<s>
Next-generation	B-Algorithm
lithography	I-Algorithm
or	O
NGL	O
is	O
a	O
term	O
used	O
in	O
integrated	O
circuit	O
manufacturing	O
to	O
describe	O
the	O
lithography	B-Algorithm
technologies	O
in	O
development	O
which	O
are	O
intended	O
to	O
replace	O
current	O
techniques	O
.	O
</s>
<s>
The	O
term	O
applies	O
to	O
any	O
lithography	B-Algorithm
method	O
which	O
uses	O
a	O
shorter-wavelength	O
light	O
or	O
beam	O
type	O
than	O
the	O
current	O
state	O
of	O
the	O
art	O
,	O
such	O
as	O
X-ray	B-Algorithm
lithography	I-Algorithm
,	O
electron	B-Architecture
beam	I-Architecture
lithography	I-Architecture
,	O
focused	O
ion	O
beam	O
lithography	B-Algorithm
,	O
and	O
nanoimprint	B-Algorithm
lithography	I-Algorithm
.	O
</s>
<s>
Many	O
technologies	O
once	O
termed	O
"	O
next	O
generation	O
"	O
have	O
entered	O
commercial	O
production	O
,	O
and	O
open-air	O
photolithography	B-Algorithm
,	O
with	O
visible	O
light	O
projected	O
through	O
hand-drawn	O
photomasks	B-Algorithm
,	O
has	O
gradually	O
progressed	O
to	O
deep-UV	O
immersion	B-Algorithm
lithography	I-Algorithm
using	O
optical	B-Algorithm
proximity	I-Algorithm
correction	I-Algorithm
,	O
inverse	O
lithography	B-Algorithm
technology	O
,	O
off-axis	B-Algorithm
illumination	I-Algorithm
,	O
phase-shift	B-Algorithm
masks	I-Algorithm
,	O
double	B-Algorithm
patterning	I-Algorithm
,	O
and	O
multiple	B-Algorithm
patterning	I-Algorithm
.	O
</s>
<s>
In	O
the	O
late	O
2010s	O
,	O
the	O
combination	O
of	O
many	O
such	O
techniques	O
was	O
able	O
to	O
achieve	O
features	O
on	O
the	O
order	O
of	O
20nm	O
with	O
the	O
193	O
nm-wavelength	O
ArF	O
excimer	O
laser	O
in	O
the	O
14	B-Algorithm
nm	I-Algorithm
,	O
10	B-Algorithm
nm	I-Algorithm
and	O
7	B-Algorithm
nm	I-Algorithm
processes	O
,	O
though	O
at	O
the	O
cost	O
of	O
adding	O
processing	O
steps	O
and	O
therefore	O
cost	O
.	O
</s>
<s>
13.5	O
nm	O
extreme	O
ultraviolet	O
(	O
EUV	O
)	O
lithography	B-Algorithm
,	O
long	O
considered	O
a	O
leading	O
candidate	O
for	O
next-generation	B-Algorithm
lithography	I-Algorithm
,	O
began	O
to	O
enter	O
commercial	O
mass-production	O
in	O
2018	O
.	O
</s>
<s>
As	O
of	O
2021	O
,	O
Samsung	O
and	O
TSMC	O
were	O
gradually	O
phasing	O
EUV	B-Algorithm
lithography	I-Algorithm
into	O
their	O
production	O
lines	O
,	O
as	O
it	O
became	O
economical	O
to	O
replace	O
multiple	O
processing	O
steps	O
with	O
single	O
EUV	O
steps	O
.	O
</s>
<s>
As	O
of	O
the	O
early	O
2020s	O
,	O
many	O
EUV	O
techniques	O
are	O
still	O
in	O
development	O
and	O
many	O
challenges	O
remain	O
to	O
be	O
solved	O
,	O
positioning	O
EUV	B-Algorithm
lithography	I-Algorithm
as	O
being	O
in	O
transition	O
from	O
"	O
next	O
generation	O
"	O
to	O
"	O
state	O
of	O
the	O
art.	O
"	O
</s>
<s>
Candidates	O
for	O
next-generation	B-Algorithm
lithography	I-Algorithm
beyond	O
EUV	O
include	O
X-ray	B-Algorithm
lithography	I-Algorithm
,	O
electron	B-Architecture
beam	I-Architecture
lithography	I-Architecture
,	O
focused	O
ion	O
beam	O
lithography	B-Algorithm
,	O
nanoimprint	B-Algorithm
lithography	I-Algorithm
,	O
and	O
quantum	B-Algorithm
lithography	I-Algorithm
.	O
</s>
<s>
Several	O
of	O
these	O
technologies	O
have	O
experienced	O
periods	O
of	O
popularity	O
,	O
but	O
have	O
remained	O
outcompeted	O
by	O
the	O
continuing	O
improvements	O
in	O
photolithography	B-Algorithm
.	O
</s>
<s>
Electron	B-Architecture
beam	I-Architecture
lithography	I-Architecture
was	O
most	O
popular	O
during	O
the	O
1970s	O
,	O
but	O
was	O
replaced	O
in	O
popularity	O
by	O
X-ray	B-Algorithm
lithography	I-Algorithm
during	O
the	O
1980s	O
and	O
early	O
1990s	O
,	O
and	O
then	O
by	O
EUV	B-Algorithm
lithography	I-Algorithm
from	O
the	O
mid-1990s	O
to	O
the	O
mid-2000s	O
.	O
</s>
<s>
Focused	O
ion	O
beam	O
lithography	B-Algorithm
has	O
carved	O
a	O
niche	O
for	O
itself	O
in	O
the	O
area	O
of	O
defect	O
repair	O
.	O
</s>
<s>
Nanoimprint	B-Algorithm
's	O
popularity	O
is	O
rising	O
,	O
and	O
is	O
positioned	O
to	O
succeed	O
EUV	O
as	O
the	O
most	O
popular	O
choice	O
for	O
next-generation	B-Algorithm
lithography	I-Algorithm
,	O
due	O
to	O
its	O
inherent	O
simplicity	O
and	O
low	O
cost	O
of	O
operation	O
as	O
well	O
as	O
its	O
success	O
in	O
the	O
LED	O
,	O
hard	B-Device
disk	I-Device
drive	I-Device
and	O
microfluidics	O
sectors	O
.	O
</s>
<s>
Electron	O
beam	O
and	O
nanoimprint	B-Algorithm
lithography	I-Algorithm
are	O
limited	O
mainly	O
by	O
the	O
throughput	O
,	O
while	O
EUV	O
and	O
X-ray	B-Algorithm
lithography	I-Algorithm
are	O
limited	O
by	O
implementation	O
and	O
operation	O
costs	O
.	O
</s>
<s>
+Hypothetical	O
NGL	O
case	O
@	O
5	O
nmNodeLeading	O
chipmakerLagging	O
chipmakerNo	O
changeWith	O
NGL180nmKrFKrF	O
-	O
130nmKrFKrF90nmArFArF65nmArFArF45/40nmArF	O
immersionArF	O
immersion32/28nmArF	O
immersionArF	O
immersion22/20nmArF	O
immersion	O
,	O
double	B-Algorithm
patterning	I-Algorithm
?	O
</s>
<s>
Skipped	O
multipatterning	B-Algorithm
costs16/14nmArF	O
immersion	O
,	O
double	O
patterning10nmArF	O
immersion	O
,	O
SADF/triple	O
patterning7nmArF	O
immersion	O
,	O
SADF/SAQP5nmSAQP	O
+	O
additional	O
lithographyNGLThe	O
difficulty	O
of	O
extending	O
optical	B-Algorithm
lithography	I-Algorithm
has	O
been	O
the	O
main	O
selling	O
point	O
of	O
NGL	O
.	O
</s>
<s>
However	O
,	O
a	O
leading	O
chipmaker	O
would	O
benefit	O
significantly	O
less	O
than	O
a	O
lagging	O
chipmaker	O
,	O
due	O
to	O
the	O
huge	O
additional	O
investments	O
in	O
extending	O
optical	B-Algorithm
lithography	I-Algorithm
up	O
to	O
its	O
current	O
state	O
.	O
</s>
<s>
Regardless	O
of	O
whether	O
NGL	O
or	O
photolithography	B-Algorithm
is	O
used	O
,	O
etching	O
of	O
polymer	O
(	O
resist	O
)	O
is	O
the	O
last	O
step	O
.	O
</s>
<s>
Ultimately	O
the	O
quality	O
(	O
roughness	O
)	O
as	O
well	O
as	O
resolution	O
of	O
this	O
polymer	O
etching	O
limits	O
the	O
inherent	O
resolution	O
of	O
the	O
lithography	B-Algorithm
technique	O
.	O
</s>
<s>
Next	O
generation	O
lithography	B-Algorithm
also	O
generally	O
makes	O
use	O
of	O
ionizing	O
radiation	O
,	O
leading	O
to	O
secondary	O
electrons	O
which	O
can	O
limit	O
resolution	O
to	O
effectively	O
>	O
20nm	O
.	O
</s>
<s>
The	O
above-mentioned	O
competition	O
between	O
NGL	O
and	O
the	O
recurring	O
extension	O
of	O
photolithography	B-Algorithm
,	O
where	O
the	O
latter	O
consistently	O
wins	O
,	O
may	O
be	O
more	O
a	O
strategic	O
than	O
a	O
technical	O
matter	O
.	O
</s>
<s>
If	O
a	O
highly	O
scalable	O
NGL	O
technology	O
were	O
to	O
become	O
readily	O
available	O
,	O
late	O
adopters	O
of	O
leading-edge	O
technology	O
would	O
immediately	O
have	O
the	O
opportunity	O
to	O
leapfrog	O
the	O
current	O
use	O
of	O
advanced	O
but	O
costly	O
photolithography	B-Algorithm
techniques	O
,	O
at	O
the	O
expense	O
of	O
the	O
early	O
adopters	O
of	O
leading-edge	O
technology	O
,	O
who	O
have	O
been	O
the	O
key	O
investors	O
in	O
NGL	O
.	O
</s>
<s>
Suppose	O
company	O
A	O
manufactures	O
down	O
to	O
28nm	O
,	O
while	O
company	O
B	O
manufactures	O
down	O
to	O
7nm	B-Algorithm
,	O
by	O
extending	O
its	O
photolithography	B-Algorithm
capability	O
by	O
implementing	O
double	B-Algorithm
patterning	I-Algorithm
.	O
</s>
<s>
If	O
an	O
NGL	O
were	O
deployed	O
for	O
the	O
5nm	B-Algorithm
node	O
,	O
both	O
companies	O
would	O
benefit	O
,	O
but	O
company	O
A	O
currently	O
manufacturing	O
at	O
the	O
28nm	O
node	O
would	O
benefit	O
much	O
more	O
because	O
it	O
would	O
immediately	O
be	O
able	O
to	O
use	O
the	O
NGL	O
for	O
manufacturing	O
at	O
all	O
design	O
rules	O
from	O
22nm	B-Algorithm
down	O
to	O
7nm	B-Algorithm
(	O
skipping	O
all	O
the	O
said	O
multiple	B-Algorithm
patterning	I-Algorithm
)	O
,	O
while	O
company	O
B	O
would	O
only	O
benefit	O
starting	O
at	O
the	O
5nm	B-Algorithm
node	O
,	O
having	O
already	O
spent	O
much	O
on	O
extending	O
photolithography	B-Algorithm
from	O
its	O
22nm	B-Algorithm
process	O
down	O
to	O
7nm	B-Algorithm
.	O
</s>
<s>
The	O
gap	O
between	O
Company	O
B	O
,	O
whose	O
customers	O
expect	O
it	O
to	O
advance	O
the	O
leading	O
edge	O
,	O
and	O
Company	O
A	O
,	O
whose	O
customers	O
do	O
n't	O
expect	O
an	O
equally	O
aggressive	O
roadmap	O
,	O
will	O
continue	O
to	O
widen	O
as	O
NGL	O
is	O
delayed	O
and	O
photolithography	B-Algorithm
is	O
extended	O
at	O
greater	O
and	O
greater	O
cost	O
,	O
making	O
the	O
deployment	O
of	O
NGL	O
less	O
and	O
less	O
attractive	O
strategically	O
for	O
Company	O
B	O
.	O
</s>
<s>
This	O
becomes	O
more	O
clear	O
when	O
considering	O
that	O
each	O
resolution	O
enhancement	O
technique	O
applied	O
to	O
photolithography	B-Algorithm
generally	O
extends	O
the	O
capability	O
by	O
only	O
one	O
or	O
two	O
generations	O
.	O
</s>
<s>
For	O
this	O
reason	O
,	O
the	O
observation	O
that	O
"	O
optical	B-Algorithm
lithography	I-Algorithm
will	O
live	O
forever	O
"	O
will	O
likely	O
hold	O
,	O
as	O
the	O
early	O
adopters	O
of	O
leading-edge	O
technology	O
will	O
never	O
benefit	O
from	O
highly	O
scalable	O
lithography	B-Algorithm
technologies	O
in	O
a	O
competitive	O
environment	O
.	O
</s>
<s>
There	O
is	O
therefore	O
great	O
pressure	O
to	O
deploy	O
an	O
NGL	O
as	O
soon	O
as	O
possible	O
,	O
but	O
the	O
NGL	O
ultimately	O
may	O
be	O
realized	O
in	O
the	O
form	O
of	O
photolithography	B-Algorithm
with	O
more	O
efficient	O
multiple	B-Algorithm
patterning	I-Algorithm
,	O
such	O
as	O
directed	O
self-assembly	O
or	O
aggressive	O
cut	O
reduction	O
.	O
</s>
<s>
mask	O
defect	O
size	O
Throughput	O
Issues	O
193	O
nm	O
1.35	O
NA	O
(	O
current	O
)	O
40	O
nm	O
34	O
nm	O
80	O
nm	O
130	O
WPH	O
water	O
immersion	O
193	O
nm	O
1.7	O
NA	O
30	O
nm	O
34	O
nm	O
60	O
nm	O
development	O
stopped	O
toxic	O
high-index	O
materials	O
needed	O
(	O
prohibitive	O
)	O
157	O
nm	O
1.7	O
NA	O
25	O
nm	O
24	O
nm	O
50	O
nm	O
development	O
stopped	O
lack	O
of	O
materials	O
;	O
CaF2	O
required	O
to	O
replace	O
fused	O
silica	O
(	O
prohibitive	O
)	O
EETimes	O
:	O
157	O
nm	O
dropped	O
from	O
Intel	O
's	O
roadmap	O
13.5	O
nm	O
0.25	O
NA	O
(	O
EUVL	B-Algorithm
)	O
~	O
30	O
nm	O
(	O
secondary	O
electrons	O
)	O
R	O
.	O
Feder	O
et	O
al.	O
,	O
J	O
.	O
Vac	O
.	O
</s>
<s>
0.4	O
nm	O
(	O
prohibitive	O
)	O
40	O
nm	O
4	O
WPH	O
(	O
prohibitive	O
)	O
ionization	O
;	O
shot	O
noise	O
;	O
throughput	O
;	O
mask	O
defects	O
X-ray	O
proximity	O
printing	O
~	O
30	O
nm	O
(	O
secondary	O
electrons	O
)	O
>	O
100	O
nm	O
depends	O
on	O
proximity	O
gap	O
;	O
matches	O
mask	O
(	O
prohibitive	O
)	O
comparable	O
to	O
optical	B-Algorithm
mask	I-Algorithm
membrane	O
(	O
prohibitive	O
)	O
;	O
source	O
(	O
possibly	O
prohibitive	O
)	O
Nanoimprint	B-Algorithm
N/A	O
(	O
matches	O
template	O
)	O
residual	O
layer	O
thickness	O
0	O
nm	O
(	O
prohibitive	O
)	O
>1	O
WPH	O
bubble	O
defects	O
(	O
possibly	O
prohibitive	O
)	O
;	O
H	O
.	O
Hiroshima	O
and	O
M	O
.	O
Komuro	O
,	O
Jpn	O
.	O
</s>
<s>
The	O
complexities	O
of	O
next-generation	B-Algorithm
lithography	I-Algorithm
development	O
have	O
always	O
spurred	O
the	O
pursuit	O
of	O
ways	O
of	O
extending	O
the	O
use	O
of	O
incumbent	O
lithographic	O
materials	O
,	O
light	O
sources	O
,	O
and	O
tools	O
.	O
</s>
