APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
P oximi y E ec in Ga e Fab ica ion Using
Pho oli hog aphy Technique
Joanna PRAZMOWSKA, Ko nelia INDYKIEWICZ, Bogdan PASZKIEWICZ,
Regina PASZKIEWICZ
Depa men o Mic oelec onics and Nano echnology, Facul y o Mic osys em Elec onics and Pho onics,
W oclaw Uni e si y o Science and Technology, Wyb zeze Wyspianskiego 27, 503 70 W oclaw, Poland
joanna.p azmowska@pw .edu.pl, ko nelia.indykiewicz@pw .edu.pl, b[email p o ec ed],
egina.paszkiewicz@pw .edu.pl
DOI: 10.15598/aeee. 15i2.2024
Abs ac . In he pape he echnological ac o s in lu-
encing es s uc u e ga e leng h we e desc ibed. The
in luence o es s uc u e ga e placemen (Scho ky
me alliza ion be ween ohmic con ac s, on mesa and on
GaN su ace) was analyzed and discussed. Mo eo e ,
a ious dis ances be ween ohmic con ac s pa hs we e
es ed. Excep o expe imen al in es iga ions, simula-
ions using ini e elemen s me hod in COMSOL we e
pe o med o he same s uc u e. The modelling e-
sul s e ealed c ucial impac o a gap beyond he mask
on he elec ic ield dis ibu ion in pho o esis laye .
The smalles alue o ela i e e o o es inge leng hs
was obse ed o inge pa s placed be ween ohmic
pa hs on mesas. I was explained by hicke li -o dou-
ble laye be ween ohmic pa hs and he smalles Y-gap
compa ed o es inge s placed on mesa and ou side o
i . Simula ion did no b ing an explana ion o la ge
alues o ela i e e o o smalle dis ance be ween
ohmic pa hs.
Keywo ds
AlGaN/GaN ansis o s, h-line li hog aphy,
p oximi y e ec .
1. In oduc ion
Con inuous inc ease o scale o in eg a ion o elec onic
de ices cause ha he op ical li hog aphy aced i s es-
olu ion limi a ion o used wa eleng h. Acco ding o
Rayleigh’s equa ion, enhancemen o esolu ion could
be assu ed by dec ease o wa eleng h o highe nu-
me ical ape u e o lens sys ems [1]. To ob ain highe
esolu ion, he wa eleng h was dec eased om G-line
(435 nm) o I-line (365 nm), u he o 248 nm (ex-
cime lase sou ce wi h K F) and o 193 nm (A F)
[2] and [3]. Also 8 a ious me hods o image o ma ion
ha e been de eloped e.g. phase shi ing me hod [1] and
[2]. Mo eo e , he e a e e o s o de elopmen o su-
pe lenses [4], ex eme ul a iole and beyond ex eme
ul a iole li hog aphy [5], su ace-plasmon pola i on
esonance [6].
Addi ionally, cons an s depending on esis ma e ial,
p ocess echnologies and image o ma ion echniques
play an impo an ole. The p oximi y e ec de ined
as a a ia ion in pa e n wid h due o p oximi y o
o he nea by ea u es is well known o elec on-beam
li hog aphy [7]. The op ical p oximi y e ec was s ud-
ied e e ing o ea u es ypical o ansis o s ab ica-
ion. In he pape he echnological ac o s in luencing
es s uc u e ga e esolu ion we e desc ibed. The in-
luence o es s uc u e ga e placemen was discussed.
Obse ed phenomena we e analyzed also based on com-
pu e simula ions esul s.
2. Expe imen al De ails
The dedica ed es s uc u es we e made du ing
AlGaN/GaN HEMT (High Elec on Mobili y T ansis-
o ) de ices ab ica ion. The AlGaN/GaN he e os uc-
u es ab ica ion in me al-o ganic apou phase epi axy
echnique was desc ibed elsewhe e [8]. Each ansis o
in he module on he wa e consis ed o wo es s uc-
u es ha di e ed in designed dis ance be ween ohmic
con ac pa hs:
• ype 1 - designed dis ance o 3 µm plus designed
leng h o es inge ,
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 358
APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
• ype 2 - designed dis ance o 4 µm plus designed
leng h o es inge ).
Bo h ypes o dedica ed es s uc u es con ained six
inge s o a ious leng hs in pu pose o indi ec analysis
o ga es leng hs and chosen ac o s in luencing i s alue.
The designed inge s leng hs we e #1 – 0.6, #2 – 0.8,
#3 – 1, #4 – 1.4, #5 – 2, #6 – 5 µm, espec i ely.
Addi ionally, he es s uc u es emb ace h ee di -
e en a eas on which he es s uc u es inge s we e
placed Fig. 1:
•a ea A - Scho ky me alliza ion on mesa be ween
ohmic con ac s,
•a ea B - Scho ky me alliza ion on mesa,
•a ea C - Scho ky me alliza ion on GaN su ace
(ou side o he mesa).
SECTION POLICIES VOLUME: XX | NUMBER: X | 2015 | MONTH
© 2015 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 2
Addi ionally, he es s uc u es emb ace h ee
di e en a eas on which he es s uc u es inge s we e
placed (Fig. 1):
• a ea A – Scho ky me alliza ion on mesa be ween
ohmic con ac s,
• a ea B – Scho ky me alliza ion on mesa,
• a ea C – Scho ky me alliza ion on GaN su ace
(ou side o he mesa).
The dedica ed es s uc u es we e ab ica ed in
AlGaN/GaN he e os uc u es by pho oli hog aphy
echnique using Ca l Suess MA56 mask aligne wo king
in h-line mode. Mesas we e e ched h ough he SiO2
mask (300 nm hick, deposi ed by plasma-enhanced
chemical apo deposi ion) in eac i e ion e ching (RIE)
sys em. Fo he RIE p ocess a Cl2:BCl3 mix u e o
gasses was used. Time o e ching equal o 70 s ga e
heigh s o mesas in he ange om 70 o 87 nm.
Me alliza ion con ac s we e deposi ed in UHV sys em by
he mal and e-beam e apo a ion. The me alliza ion s ack
o Ti/Al/Mo/Au he mally o med in apid he mal
annealing sys em (a 820°C o 60 s) was used as ohmic
con ac . Scho ky con ac s we e o Ru/Au (30/150 nm)
double laye .
Mesa s uc u es pa e ns we e made in s anda d
li hog aphy (using Mic oposi S1813 Pho o Resis -
Shipley) while ohmic and Scho ky con ac s we e
ab ica ed in li -o echnology using double laye –
Shipley Mic oposi LOL 2000 and Megaposi SPR 700-
1.0 (DOW). Pa e n was ans e ed om ch omium mask
in acuum con ac ( he acuum seal in la es o o m a
chambe be ween mask and sample, which is hen
e acua ed). The wa eleng h o exposu e UV ligh was
405 nm and i s in ensi y o 18 mW/cm2. The p e-bake
ime, ime o exposu e, LOL2000 and S1813 hicknesses,
ime o de elopmen as well as ul asounds powe du ing
de elopmen we e op imized o HEMTs ga es
ab ica ion. Addi ionally, s ep o edge bead emo e o
minimize he dis ance o he mask and sample du ing
exposu e was applied.
Finge s leng hs o es s uc u es we e measu ed
wi hin a se ies o samples made in simila en i onmen al
condi ions o li hog aphy p ocess. The yield o each
sample exceeded 90 %. The leng hs we e measu ed
epea edly nea he middle o each inge based on
scanning elec on mic oscope (SEM) images.
Quali a i e analysis o he elec ic ield dis ibu ion in
he Y-gap, pho o esis and LOL (Fig. 2) du ing exposing
was pe o med based on simula ions using fini e elemen
analysis (FEA) by COMSOL – he comme cial so wa e.
The s uc u e used o simula ions was simila wi h ha
ob ained in expe imen s. I con ains ( op-down):
• mask, wi h ch omium a eas and he X-gap (wid h
as designed es inge leng h) – illed wi h
acuum du ing exposu e,
• Y-gap, esul ing om non-uni o m spin coa ing o
he pho o esis due o ohmic con ac s p esence on
he AlGaN/GaN su ace – illed wi h acuum
du ing exposu e,
• pho o esis and LOL laye ,
• ohmic me alliza ion pa hs placed in he dis ances
as in expe imen s (i.e. 3 and 4 µm plus designed
ga e leng h, depending on ype o he s uc u e),
• AlGaN/GaN s uc u e.
Fig. 2: The s uc u e used o simula ions in COMSOL.
The exposu e pa ame e s o e ically inciden ligh in
simula ions we e he same as o expe imen al pa o he
in es iga ion. In he able 1 he e ac i e index alues o
used ma e ials o 405 nm wa eleng h a e shown.
Table 1 Re ac i e index alues o used ma e ials
SPR
700
LOL
2000
Al2O3 AlGaN GaN Glass Me alliza ion
1.7
1.6
1.76
2.25
2.55
1.5
1.52
3. Resul s
In he i s s ep he mean alue o inge s leng hs in h ee
a eas was es ima ed (Fig. 3). Addi ionally, he s anda d
de ia ion was calcula ed. I s alue was he smalles o
inge s pa s loca ed be ween he ohmic con ac s pads.
S anda d de ia ions o leng hs o inge s designed o 1
µm, as ga e leng h o HEMT s uc u es, we e simila .
A B C
Fig. 1: SEM image o es s uc u e.
Fig. 1: SEM image o es s uc u e.
The dedica ed es s uc u es we e ab ica ed in
AlGaN/GaN he e os uc u es by pho oli hog aphy
echnique using Ca l Suess MA56 mask aligne wo king
in h-line mode. Mesas we e e ched h ough he SiO2
mask (300 nm hick, deposi ed by plasma-enhanced
chemical apo deposi ion) in Reac i e Ion E ching
(RIE) sys em. Fo he RIE p ocess a Cl2:BCl3mix-
u e o gasses was used. Time o e ching equal o 70 s
ga e heigh s o mesas in he ange om 70 o 87 nm.
Me alliza ion con ac s we e deposi ed in UHV sys em
by he mal and e-beam e apo a ion. The me alliza ion
s ack o Ti/Al/Mo/Au he mally o med in apid he -
mal annealing sys em (a 820 ◦C o 60 s) was used
as ohmic con ac . Scho ky con ac s we e o Ru/Au
(30/150 nm) double laye .
Mesa s uc u es pa e ns we e made in s anda d
li hog aphy (using Mic oposi S1813 Pho o Resis -
Shipley) while ohmic and Scho ky con ac s we e ab-
ica ed in li -o echnology using double laye - Ship-
ley Mic oposi LOL 2000 and Megaposi SPR 700 – 1.0
(DOW). Pa e n was ans e ed om ch omium mask
in acuum con ac ( he acuum seal in la es o o m
a chambe be ween mask and sample, which is hen
e acua ed). The wa eleng h o exposu e UV ligh
was 405 nm and i s in ensi y o 18 mW·cm−2. The
p e-bake ime, ime o exposu e, LOL2000 and S1813
hicknesses, ime o de elopmen as well as ul asounds
powe du ing de elopmen we e op imized o HEMTs
ga es ab ica ion. Addi ionally, s ep o edge bead e-
mo e o minimize he dis ance o he mask and sample
du ing exposu e was applied.
Finge s leng hs o es s uc u es we e measu ed
wi hin a se ies o samples made in simila en i onmen-
al condi ions o li hog aphy p ocess. The yield o each
sample exceeded 90 %. The leng hs we e measu ed e-
pea edly nea he middle o each inge based on Scan-
ning Elec on Mic oscope (SEM) images.
X
Y-gap
pho o esis –890 nm
LOL –240 nm
ohmic me alliza ion –210 nm
AlGaN/GaN –7.9 mm/25 nm
Al2O3-1 mm
Fig. 2: The s uc u e used o simula ions in COMSOL.
Quali a i e analysis o he elec ic ield dis ibu ion
in he Y-gap, pho o esis and LOL (Fig. 2) du ing ex-
posing was pe o med based on simula ions using Fi-
ni e Elemen Analysis (FEA) by COMSOL - he com-
me cial so wa e. The s uc u e used o simula ions
was simila wi h ha ob ained in expe imen s. I con-
ains ( op-down):
•mask, wi h ch omium a eas and he X-gap (wid h
as designed es inge leng h) - illed wi h acuum
du ing exposu e,
•Y-gap, esul ing om non-uni o m spin coa ing o
he pho o esis due o ohmic con ac s p esence on
he AlGaN/GaN su ace - illed wi h acuum du -
ing exposu e,
•pho o esis and LOL laye ,
•ohmic me alliza ion pa hs placed in he dis ances
as in expe imen s (i.e. 3 and 4 µm plus designed
ga e leng h, depending on ype o he s uc u e),
•AlGaN/GaN s uc u e.
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 359
APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
The exposu e pa ame e s o e ically inciden ligh
in simula ions we e he same as o expe imen al pa
o he in es iga ion. In he Tab. 1 he e ac i e in-
dex alues o used ma e ials o 405 nm wa eleng h a e
shown.
Tab. 1: Re ac i e index alues o used ma e ials.
SPR
700
LOL
2000 Al2O3AlGaN GaN Glass Me alli-
za ion
1.7 1.6 1.76 2.25 2.55 1.5 1.52
3. Resul s
In he i s s ep he mean alue o inge s leng hs in
h ee a eas was es ima ed (Fig. 3). Addi ionally, he
s anda d de ia ion was calcula ed. I s alue was he
smalles o inge s pa s loca ed be ween he ohmic
con ac s pads. S anda d de ia ions o leng hs o inge s
designed o 1 µm, as ga e leng h o HEMT s uc u es,
we e simila .
# 1 # 2 # 3 # 4 # 5 # 6
0
1
2
3
4
5
6
7 b e w e e n o h m i c p a h s o n m e s a
o n m e s a
o n G a N
m e a n ( µm )
i n g e n o .
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
s a n d a d d e i a i o n ( µm )
Fig. 3: Mean alues and s anda d de ia ion o inge leng hs o
es s uc u es.
The ela i e e o o inge s leng hs placed on h ee
a eas o bo h ypes o es s uc u es is p esen ed in
Fig. 4(a) and Fig. 4(b).
The smalles alue o ela i e e o was ob ained
o inge pa s placed be ween ohmic pa hs on op o
mesas o s uc u es o ype 1 (depic ed as a ea A in
Fig. 1) as well as ype 2 (depic ed as a ea B in Fig. 1).
Due o la ge heigh o ohmic con ac s (Fig. 5) com-
pa ed o li –o double laye heigh he e lec ion on
he me alliza ion slope and i s i egula i ies was ex-
pec ed o leng hen he inge s. The obse ed inge s
leng h could be a consequence o hicke li -o double
laye be ween ohmic pa hs. The hicke li -o double
laye is an e ec o he spin-o echnique used o sam-
ples coa ing by esis s. The ime o exposu e as well as
ime o de elopmen was equal o whole sample hus
hicke laye o esis s could gi e sho e inge s. Fo
# 1 # 2 # 3 # 4 # 5 # 6
0
5 0
100
150
e l a i e e o ( % )
i n g e n o .
b e w e e n o h m i c p a h s o n m e s a
o n m e s a
o n G a N
y p e 1
(a) Type 1.
# 1 # 2 # 3 # 4 # 5 # 6
0
5 0
100
150 y p e 2
e l a i e e o ( % )
i n g e n o .
b e w e e n o h m i c p a h s o n m e s a
o n m e s a
o n G a N
(b) Type 2.
Fig. 4: Rela i e e o o inge s leng hs placed on h ee a eas
o bo h ypes o es s uc u es.
Fig. 5: SEM image o es s uc u e inge be ween ohmic con-
ac s pa hs.
inge s #1, #2 and #3 obse ed leng h was la ge o
pa s placed on GaN su ace compa ed o hose placed
on op o mesa. Leng hening o inge s wi hin his a ea
could be caused by e lec ing o exposu e UV-ligh on
whiske s ha occu ed on GaN su ace and u he ex-
posu e o pa e ns Fig. 6.
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 360
APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
SECTION POLICIES VOLUME: XX | NUMBER: X | 2015 | MONTH
© 2015 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 3
The ela i e e o o inge s leng hs placed on h ee
a eas o bo h ypes o es s uc u es is p esen ed in Fig.
4 a) and b).
#1 #2 #3 #4 #5 #6
0
1
2
3
4
5
6
7
be ween ohmic pa hs on mesa
on mesa
on GaN
mean (µm)
inge no.
0.0
0.1
0.2
0.3
0.4
0.5
s anda d de ia ion (µm)
Fig. 3: Mean alues and s anda d de ia ion o inge leng hs o es
s uc u es.
a)
#1 #2 #3 #4 #5 #6
0
50
100
150
ela i e e o (%)
inge no.
be ween ohmic pa hs on mesa
on mesa
on GaN
ype 1
b)
#1 #2 #3 #4 #5 #6
0
50
100
150
ype 2
ela i e e o (%)
inge no.
be ween ohmic pa hs on mesa
on mesa
on GaN
Fig. 4: Rela i e e o o inge s leng hs placed on h ee a eas o bo h
ypes o es s uc u es.
The smalles alue o ela i e e o was ob ained
o inge pa s placed be ween ohmic pa hs on op o
mesas o s uc u es o ype 1 (depic ed as a ea A in Fig.
1) as well as ype 2 (depic ed as a ea B in Fig. 1). Due o
la ge heigh o ohmic con ac s (Fig. 5) compa ed o li –
o double laye heigh he e lec ion on he me alliza ion
slope and i s i egula i ies was expec ed o leng hen he
inge s. The obse ed inge s leng h could be a
consequence o hicke li -o double laye be ween
ohmic pa hs. The hicke li -o double laye is an e ec
o he spin-o echnique used o samples coa ing by
esis s. The ime o exposu e as well as ime o
de elopmen was equal o whole sample hus hicke
laye o esis s could gi e sho e inge s. Fo inge s #1,
#2 and #3 obse ed leng h was la ge o pa s placed on
GaN su ace compa ed o hose placed on op o mesa.
Leng hening o inge s wi hin his a ea could be caused
by e lec ing o exposu e UV-ligh on whiske s ha
occu ed on GaN su ace and u he exposu e o pa e ns
Fig. 6.
Fig. 5: SEM image o es s uc u e inge be ween ohmic con ac s
pa hs.
Fig. 6: SEM image o es s uc u e inge on a ious a eas.
The ela i e e o o es s uc u es inge s leng hs
(Fig. 7) indica ed a ec ion o ohmic con ac pa hs
dis ance on he leng hs.
Fig. 7: The ela i e e o o es s uc u es inge s leng hs o a ious
ohmic con ac pa hs dis ance.
La ge alues o ela i e e o we e obse ed o
smalle dis ance hus in luence o UV-ligh sca e ing on
ohmic con ac s slopes could no be excluded. As a esul
o he simula ions he elec ic ield dis ibu ion in he
es inge
on mesa
es inge
on GaN
Fig. 6: SEM image o es s uc u e inge on a ious a eas.
The ela i e e o o es s uc u es inge s leng hs
(Fig. 7) indica ed a ec ion o ohmic con ac pa hs dis-
ance on he leng hs.
# 1 # 2 # 3 # 4 # 5 # 6
0
5 0
100
150
e l a i e e o ( % )
pa h no.
y p e 1
y p e 2
Fig. 7: The ela i e e o o es s uc u es inge s leng hs o
a ious ohmic con ac pa hs dis ance.
La ge alues o ela i e e o we e obse ed o
smalle dis ance hus in luence o UV-ligh sca e ing
on ohmic con ac s slopes could no be excluded. As
a esul o he simula ions he elec ic ield dis ibu-
ion in he s uc u e was ob ained. Only he issue o
h ee designed inge leng hs (i.e. 0.6 µm, 1 µm and
5µm) we e selec ed o u he discussion. The case o
he smalles designed inge leng h (0.6 µm) o bo h
dis ances be ween ohmic con ac s pa hs is p esen ed in
Fig. 8.
Addi ionally, elec ic ield p o iles in ho izon al lines
in ou bounda ies egions we e es ima ed. The lines
we e loca ed be ween:
•I - mask and pho o esis laye o Y-gap,
•II - LOL and ohmic con ac su ace,
•III - ohmic con ac and AlGaN/GaN,
•IV - unde he Y-gap.
(a) 0.6 µm and 3.6 µm.
(b) 0.6 µm and 4.6 µm.
Fig. 8: Elec ic ield dis ibu ion in s uc u e o designed in-
ge leng h and dis ance be ween ohmic con ac s as de-
pic ed ( = 7.402283 ·1014, su ace: Elec ic ield no m
(V/m)).
The elec ic ield p o iles o designed inge leng h
and dis ance be ween ohmic con ac s 0.6 µm and
3.6 µm in Fig. 9(a) and 0.6 µm and 4.6 µm in Fig. 9(b).
The analysis o p o iles indica ed expec ed di ac-
ion on he edges o he ch omium laye co ne s. The
shadowing egion o bo h cases was no e iden as
well as sca e ing on he ohmic con ac s pa hs su aces.
A signi ican in luence o Y-gap p esence on he elec ic
ield dis ibu ion c ould be also obse ed o he in es i-
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 361
APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
2 3 4 5 6 7 8
0
200
400
600
800
1000
1200
E l e c i c i e l d ( V / m )
d i s a n c e ( µm )
I
II
III
IV
(a) 0.6 µm and 3.6 µm.
2 3 4 5 6 7 8
0
200
400
600
800
1000
1200
E l e c i c i e l d ( V / m )
d i s a n c e ( µm )
I
II
III
IV
(b) 0.6 µm and 4.6 µm.
Fig. 9: Elec ic ield p o iles in s uc u e o designed inge
leng h and dis ance be ween ohmic con ac s as depic ed.
ga ed sample as well as o s uc u e o designed inge
leng h and dis ance be ween ohmic con ac s 1 µm and
4µm in Fig. 10(a) and 1 µm and 5 µm in Fig. 10(b).
Op ical e ec s occu ing du ing exposu e we e also
ema kable o he samples o designed inge leng h
and dis ance be ween ohmic con ac s as 5 µm and 8 µm
and 5 µm and 9 µm (no shown) p esen ed in Fig. 11(a)
and Fig. 11(b).
The desi ed elec ic ield dis ibu ion and p o iles in
pho o esis a e p esen ed in Fig. 12. F om he simu-
la ion s uc u e he Y-gap was excluded. Compa ed
o p e iously shown elec ic ield dis ibu ions ha in
Fig. 12 has egula shape wha pe mi ed o ob ain
designed inge leng h. The phenomena indica e g ea
in luence o Y-gap on esul ing ga e leng hs.
The simula ion esul s shown he signi ican in lu-
ence o Y-gap on he elec ic ield dis ibu ion in he
pho o esis . The smalles alue o ela i e e o was
SECTION POLICIES VOLUME: XX | NUMBER: X | 2015 | MONTH
© 2015 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 4
s uc u e was ob ained. Only he issue o h ee designed
inge leng hs (i.e. 0.6 µm, 1 µm and 5 µm) we e selec ed
o u he discussion. The case o he smalles designed
inge leng h (0.6 µm) o bo h dis ances be ween ohmic
con ac s pa hs is p esen ed in Fig. 8.
a)
b)
Fig. 8: Elec ic ield dis ibu ion in s uc u e o designed inge leng h
and dis ance be ween ohmic con ac s a) 0.6 µm and 3.6 µm, b)
0.6 µm and 4.6 µm ( eq=7.402283e14, su ace: Elec ic ield
no m (V/m))
Addi ionally, elec ic ield p o iles in ho izon al lines
in ou bounda ies egions we e es ima ed. The lines we e
loca ed be ween:
• I – mask and pho o esis laye o Y-gap
• II – LOL and ohmic con ac su ace
• III – ohmic con ac and AlGaN/GaN
• IV – unde he Y-gap.
The elec ic ield p o iles o designed inge leng h
and dis ance be ween ohmic con ac s a) 0.6 µm and
3.6 µm, b) 0.6 µm and 4.6 µm a e p esen ed in Fig. 9.
a)
b)
Fig. 9: Elec ic ield p o iles in s uc u e o designed inge leng h
and dis ance be ween ohmic con ac s a) 0.6 µm and 3.6 µm, b)
0.6 µm and 4.6 µm.
The analysis o p o iles indica ed expec ed di ac ion
on he edges o he ch omium laye co ne s. The
shadowing egion o bo h cases was no e iden as well
as sca e ing on he ohmic con ac s pa hs su aces. A
signi ican in luence o Y-gap p esence on he elec ic
ield dis ibu ion c ould be also obse ed o he
in es iga ed sample as well as o s uc u e o designed
inge leng h and dis ance be ween ohmic con ac s a) 1
µm and 4 µm and b) 1 µm and 5 µm in Fig. 10.
a)
b)
(a) 1 µm and 4 µm.
SECTION POLICIES VOLUME: XX | NUMBER: X | 2015 | MONTH
© 2015 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 5
Fig. 10: Elec ic ield dis ibu ion in s uc u e o designed inge leng h
and dis ance be ween ohmic con ac s a) 1 µm and 4 µm, b) 1
µm and 5 µm
Op ical e ec s occu ing du ing exposu e we e also
ema kable o he samples o designed inge leng h and
dis ance be ween ohmic con ac s as 5 µm and 8 µm and 5
µm and 9 µm (no shown) p esen ed in Fig. 11 a) and b).
a)
b)
Fig. 11: Elec ic ield dis ibu ion a) and p o iles b) in s uc u e o
designed inge leng h and dis ance be ween ohmic con ac s o
5 µm and 8 µm.
The desi ed elec ic ield dis ibu ion and p o iles in
pho o esis a e p esen ed in Fig. 12. F om he simula ion
s uc u e he Y-gap was excluded. Compa ed o
p e iously shown elec ic ield dis ibu ions ha in Fig.
12 has egula shape wha pe mi ed o ob ain designed
inge leng h. The phenomena indica e g ea in luence o
Y-gap on esul ing ga e leng hs.
a)
b)
Fig. 12: Elec ic ield dis ibu ion a) and p o iles b) in s uc u e o
designed inge leng h o 1 µm wi hou Y-gap.
The simula ion esul s shown he signi ican in luence o
Y-gap on he elec ic ield dis ibu ion in he pho o esis .
The smalles alue o ela i e e o was obse ed o
inge pa s placed be ween ohmic pa hs on op o mesas.
Apa om hicke li -o double laye be ween ohmic
pa hs he phenomenon could be a esul o occu ence o
he smalles Y-gap compa ed o ha o es inge s
placed on mesa and ou side o i . The e ec o sepa a ion
dis ance was s udied al eady in [8]. Resul s o simula ion
did no gi e an explana ion o la ge alues o ela i e
e o o smalle dis ance be ween ohmic pa hs.
4. Conclusions
In he pape , he echnological ac o s in luencing es
s uc u e ga e leng h we e desc ibed. The s anda d
de ia ion o inge s leng h was he smalles o inge s
pa s loca ed be ween he ohmic con ac s pa hs. Also he
(b) 1 µm and 5 µm.
Fig. 10: Elec ic ield dis ibu ion in s uc u e o designed in-
ge leng h and dis ance be ween ohmic con ac s as de-
pic ed.
SECTION POLICIES VOLUME: XX | NUMBER: X | 2015 | MONTH
© 2015 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 5
Fig. 10: Elec ic ield dis ibu ion in s uc u e o designed inge leng h
and dis ance be ween ohmic con ac s a) 1 µm and 4 µm, b) 1
µm and 5 µm
Op ical e ec s occu ing du ing exposu e we e also
ema kable o he samples o designed inge leng h and
dis ance be ween ohmic con ac s as 5 µm and 8 µm and 5
µm and 9 µm (no shown) p esen ed in Fig. 11 a) and b).
a)
b)
Fig. 11: Elec ic ield dis ibu ion a) and p o iles b) in s uc u e o
designed inge leng h and dis ance be ween ohmic con ac s o
5 µm and 8 µm.
The desi ed elec ic ield dis ibu ion and p o iles in
pho o esis a e p esen ed in Fig. 12. F om he simula ion
s uc u e he Y-gap was excluded. Compa ed o
p e iously shown elec ic ield dis ibu ions ha in Fig.
12 has egula shape wha pe mi ed o ob ain designed
inge leng h. The phenomena indica e g ea in luence o
Y-gap on esul ing ga e leng hs.
a)
b)
Fig. 12: Elec ic ield dis ibu ion a) and p o iles b) in s uc u e o
designed inge leng h o 1 µm wi hou Y-gap.
The simula ion esul s shown he signi ican in luence o
Y-gap on he elec ic ield dis ibu ion in he pho o esis .
The smalles alue o ela i e e o was obse ed o
inge pa s placed be ween ohmic pa hs on op o mesas.
Apa om hicke li -o double laye be ween ohmic
pa hs he phenomenon could be a esul o occu ence o
he smalles Y-gap compa ed o ha o es inge s
placed on mesa and ou side o i . The e ec o sepa a ion
dis ance was s udied al eady in [8]. Resul s o simula ion
did no gi e an explana ion o la ge alues o ela i e
e o o smalle dis ance be ween ohmic pa hs.
4. Conclusions
In he pape , he echnological ac o s in luencing es
s uc u e ga e leng h we e desc ibed. The s anda d
de ia ion o inge s leng h was he smalles o inge s
pa s loca ed be ween he ohmic con ac s pa hs. Also he
(a) Elec ic ield dis ibu ion.
0 2 4 6 8 1 0
0
500
1000
1500
2000
E l e c i c i e l d ( V / m )
d i s a n c e ( µm )
I
II
III
IV
(b) Elec ic ield p o iles.
Fig. 11: Elec ic ield dis ibu ion and p o iles in s uc u e o
designed inge leng h and dis ance be ween ohmic
con ac s o 5 µm and 8 µm.
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 362

APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
(a) Elec ic ield dis ibu ion.
02468
0
200
400
600
800
1000
1200
E l e c i c i e l d ( V / m )
d i s a n c e ( µm )
m a s k / p h o o e s i s
p h o o e s i s / L O L
L O L / s e m i c o n d u c o
(b) Elec ic ield p o iles.
Fig. 12: Elec ic ield dis ibu ion and p o iles in s uc u e o
designed inge leng h o 1 µm wi hou Y-gap.
obse ed o inge pa s placed be ween ohmic pa hs
on op o mesas. Apa om hicke li -o double laye
be ween ohmic pa hs he phenomenon could be a e-
sul o occu ence o he smalles Y-gap compa ed o
ha o es inge s placed on mesa and ou side o i .
The e ec o sepa a ion dis ance was s udied al eady in
[8]. Resul s o simula ion did no gi e an explana ion
o la ge alues o ela i e e o o smalle dis ance
be ween ohmic pa hs.
4. Conclusions
In he pape , he echnological ac o s in luencing es
s uc u e ga e leng h we e desc ibed. The s anda d
de ia ion o inge s leng h was he smalles o in-
ge s pa s loca ed be ween he ohmic con ac s pa hs.
Also he smalles alue o ela i e e o was ob ained
o inge pa s placed be ween ohmic pa hs on op o
mesas independen ly o bo h alues o dis ance be-
ween ohmic con ac pa hs. The ela i e e o o es
s uc u es inge s leng hs indica ed a ec ion o ohmic
con ac pa hs dis ance on he leng hs. La ge alues
o ela i e e o we e obse ed o smalle dis ance be-
ween ohmic con ac s.
The simula ion esul s e eal g ea impac o Y-gap
p esence unde he mask on he elec ic ield dis i-
bu ion in he pho o esis . The smalles alue o ela-
i e e o o inge pa s placed be ween ohmic pa hs
on op o mesas could be a esul o occu ence o he
smalles Y-gap compa ed o ha o es inge s placed
on and ou side o mesa. Resul s o simula ion did no
b ing any explana ion o la ge alues o ela i e e o
o smalle dis ance be ween ohmic pa hs.
Acknowledgmen
This wo k was co- inanced by he Eu opean Union
wi hin Eu opean Regional De elopmen Fund, h ough
g an Inno a i e Economy (POIG.01.01.02-00-008/08-
05), by Na ional Cen e o Resea ch and De el-
opmen h ough Applied Resea ch P og am g an
no. 178782, p og am LIDER no. 027/533/L-
5/13/NCBR/2014, Na ional Cen e o Science un-
de he g an s no. 2015/19/B/ST7/02494 and DEC-
2012/07/D/ST7/02583, by W oclaw Uni e si y o Sci-
ence and Technology s a u o y g an s and Slo ak-
Polish In e na ional Coope a ion P og am.
Re e ences
[1] OKAZAKI, S. Resolu ion limi s o op ical li hog-
aphy. Jou nal o Vacuum Science &Technology
B, Nano echnology and Mic oelec onics: Ma e-
ials, P ocessing, Measu emen , and Phenomena.
1991, ol. 9, iss. 1, pp. 2829–2833. ISSN 2166-2746.
DOI: 10.1116/1.585650.
[2] HARRIOTT, L. R. Limi s o li hog aphy. P oceed-
ings o he IEEE. 2001, ol. 89, iss. 3, pp. 366–374.
ISSN 1558-2256. DOI: 10.1109/5.915379.
[3] ITO, T. and S. OKAZAKI. Limi s o li hog-
aphy. Pushing he limi s o li hog aphy. 2000,
ol. 409, iss. 1, pp. 1027–1031. ISSN 0028-0836.
DOI: 10.1038/35023233.
[4] FANG, N., H. LEE, C. SUN and X. ZHANG. Sub-
Di ac ion-Limi ed Op ical Imaging wi h a Sil-
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 363
APPLIED PHYSICS VOLUME: 15 |NUMBER: 2 |2017 |JUNE
e Supe lens. Science. 2000, ol. 308, iss. 5721,
pp. 534–537. ISSN 1095-9203. DOI: 10.1126/sci-
ence.1108759.
[5] MOJARAD, N., J. GOBRECHT and Y. EKINCI.
Beyond EUV li hog aphy: a compa a i e s udy o
e icien pho o esis s pe o mance. Scien i ic Re-
po s. 2015, ol. 5, no. 9235, pp. 1–5. ISSN 2045-
2322. DOI: 10.1038/s ep09235.
[6] LUO, X. and T. ISHIHARA. Subwa eleng h
pho oli hog aphy based on su ace-plasmon
pola i on esonance. Op ics Exp ess. 2004,
ol. 12, iss. 14, pp. 3055–3065. ISSN 1094-4087.
DOI: 10.1364/OPEX.12.003055.
[7] LEUNISSEN, L. H. A., R. JONCKHEERE,
U. HOFMANN, N. UNAL and C. KALUS.
Expe imen al and simula ion compa ison o
elec on-beam p oximi y co ec ion. Jou nal o
Vacuum Science and Technology B: Mic o-
elec onics and Nanome e S uc u es. 2004,
ol. 22, iss. 6, pp. 2943–2947. ISSN 1071-1023.
DOI: 10.1116/1.1808742.
[8] WOSKO, M., B. PASZKIEWICZ, A. VINCZE, T.
SZYMANSKI and R. PASZKIEWICZ. GaN/AlN
supe la ice high elec on mobili y ansis o he -
e os uc u es on GaN/Si(111). Physica S a us So-
lidi (B). 2015, ol. 252, no. 5, pp. 1195–1200.
ISSN 1071-1023. DOI: 10.1002/pssb.201451596.
[9] BAEK, S., G. KANG, M. KANG, C.-W. LEE and
K. KIM. Resolu ion enhancemen using plasmonic
me amask o wa e -scale pho oli hog aphy in he
a ield. Scien i ic Repo s. 2016, ol. 6, no. 30476,
pp. 1–8. ISSN 1071-1023. DOI: 10.1038/s ep30476.
Abou Au ho s
Joanna PRAZMOWSKA ecei ed he M.Sc.
deg ee in Elec onics om W oclaw Uni e si y o
Technology, Poland in 2005 and Ph.D. deg ee om
W oclaw Uni e si y o Technology (W UT) in 2011.
Now she is assis an p o esso a W UT. He esea ch
in e es emb aces echnology o semiconduc o de ices
i.e. li hog aphy p ocess de elopmen o elec onic,
op oelec onic de ices as well as gas senso s.
Ko nelia INDYKIEWICZ bo n in Wy zysk,
Poland in 1986, is a Ph.D. s uden a W oclaw Uni-
e si y o Technology, Depa men o Mic oelec onics
and Nano echnology. She wo ks in a mul idisciplina y
eam doing esea ch conce ning AlGaN/GaN based
ansis o s.
Bogdan PASZKIEWICZ ecei ed his M.Sc.
deg ee in Elec ical Enginee ing om S . Pe e sbu g
Elec o echnical Uni e si y, S . Pe e sbu g, Russia in
1979 and Ph.D. deg ee om he W oclaw Uni e si y
o Technology in 1997. Now he is assis an p o esso
a W UT. His esea ch is ocused on he design
and pa ame e e alua ion o ni ides-based de ices:
HEMTs and senso s.
Regina PASZKIEWICZ ecei ed he M.Sc.
deg ee in Elec ical Enginee ing om S . Pe e sbu g
Elec o echnical Uni e si y, S . Pe e sbu g, Russia in
1982 and Ph.D. deg ee om he W oclaw Uni e si y
o Technology in 1997. Now she is ull p o esso a
W UT. He esea ch is ocused on he echnology
o (Ga, Al, In) N semiconduc o s, mic owa e and
op oelec onic de ices echnological p ocesses de elop-
men .
c
2017 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 364