicated that the surfaces were contaminated. However, after exposure to UV/ozone in box 1, the same tests always indicated clean surfaces. The cleanliness of such UV/ozone-cleaned surfaces has been verified on ntimerous occasions, in the author's laboratory and elsewhere, by AES and electron spectroscopy for chemical analysis (ESCA).(l, 3 , 4 ,1 3 -1 6 ) Figure 4 shows cleaning.(16)
Auger
Ten minutes contamination on
of an
detectab'lity
level,
spectra
before
and
UV/ozone
UV/nrnne aluminum about
cleaning reduced ;-he surface thin film to below the AES one percent of a monolayer. The
effectiveness of UV/ozone cleanin- ha-scattering
after
spectrozcopy/secondary
.rmedco-f ion
mass
by ion spectroscopy
(ISS/SIMS) . (17)
A number of quartz wafers were precleaned and exposed to the UV light in box 1 until clean surfaces were obtained. Each of the wafers was then thoroughly contaminated with human skin oil, which has been a difficult zontaminant to remove. (The skin oil was applied by rubbing the wafer on the forehead of one of the researchers.) The wafers were ptecleaned again, groups of wafers were exposed tc each of the four earlier?
and
measured,
as
indicated
were
placed
samples
the
-Lime needed
UV/ozone combinations mentioned to
by the steam within
5
mm
attain
a
test. of
clean
surface
was
In each
UV box,
the
the
UV
scurce
(where
the tempetature was abeut 70 0 C). The wafers exposed to 25'.7 nm + 184.9 nrm + ozone in UV box 1 became clean in 20 s. The samples exposed to 253.7 nm + ozone in
*
S
0
MI I
AI
w
I'll
2
A
0t
•
0 A •
20
"Electron
400
600
80
1000
energy (eV,
Figure 4. Auger spectra of evaporated aluminum film on silicon suuatrate: (a) before UV/ozone cleaning; (b) after UV/ozone cleaning.
• •
UV box 2 reached the clean condition in 90 s. Samples exposed to 253.7 nm without ozone were cleaned within one hour, and samples to ozone without Sexposed UV light were cleaned Within ten hours. The results are summarized in Table 3.
Table 3. Exposure types vs.
cleaning times.
Exposure Type
Cleaning Time
"Block light" (>300 nm)
No cleaning
03, no UV
10 hours
253.7 nm, no 03
1 hour
253.7 nm + 03
90 sec
"253.7
Therefore, ozone and
nm + 18-,.9 nm + 03 one may conclude that, ozone without UV light
20 sec while both UV light without can produce a slow cleaning
effect in air, the combination of short-wavelength UV light and of ozone,
such as is obtained from a quartz UV lamp, produces a clean
surface orders of magnitude faster. Although
the
organic molecules,
184.9
nm
radiation
is
also
absurbed
by many
it
was not possible from these experiments to isolate the cleaning effect of the 184.9 nm radiation. The ozone As is discussed below, concentrations had not been measured. within each box the ozone concentrations vary with distance from tne UV source.
3.2
Distance Between the Sample and UV Source
Another variable that can greatly affect the cleaning rate is the distance between the sample and the UV source. Because of the shzipcs of the UV tubes and of the Alzak reflectors above the tubes and below the samples, the lamps in both boxes were essentially plane sources. Therefore, one way conclude that the intensity of UV light reaching a sample would be nearly independent of
distance. However, this is not so when ozone is present, because ozone has a broad absorption band( 7 ,18,1 9 ) centered at about 260 nm, as is shown in Figure 3. At 253.7 nm, the absorption coefficient is 130 cm-I atm- 1 . The intensity I of the 253.7 rm radiation reaching a sample therefote decreases as I = Ioe-130pd
where p is
the average ozone pressure between the sample and UV source in atmospheres at 0oC, and d is the distancez to sample in centimeters. When a quartz UV tube is used, both ozone concentration and the UV radiation intensity decrease distance from the UV source.
the the the with
Two sets of identically precleaned samples were placed in UV box 2. One set was placed within 5 mm of the UV tube, the other was placed at the bottom of the box about 8 cm from the tube. With the ozone generator off, there was less than a 30 percent Sdifference in the tLme it took for the two sets of samples to attain
min.
a
minimal
(--14o)
contact
angle,
about
60
min
versus
75
When
the experiment was repeated with the ozone generator on, the samples near the tube became clean nearly ten times faster (about 90 s versus 33 min). Similarly, in UV box 1, samples placed within 5 mm of the tube were cleaned in 20 s versus 20-30 min for samp.ls placed near the bottom of the box at a distance of 13 cm. Therefore, to maximize the cleaning rate, the samples should be placed as close as practicable to the UV source.
3.3
The Contaminants Vig et al.
procedure
on
tested the effectiveness of the UV/ozone cleaning a variety of contaminants. Among the contaminants
were:
11
(1) human
skin
oils
(wiped
from
the
forehead
of
one
of the researchers), (2)
contamination
adsorbed
during
prolonged
to air, cutting oil,(20) beeswax and rosin mixture, lapping vehicle,(21. mechanical vacuum pump oil,( 2 2 ) DC 704 silicone diffusion pump oil,(
(3) (4) (5) (6) (7)
DC 705 silicone diffusion pump oil,( silicone vacuum grease,(23) acid (solder) flux,( 2 4 )
(8) (9) (10) (11) (12)
23
)
23
)
exposure
rosin flux from a rosin core lead-tin solder cleaning solvent residues, including acetone, ethanol,
methanol,
isopropyl alcohol,
trichloroethane,
and trichlorotrifluoroethane. The contaminants were applied with swabs to clean, polished quartz wafers. The amount ot contamination was not measured. However, each time a swab was used in the application, it was obvious to the unaided eye that the sampLes had been thoroughly contaminated. After contamination, the wafers were precleaned, then exposed to UV/ozone by placement within a few millimeters of the tube in UV box 1. After a 60 s exposure, the steam test and AES indicated that all traces of the contaminants had been rcmoed. Using silicon
peaks
containing fluids
AES,
was,
no due
differentiation
coulc
be
made
to
and
those
due
to
removal
of
quartz
contaminants. therefore,
The
silicone
also tested on Alzak,
between the
the
silicon-
diffusion
pump
which normally has a
silicon-free oxide surface, ana on gold. Following UV/ozone cleaning, AES examination both of the Alzak and the gold surfaces showed no presence of silicon.
12 *
During the course of their studies, colleagues
working
on
ion
Vig et al.
implantation
for
learned from
integrated
circuits
that the usual wet-cleaning procedures (with hot acids) failed to remove the photoresist from silicon wafers that had been exposed to radiation in an ion-implantation accelerator, presumably because of cross-linking of the photoresist. Ion-implanted silicon wafers, each with approximately a lum coating of exposed Kodak Micro Resist 717,(25) were placed within a few millimeters of the source in UV box 1. After an overnight ('-'10 h) exposure to UV/ozone, all traces of the photoresist were removed from the wafers, as confirmed by AES. Films of carbon, quartz
surfaces
vacuum-deposited
conductive
for study
onto quartz
in
to make
an electron
the
microscope,
were also successfully removel by exposure to UV/ozone. Inorganic contaminants, such as dust and salts, cannot be removed by UV/ozone and should be removed in the precleaning procedure. UV/ozone for
has also
destruction
of
been
highly
used toxic
for
waste-wateL
compounds. (26-29)
treatment
and
Experimental
work in connection with these applications has shown that UV/ozone can convert a wide variety of organic and some inorganic species to relatively harmless, mostly volatile products such as CO2 , CO, H2 0, and N2. Compounds which have been destroyed successfully in water by UV/ozone include: ethanol, acetic acid, glycine, glycerol, palmitic acid; organic nitrogen, phosphorous and sulfur compounds; cyanides;
potassium cyanide; complexed Cd, Cu, Fe, and Ni photographic wastes, medical wastes, secondary
effluents; chlorinated organics pentachlorophenol, dichlorobenzene, malathion, Baygon, Vapam, and shown(
30
)
that
using
the
and pesticides such as dichlorbutane, chloroform, DDT. It has also been
combination
of
UV
and
ozone
is
more
effective than using eithec one alone in destroying microbial contaminants (E. coli and streptococcus faecalis) in water.
13
3.4
The Prccleaning Contaminants
carbon
films
precleaning,
removed
such
can 1'ut,
without
be in
as
thick
removed general,
precleaning.
For
ivhotoresist with gross
coatings
UV/ozone,
and
pure
without
any
contamination
example,
when
a
cannot
clean
be
quartz
wafer was coated thoroughly with human skin oils and placed in OV box 1 (Figure 1) witnout any precleaning, even prolonged exposure to UV/ozone failed to produce a low-contact-angle surface, because human skin oils contain materials such as inorganic salts which cannot 'e removed by photosensitized oxidation. The UV/ozone removed silicones from surfaces which had been precleaned, as described earlier, and also from surfaces which had simply been wiped with a cloth to leave a thin film. However, when the removal of a thick film was attempted, the UV/ozone removed most of the tilm upon prolonged exposure, but it also left a hard, cracked respond
residue on thc suzrface, possibly because many chemicals to radiatitn in various ways, depending upon whether or
not oxygen is piesent. For instance, in the presence of oxygen, many polymers degrade when irradiated; whereas, in the absence of oxygen (as would be the case for the bulk of a thick film) these same polymers crosslink. In the study of the radiation degradation of polymers in air, the "results obtained with thin fils are ... ly different fyow those obtained using (31) thick specimen... -For the UV/ozone cleaning procedure to perform reliably,
the
surfaces must be precleaned: first, to remove contaminants such as dust and salts, which cannot be changed into volatile products by the oxidizing action of UViozone, and, second, to remove thick films the bulk of which could be transformed into a UV-resistant film by the crosslinking action of the UV light chat penetrates the surface.
3.5
The Substrate
The UV/ozone cleaning process has been used with success on a variety of surfaces, including: glass, quartz, mica, sapphire, ceramics, metals, silicon, gallium arsenide, and a conductive pol.yimide cement. Quartz and sapphire are especially easy to clean with UV/ozone since these materials are transparent to short wavelength UV. For example, when a pile of thin quartz plates, approximately two centimeters deep, was cleaned by UV/ozone, both sides of all the plates, even those at the bottom of the pile, were cleaned by the process. Since sapphire is even more transparent, it, too, could probably be cleaned the same way. When flat quartz plates were placed on top of each ocher so that there could have been little or no ozone circulation between the plates, it was possible to clean both sides of the plates by the UV/ozone cleaning method. (Reference 32 shows that photocatalytic oxidation of hydrocarbons, without the presenc' of gaseous oxygen, can occur on some oxide surfaces.)
When white alumina ceramic substrates were cleaned by UV/ozone, the surfaces were cleaned properly. However, the sides facing the UV became yellow, probably due to the production of UV induced color centers. After a few minutes at high temperatures (>1600C),
* *.
the white color returned.
Metal surfaces could be cleaned by UV/ozone without any problems, so loig as the UV exposure was limited to the time required to ,.odu-e a clean surface. (This time should be approximatcŽ' one rinute or less for surfaces which have been properly prec?.eained.) However, prolonged exposure of oxide-forming metals
to UV
for examole, with
sheets
light
can
blackened of
Kovzr,
produce
rapid
corrosion.
within one hour
in
stainless
(type
steel
15
Silver
UV box 302) ,
samples,
1. Experiments gold,
silver,
and
copper
showed
the stainless and
copper
darker
on
that, and
the gold
oxidized
on
beth
sides,
but
facing
away
from
sides
electroless
gold-plated
for several
days,
the
Apparently,
pinholes
in
covered
the
observed
in
the
in
nickel
a powdery
parts
nickel
ne,•rly
appeared
black
gold plating
gold
and
unchanged; the
to
the
The
the
UV
when
substantially
increased
boxes
to
increase
under
When
UV/ozone
appeared
surface nickel
on
through
eventually
corrosion
was
also
The ratep,
of
a beaker
the
were
source.
gradually
oxidized
completely.
the silver layers
UV
stored
coating the
oxiýie
the
were
diffused
the Kovar,
UV box 2 when no ozone was being generated.
of corrosion placed
UV irradiation,
steel,
the
parts.
upon extended
water
humidity.
Even
was
Kovar
showed signs of corrosion under such conditions. The known in short
corrosion
may
be
explained
as
the science of air-pollution control, UV
wavelength L,
controlled
sulfuric
atmospheres
in
acids.
the
in
air,
in
combine with water vapor
arid
iziLLic
impurities
light,
nitrogen and sulZur, !IilSL
possibly
follows:
as
is
the presence of
such as
of
oxides
to form a corrosive
Therefore,
the
UV box may minimize
the
use
of
corrosion
problem. Since
UV/ozone
dissociates
useful means of cleaning
organic
molecules,
some organic materials,
it
may
just as
be
a
etching
and electropolishing are sometimes useful for cleaning metals. The process has been used successfully to clean quartz resonators which
have
Teflon
(TFE)
a
angles
weight
to
loss
measured
UV/ozone of
clean
in
2.5
UV
box
1
percent.(
34
cement.( for
)
33
)
ten days
Also,
the
a piece of Teflon was placed next
plates ir creased after to the plates in a UV box.( 3 5 )
Similarly,
from
V'ton
on
polyimide
silver-filled
with
tape exposed
experienced contact
bonded
been
shavings
taken
quartz
an
O-ring
experienced
a
weight loss of 3.7 percent after 24 hours in UV box 1. At the end of
the
24
Semicon,''ctor without
hours,
surfaces
adversely
the
Viton
have
affecting
surfaces
been
successfully
the functioning
16
had of
become
sticky.
UV/ozone
cleaned
the
dpvices.
For
example,
after a 4K static
UV/ozone for continued IC
had
120 min in
to
made
integrated
a commercial
function
been
RAM
without
using
any
circuit was exposed
UV/ozone change
in
silicon
n-channel
cleaner,
the device
performance.
gate
to
(This with
technology,
1 to 1.5 um junction depths.)(36) 3.6
Rate Enhancement Techniques UV/ozone
by
investigated p.-
cleaning
"rate
enhancement" Chiu. (37)
and
Zafonte
techniques
have
been
gas
phase
on
Experiments
enhancement
techniques included a comparison of the cleaning rates
in
dry oxygen,
dry air,
moist air,
and moist oxygen.
The moist air
and moist oxygen consisted of gases that had been bubbled through water.
Oxygen
Experiments on liquid enhancement techniques consisted
aiso tried. of
-.
that had been bubbled through hydrogen peroxide was
a drop-wise
addition either of distilled
peroxide
solutions
surfaces.
Most of the
of
variou3
sample
or
concentrations
surfaces
to
of hydrogen the
sample
consisted of various
types
of photoresist on silicon wafers. The gas phase "enhancentent" to
slight
increases
A/min without water
and
both
for
implanted
the
in
and
of
liquid-phase rate
by
UV/ozone,
1500-1600
1016
800
A/min
for
removal
to 900
(to
The
The
techniques 100
to
200
heavily
ion
were
not
)
"enhanced"
or
not.
A/min
negative
for positive photoresists( 3 8 )
of UV/ozone by one manufacturer were reported equipment. (38) The fast removal rate was achieved at using a 253.7 nm source of UV,
(3-20
enhancement).
atoms/cm2
whether
of
negligible
enhancement
resists.
to
rates 0
with
enhancements
(1015
removal
photoresist
3-30 A/min
implanted
affected
Photoresist
vs.
significant
resists
techniques resulted in
rates
peroxide
non-ion
significantly
photoresists
in
enhancement
hydrogen
resulted
A/min)
_
water
cleaning 300 0 C by
*"
a cilent discharge ozone generator, a heater built into the cleaning chamber, and by using oxygen from
|.."
17
a gas cylinder to generate the ozone. UV/ozone cleaner
is
shown in
A schematic drawing of this
Figure 5.
The photoresist stripping rate vs.
substrate temperature for three
different photoresists is
Figure 6.
shown in
FILTER
02
FLOW
METER
=
N2
PURGE
Figure 5.
CLEANING
GENERATOR
CHAMBER
EXHAUST
Schematic drawing of a UV/ozone cleaner that uses a
silent-discharge
4.
OZONE
ozone generator.
THE MECHANISM OF UV/OZONE CLEANING The
available
evidence
indicates
that
UV/ozone
cleaning
primarily the result of photocensitized
oxidation processes,
represented
The contaminant
are
schematically
excited
wavelength
and/or
UV light.
simultaneously
in
Figure
dissociated Atcmic
when
2
by
oxygen
is
7.
the
absorption
and ozone(
dissociated
by
1
S,1
9
the
is
as is
molecules of
short
) are produced absorption
of
UV with wavelengths less than 245.4 nm. Atomic oxygen is also produced(1 8 , 1 9 ) when ozone is dissociated by the absorption of the
UV
and
contaminant
lonqer molecules
wavelengths and
the
18
of
radiation. free
The
radicals
excited produced
1600 -
- 1400 "E 1200 -
Z, 1000
/
AOMR 83
A
OOFPR 800
*AZ-1350
-
o
A
S800-
000
400
0"
-"
200 -
3000
2000
1000
00
Substrate Temperature (CC) Figure 6.
Photoresist stripping rate vs. substrate temperature
by
the
dissociation
of
oxygen to form simpler,
The ground just
ene,.gy
state below
0
contaminant
245.4
molecules
volatile molecules,
required atoms
for three types of photoresists.
to
dissociate
corresponds nm
the
to
react
with
atomic
such as C0 2 , H2 0,
and
02
into
two
at
and
an
245.4
molecule
nm.
absorption
However,
of
02
is
very
weak. (7,18,19) 3..The absorption coefficient .....r 2q, as th decreasi..- wavelengths, •ji, iU,u
increases rapidly below Yigure 2. is shown in
For producing
is
03,
a convenient wavelength
by low-?-essure Hg Similarly, since
discharge lamps in fused most organic molecules
absorption band between 200 nm and 300 nm, emitted
by
the
the 184.9 nm emitted
same
lamps
is
19
quartz have
envelopes. a strong
the 253.7 nm wavelength
useful
for
excitinq
or
IONS CONTPMINANT FREE RRDICALS MOLECULES + h -->EXCITED MOLECULES NEUTRAL MOLECULES VOLATILE MOLECULES (CO?2, H20, N2 ,etc.) 0? + hv2
> 0,03
Figure 7.
Simplified
UV/ozone cleaning
dissociating dissociate is
contaminant
relatively
actual
photochemical
more
complex
presence
As
alone.
was UV
light
2,000
et
al.( 102
to
are
materials
that
nm;
300 nm.
shown
in ,
2 9
10
ozone
is
shown
during
Figure
such
heat
UV
studies found
-fold, as
the
absorption
in
Figure
UV/ozone
7.
For is
3.
a
The
cleaning
example,
the
promoted by the
C0 the
that
and 2
combination
produced
than
had 4
the
to
such as N 2 and CO 2 .
their )
required
The absorption re'aches
as
in
energy
however,
occurring
faster
increased mild
1,140
previously, and
times 2 6
ozone
temperature
F
described
with
found
that
The
of ozonc by 184.9 nm photons
Similarly,
Prengle
processes
of other molecules,
wavelength to
to
nm wavelength,
than
rate of production
200
molecules.
weak above
maximum near the 253.7
of
process.
ozone corresponds
by ozone
are
schematic representation
,
the 1120,
reaction
increases
20
the
clean liJht
of
short
surfaces
about
alone
of
wastwM-er
UV
enhances
products and
N2.
rater. OV/ozone
of
or
ozone
tLeatment, the
the
reaction reactions
Increasing Mattox(
3
cleaning
9)
the also
rates.
Bolon and Kunz,(1)
on the other hand,
had found that the rate of
ozone depolymerization of photoresists significantly between 1000c and 3000C. The
did not change rate of destruction
of microorganisms was similarly insensitive increase from room temperature to 40oC.(30)
to a temperature One manufacturer
of U//ozone cleaning equipment claims that the rate photoresist stripping by UV/ozorne increases severalfold as temperature is raised from 200C to 300 0 C.(38)
5.
of the
UV/OZONE CLEANING IN VACUUM SYSTEMS Soweil et al.(2)
reported
oxygen was present in a desorbocd gases from the irradiation,
that,
vacuum walls
when
10-4 torr pressure
system, of the
short-wavelength system. During
the partial pressure of oyygen decreased,
of UNV UV
while that
of CO 2 and H20 increased. One must exercise caution in using a mercury UV scurce in a vacuum system because, should the lamp envelope break or leak, mercury can enter and ruin the usefulness of the system. Mercury has a high vapor pressure; its complete removal from a vacuum chamber is a difficult task. Other types of UV sources, such as xenon or deuterium lamps, may be safer to use in vacuum systems. The UV light can also be radiated into systems through sapphire or quartz
windows,
or through deep-UV fiber optic bundles. A small partial pressure of oxygen should be present during UV cleaning. Caution cryopumped potentially
must vacuum
also
be
exercised
system,
explosive,(4
0)
when
since
particularly
using
UV/ozone
cryopumped during
in
ozone
regeneration
a is of
the cryopump. A convenient method of dealing with this potential hazard is to use two kinds of UV sources, one an ozonekiller" source.(41) an "ozone the other source, generating (See next section.)
21
6.
SAFETY CONSIDERATTONS
,In aware
constructing
of
the
wavelength
safety
UV
light
within a short attached
to
are
opened,
thc
completely cleaning proper
In
the
that
clothing
and
The
and
can
now
well known
factor
result
in in
azre
solar
skin cancer.(
can
c,-4i UV
4 4
)
been
found
as
short
mutations
be
eye
injury
switches
when
the doors If
without
the
case
the being
if
a
bonder),
a
wire
UV
safety
UV
then
glasses
wavelengths
that
needed
these wavelengths
Another
is nm
most
Because are
with
mutation.
the
prime
a
significant
portion
not
is
to which people Ozone
a
potential
22
lIV
wavelength filters
not
exposed
normally a
study
out
of
to the
found that in causing
nm wavelengths.
Therefore,
be
the
exposed
to
short
low doses of
which
is
highly
toxic.
In
one must ensure that the ozone
are exposed is
solar
damage to human cells.
ozone,
setting up a UV cleaning facility,
risk
effective
in
is
causative
of
for UV/ozone cleaning because even can cause significant
It
the atmosphere
however,
above-300
by human
crosslinks,
cancer rates, it was times more effective
personnel
safety hazard
standard. (45)
and
the
nm.
cell
is
)
absorbed
DNA-protein
and
290-320 be
254
than were the
essential
4 3
,
humans
as
to
radiation
effects of UV radiation on skin the 254 nm wavelength was many
I:
short-
al.,
used
strongly
death,
The to
is lead
4 2
cancer,(
wavelengths,
wavelengths
OSHA
(e.g.,
radiation
cancer,
that
human eye
UV
absorption
in
shorter
levels
that,
be
into
causing skin cancer.
is
lamps
incorporated
region for
it
and
be
to
automatically.
as might
has
cell
off
should
exposure skin
manner
example,
radiation
the
with
used by Vig et
shut UV
one
should be worn to prevent skin and eye damage.
DNA.
factor
facility,
serious
a
eye protection
Short-wavelength cellular
such
the
(for is
facility
in
lamps
enclosed
cause
the UV boxes
doors
UV
cleaning
associated
which can
demands
side f]aps)
UV/ozone
hazards
time.
are
application
a
do
not exceed hazard
in
0.1 a
ppm,
the
cryopumped
L vacuum system because cryopumped ozone can become explosive under certain conditions.(40)
is
One method of minimizing the hazards associated with ozone to use two types of short-wavelength ultraviolet sources for
UV/czone a
cleaning(
low-pressure
4 1 ):
one,
mercury
an
light
ozone in
a
generating fused
quartz
UV lamp, envelope,
e.g., the
other,
a UV lamp that does not generate ozone but which emits one or more wavelengths that are strongly absorbed by ozone, e.g., a low-pressure mercury light in a high-silica glass tube which emits at 253.7 nm but not at 184.9 nm. Such a non-ozone generating UV source can be used as an "ozone killer." For example, in one cryopumped
vacuum system,
UV/ozone cleaning was performed in up to 20 torr of oxygen. After the cleaning was completed and the ozone-generating UV lamp was turned off, ten minutes of "ozone killer" UV light reduced the concentration of ozone to less than 0.01 ppm, a level that is safe for cryopumping.( 4 6 ) Therefore, with the "ozone killer" lamp, ozone concentrations were reduced by Without the at least a factor of 100 within ten minutes. "ozone killer" lamp, the half-life of ozone is three days at 20 0 C.(47)
The decomposition
of ozone can also
be greatly
accelerated
in through the use of catalysts. For example, prior to 1980, high-flying aircraft, ozone was found co be a causative factor for flight personnel and passengers experiencing headaches, eye, nose and throat irritations and chest pains. Passing the aircraft cabin air through a precious metal catalytic converter reduced the ozone concentration from the 1 ppm to 2 ppm level present in the troposphere to the low levels required for passenger comfort and safety. (48)
j
23
7.
UV/OZONF CLEANING FACILITY CONSTRUCTION
The cleaning UV/ozone.
materials facility
chosen
should
Polished
for
the
construction
remain uncorroded
aluminum
with
a
of
a
UV/ozone
by extended exposure to
relatively
thick
anodized
6
It is such as Alzak,( ) is one such material. oxide layer, resistant to corrosion, has a high thermal conductivity, which helps to prevent heat buildup, and is also a good reflector of short wavelength UV. Most other metals, including silver, are poor reflectors in
the UV range.
Initially,
Vig et al.
used an ordinary shop-variety aluminum
sheet for UV box construction,
which was found not to be a good
material because, in time, a thin coating of white powder (probably aluminum oxide particles) appeared at the bottom of the boxes. Even in a UV box made of standard Alzak, after a couple of years' usage, white spots appeared on the Alzak, probably due to pinholes in the anodization. To avoid the possibility of particles being generated inside the UV/ozone cleaning facility, the facility should be inspected periodically for signs of corrosion. Using "Class M" Alzak may also aid in avoiding particle generation, since this material has a much thicker oxide coating, and is made for "exterior marine service," whereas standard Alzak is for "mild interior service." Some commercially available UV/ozone cleaners are now constructed of stainless steel.( 4 9 , 5 0 ) To date, no corrosion problems have been reported with such cleaners. Organic materials should not be present in the UV cleaning box. For example, the plastic insulation usually found on the lads of UV lamps should be replaced with inorganic insulation such as glass or ceramic. The box should be enclosed so as to minimize recontamination by circulating air, and to prevent accidental UV exposure and ozone escape.
24
The
most
light are pure
widely
the mercury
fused quartz
available
sources
arc lamps.
envelopes
of
short-wavelength
Low-pressure
operate near
mercury
UV
lamps
room temperature,
in
emit
approximately 90 percent at the 253.7 nm wavelength, and generate sufficient ozone for effective surface cleaning. Approximately five percent of the output of these lamps is at 184.9 nm. Mediuumand high-pressure UV lamps( 7 ) generally have a much higher output in the short wavelength UV range. These lamps also emit a variety of additional wavelengths below 253.7 nm, which may enhance their cleaning action. However, they operate at high temperatures lifetime,
(the
higher
envelopes cost,
are
near
and present
red-hot),
have
a
shorter
a greater safety hazard.
The
mercury tubes can be fabricated in a variety of shapes to fit different applications. In addition to mercury arc lamps, microwave-powered mercury vapor lamps are also available.(51)
Other available sources of short-wavelength UV include xenon lamps and deuterium lamps. These lamps must also be in an envelope transparent to short-wavelength UV, such as quartz or sapphire, if no separate ozone generator is cleaning facility, one should
to be used. In setting up a UV choose a UV source which will
generate enough UV/ozone to allow oxidation of contaminants. However,
for rapid photosensitized too high an output at
the ozone-generating wavelengths can be counterproductive because a high concentration of ozone can absorb most of the UV light before it
reaches the samples.
The samples should be placed
as close to reaching the parts to be which can be parts to be
the UV source as possible to maximize the intensity samples. In the UV cleaning box 1 of Vig et al., the cleaned are placed on an Alzak stand the height of adjusted to bring the parts close to the UV lamp. The cleaned can also be placed directly onto the tube
if
is
the
box
built
so that
the tube
is
on the bottom of the
box.(52) An alternative to using low-pressure mercury lamps in fused quartz envelopes is to use an arrangement similar to that of 25
box
2,
shown
in
Figure
1. Such
available
commercially, (38)
and
UV
source
that
in
Figure
5.
a
shown
cleaning with
rate
that
UV/ozone
is
UV/ozone
thh
253.7
nm
of
this
manufacturer faster
mL:ch
cleaners
cleaner,
now
uses silent-discharge-generated
generates The
a
that
than
do
that which
not
ozone
wavelength, cleaner
contain
is
also as
claims
is a
achievable
separate
ozone
generators.
This cleaner also uses oxygen from a gas cylinder and
a built-in
sample heater
that
may
further
increase
the cleaning
rate.
8.
APPLICATIONS The
UV/ozone
applications.
cleaning
A major
film deposition,
procedure
use
as is
is
is
substrate
widely used in
now
used
cleaning
in
numerous
prior
the quartz crystal
to
thin
industry
during the manufacture of quartz crystal resonators for clocks and frequency control. There is probably no other device of which the performance For that
is
so critically the
example, the
week,
contamination that
no
adsorption
or
in
the
hermetically
from 106.
such
The
rate
more
cleanliness
sealed
fabrication
laboratory,
week!
In
the
riethod
which
of the
a
1010
per
monolayer
of
on
frequency
must
masks,
author's
by about
therefore
be
relies
also on
quartz
being
the
the 10-4
resonator
at several
points
and storage containers.
applied
adhesion
an ultrahigh vacuum.(1 4 ,53-55) surfaces will weld together
such
such as for cleaning and storing
resonator parts, is
is
resonator
parts
UV/ozone has been used
during the fabrication sequence,
process
two
MHz
contamination transfer within tha-n less is enclosure resonator
per
The
5
cleanliness.
of
monolayers
metal tools,
than
changes
surface
one
desorption
a device
surface
upon
for
requirement
change
frequency
whereas
one part
aging
dependent
in
between
a
hermetic clean
sealing
surfaces
in
it has been shown that metal under near-zPro forces if the
26
are
surfaces
of
a
been
ceramic
flatpack
shown(53-55)
that
enclosed it
gasket
gold
is
quartz
feasible
gold
between
in
seals
hermetic
excellent
providing
currently
A
surfaces
aluminum oxide sealing
cleaned)
(QV/-)one
metallized
clean.
atomically
production
the
resonator.
It
has
to achieve hermetic
gasket between by pressing a clean aluminum unmetallized aluminum oxide ceramic surfaces. The same adhesion phenomenon between clean
is
also seals
two
clean,
(UV/ozone cleaned)
gold surfaces has been applied to the construction of a novel surface contaminant detector.( 5 6e57) The rate of decrease in the coefficient
of
adhesion between
used as a measure of
freshly cleaned
gold contacts
is
level in
the gaseous condensable contaminant
the atmosphere. The process has also been applied of
wire
bonds,
example,
it
bonding
especially
has
been
process
contaminants dependence
is
shown(
highly
58
to improve the reliability
at
reduced
,59)
that
temperature
are present on the bonding can
be
greatly
reduced
temperatures.
the
thermocompression
dependent surfaces.
by
UV/ozone
is shown in methods on
shear
was
method
effective
surfaces.(60) substrate
UV/ozone of
UV/ozone
surfaces
cleaning also
during
the
being
used
processing
for of
organic
The temperature of
Figure 8. gold ball to
be
the
cleaning thin
film
the In a bond most gold
from
contaminants
cleaning is
found
when
cleaning
surfaces just prior to bonding, as study of the effects of cleaning strength,
For
alumina hybrid
circuits.(61) A number of cleaning methods were tested when appearance traced
of
to
thermal/flash
organic
protective
contaminants
on
the nonuniform
electrooptic the
goggles
electrooptic
was
wafers.
UV/ozone proved to be the most effective method for removing these contaminants, the goggles.(
and thus it 62
was chosen for use in
)
27
the production
of
100 900
-90
W
uV CLEANED,