Direct Synthesis of Photosensitive Poly(benzoxazole)
Kazuya Ebara, Yuji Shibasaki, and Mitsuru Ueda*
Department of Organic and Polymeric Materials
Graduate School of Science and Engineering, Tokyo Institute of Technology,
2-12-1 O-okayama, Meguro-ku, Tokyo 152-8552, Japan
Poly(o-hydroxy-amide) (3) as a poly(benzoxazole) precursor was prepared by the
polycondensation of 2,2-bis(3-amino-4-hydroxyphenyl)hexafluoropropane (1) and
N,N`-isophtaloylbis(benzoxazoline-2-thione) (2) in N-methylpyrrolidinone. The resist
from the polymerization solution without isolation of 3 was formulated with
diazonaphtoquinone. A clear positive pattern with 8 m line and space could be
obtained when the film was exposed to g-line (436 nm), followed by developing with a
0.8 wt% aqueous tetramethylammonium hydroxide (TMAH) solution, where the
sensitivity and contrast were determined as 80 mJ/cm 2 and 2.2 respectively. The
positive image was successfully converted to poly(benzoxazole) pattern by thermal
treatment.
Keywords: photoresist, direct synthesis, poly(o-hydroxy-amide), poly(benzoxazole)
1. Introduction
Photosensitive polyimides (PSPIs) are
widely used as protection and insulation
layers of very large scale intergrated circuit,
multichip
modules
for
computer,
telecommunications
and
thermal
heads
because they simplify processing and do not
need a photoresist used in the microlithograpy
or a toxic etchant such as hydrazine [1].
Most of the standard resist materials
used in semiconductor manufacturing
are positive type resists based on
novolac resins with diazonaphtoquinone
(DNQ), where DNQ acts as a dissolution
inhibitor for aqueous base development
of
novolac
resins.
Therefore,
positive-working PSPIs that can be
developed with an aqueous base solution
have been attracting great interest.
A conventional polyimide precursor,
poly(amic acid) (PAA) is very soluble in an
aqueous alkaline solution, and the dissolution
rate is essentially too high to get a sufficient
dissolution contrast. Thus, the dissolution rate
of PAA has to be reduced by prebaking or
post exposure bake. These methods, however,
don’t afford a reliable dissolution contrast. To
overcome this problem, polyimide containing
hydroxy groups and DNQ, or o-nitrobenzyl
ester of polyamic acid have been reported
[2,3,4].
Poly(o-hydroxy-amide)
is
a
poly(benzoxazole)
precursor,
attracting
increasing
attention
as
an
alkaline
developable positive -type photosensitive
polymer because of its appropriate dissolution
rate toward a 2.38 wt% tetramethylammonium
hydroxide (TMAH) aqueous solution, and the
disappearance of phenol moiety after
benzoxazole ring formation.
A chlorine free synthetic method using a
condensing agent is generally applied to
prepare poly(o-hydroxy-amide) from aromatic
dicarboxylic acid and bis(o-aminophenol),
and polymer is isolated by precipitation from
a polymerization solution [5-9].
For the
resist formulation, the polymer should be
again dissolved in a solvent. This procedure is
troublesome. Therefore, it is of interest to
develop a more simple procedure, that is, the
resist formulation from the polymerization
solution
without
isolation
of
poly(o-hydroxy-amide). We now report a
direct
resist
formulation
method
for
photosensitive polybenzoxazole.
2. Experimental
2.1. Materials
Acetone was purified by distillation and
stored
over
4-A
molecular
sieves.
N-Methyl-2-pyrrolidinone
(NMP)
was
purified by vacuum distillation from calcium
hydride and stored over 4-A molecular sieves.
2,2-bis(3-amino-4-hydroxyphenyl)hexafluoro
-propane (1) was purified by recrystallization
from hexane / tetrahydrofuran. Other reagents
including
1-{1,1-bis[4-(2-diazo-1(2H)naphthalenone-5sulfonyloxy)phenyl]-ethyl}-4-{1-[4-(2-diazo1(2H)naphthalenone-5-sulfonyloxy)phenyl]m
ethylethyl}benzene (S-DNQ) were obtained
commercially and used as received.
N,N’-Isophtaloylbis(benzoxazoline -2-thione)
(2)
To solution of 2-mercaptobenzoxazole
(MB) (0.906 g, 12 mmol) and triethylamine
(TEA) (2 mL, 14 mmol) in acetone (10 mL),
was dropwise a solution of isophtaloyl
chloride (1.22 g, 6 mmol) in acetone (10 mL).
After 1 h the mixture was poured into water.
The precipitate was collected, washed with
water, and dried in vacuo over P 2 O 5 . It was
recrystallized from dichloromethane to give
white-yellow needles. Yield 1.45 g (56 %)
mp: 198 ℃ (lit. 201-202 ℃)[10]: IR (KBr)
ν 1720 (C=O), 1604 (Ar), 1355 cm -1 (C=S);
1
H-NMR(CDCl 3 ) σ 8.38 (t, J = 1.8 Hz, 1H),
8.15 (dd, J = 7.7 and 1.8 Hz, 2H), 7.68 (t, J =
7.7 Hz, 1H), 7.59-7.58 (m, 2H), 7.41-7.33 (m,
6H); 13 C-NMR(CDCl 3 ) σ179.0, 167.3, 147.4,
135.4, 132.7, 132.0, 130.2, 129.3, 126.2,
125.7, 113.6, 110.4.
2.2. Synthesis of poly(o-hydroxy-amide) (3)
from 1 and 2
To a solution of 2 (0.648 g, 1.5 mmol) in
NMP (2.7 mL) was added 1 (0.549 g, 1.5
mmol). The solution was stirred at room
temperature for 3 h, and poured into 3w%
sodium carbonate aqueous solution (150 mL)
to precipitate the polymer. The obtained
polymer was washed with water, dried in
vacuo over P 2 O 5 . The Yield was 0.700 g
(94%): IR (KBr): ν 3300 (OH), 1654 (C=O),
1608
(Ar)
and
1253
cm -1
(CF 3 );
１
H-NMR(DMSO-d 6 ): σ 10.0 (brs, 1H),
9.62 (s, 1H), 8.51 (s, 1H), 8.12 (d, J = 8.0
Hz, 2H), 7.95 (s, 1H), 7.64 (t, J = 8.0 Hz, 1H),
7.01 (brs, 4H)
2.3. Dissolution Rate
To a polymerization solution was directly
added S-DNQ (30 wt% of polymer 3). Films
spin-cast on silicone wafer were pre-baked
at115 ℃ for 3 min and exposed through a
filtered super-high pressure mercury lamp
SH-200. Imagewise exposure was carried out
in a contact mode. Exposed films were
subjected to developing with a 0.8 wt%
TMAH aqueous solution at room temperature.
2.4. Photosensitivity
A 1  m thick film obtained from a
polymerization solution and S-DNQ (30 wt%
of polymer 3) on silicone wafer was exposed
to 436 nm light through the filtered
super-high-pressure mercury lamp. The
characteristic curve was obtained by
normalized film thickness against exposed
energy.
2.5. Measurement
Infrared spectra were recorded on a Horiba
FT-720 spectrometer. 1 H and 13 C-NMR
spectra were obtained on a Bruker DPX-300
spectrometer.
Molecular
weights
were
determined
by
a
gel
permeation
chromatography (GPC) with polystyrene
calibration using a HITACHI D-7100-G
equipped with column in THF at 40 ℃ .
Thermal analyses were performed on a Seiko
EXSTAR 6000-TG / DTA6300 thermal
analyzer at a heating rate of 10 ℃・min -1 for
thermogravimetric (TG) under nitrogen.
3. Results and Discussion
3.1. Synthesis of poly(o-hydroxy-amide) (3)
In a previous paper [10], we reported that
active diamides derived from MB reacted
with aromatic diamines at room temperature,
producing polyamides with high molecular
weights. Based on this information, the
synthesis of polymer 3 was carried out by the
addition of 1 into 2 solution in NMP. The
polycondensation proceed smoothly, giving 3
in 94 % yield.
O
O
O
S
S
+ HO
O
in Figure 2.
NH2
CF3
N
N
OH
NMP, r.t. 3 h
CF3
H2N
1
2
O
HO
O
CF3
N
H
CF3
OH
N
H
+ 2 HN
n
S
O
MB
3
The structure of polymer 3 was confirmed
as a desired poly(o-hydroxy-amide) by IR and
1
H-NMR spectroscopies. The IR spectrum of
polymer 3 showed characteristic O-H,
carbonyl of amide group, C=C, and CF 3 bands
at, 3300, 1654, 1608, and 1253 cm -1 ,
respectively. The 1 H-NMR spectrum of
polymer 3 is presented in Figure 1.
O
O
c
e
N
H
HO
d
f
CF3
N
Hb
OH n
a
CF3
g
g
b
c
de
f
a
10.0
9.5
9.0
8.5
8.0
7.5
7.0
6.5
ppm
1
Fig. 1. H-NMR spectrum of polymer 3 in
DMSO-d 6 .
All peaks were well assigned to the
expected polymer structure. The molecular
weight of polymer 3 was estimated (relative
to a polystyrene standard) by gel permeation
chromatography (GPC) in THF. The
chromatogram of polymer 3 was a unimodal
distribution with a polydispersity of 3.9.
The weight-average molecular weight (M w )
was determined as 21,000.
3.2. Lithographic Evaluation
The leaving group of active amide, MB in a
polymerization solution does not have
absorption above 330 nm. In fact, the
UV-visible spectrum of a 1  m thick film
obtained directly from a polymerization
solution is a similar to that of a film obtained
from polymer 3, and showed an excellent
transparency above 400 nm. Thus, S-DNQ
having a strong absorption in the range of
320-440 nm was selected as a photoactive
compound. The UV-visible spectra of the film
casted from the polymerization solution
containing 30 wt% of S-DNQ and its changes
upon irradiation with 436 nm light are shown
Fig. 2. UV-visible spectral changes of polymer 3
containing 30 wt% S-DNQ.
The characteristic absorption band at
around 425 nm decreased rapidly with
increasing with an irradiation dose.
The dissolution rate of resists depends on a
concentration of TMAH aqueous solution and
a S-DNQ loading. The dissolution rate was
estimated by measuring the film thickness
after development. Polymer 3 film was easily
soluble in a 2.38 wt% of TMAH aqueous
solution even in the presence of S-DNQ.
After a preliminary optimization study of the
concentration of TMAH aqueous solution, we
decided to use the 0.8 wt% of TMAH aqueous
solution.
The dissolution rates of exposed and
unexposed areas in the presence of the
various DNQ contents were studied. The
result presented in Figure 3 shows that the
dissolution rate of the unexposed area
decreased with increasing the S-DNQ content,
and reached to zero under 30 wt% loading of
S-DNQ. On the other hand, the solubility of
exposed area was almost unchanged even
increasing the S-DNQ content. Thus, the
suitable S-DNQ content was determined to be
30 wt% of polymer 3.
Fig. 3. Relationship between S-DNQ content and
dissolution rate.
Based on these optimization studies, we
formulated a photosensitive polymer system
consisting of polymer 3 and 30 wt% of
S-DNQ. The sensitivity curve for 2.7  m
thick resist (Figure 4) indicated that the
sensitivity (D o ) and contrast (r o ) were 80 mJ/
cm 2 and 2.2 with 436 nm light, respectively.
Fig. 4. Exposure characteristic curve for polymer
3 containing 30 wt% S-DNQ.
Figure 5 presents SEM micrograph of the
contact-printed image obtained using the
resist system composed of polymer 3 (70
wt%) with S-DNQ (30 wt%), after exposure
to 200 mJ/cm2 , and developed with the 0.8
wt% TMAH aqueous solution at room
temperature. The clear positive pattern with
8  m line and space was obtained.
3.3. Thermal treatment of the resist system
Thermal behavior of the resist system
obtained directly from the polymerization
solution and S-DNQ was studied by TG
(Figure 6). The weight losses were observed
from 150 to 330 ℃ due to decomposition of
S-DNQ and MB, and elimination of water by
formation of oxazole ring. The transformation
was also confirmed by the appearance of an
absorption band at 1627 cm-1 characteristic of
an oxazole ring, and disappearing at 3300 and
1654 cm -1 characteristic of hydroxy and amide
groups, respectively.
Fig. 6. TG curve of polymer 3 obtained directly
from polymerization solution and S-DNQ.
Thus, the isothermal treatment of this
resist system was carried out at 350 °C for 30
min under nitrogen.
Figure 7 shows TG curves of i) a
conventional poly(benzoxazole) ( PBO) and,
ii) the PBO obtained by the thermal treatment
of the resist system. Both TG curves are same,
indicating the satisfactory formation of PBO
accompanied with quantitative decomposition
of MB and S-DNQ.
The printed film was heated at 150 ℃ for
20 min, 250 ℃ for 30 min, then 350 ℃ for
30 min to convert into the PBO film. Figure 8
shows a SEM micrograph of the PBO film.
The positive image in the resist remained
unchanged.
Fig. 7. TG curves of PBOs.
Fig. 5. SEM micrograph of pattern from the
system consisting of polymer 3 and S-DNQ.
This direct resist formulation method is
advantageous compared to the conventional
two step one.
Fig. 8. SEM image of thermally treated pattern.
4. Conclusion
Poly(o-hydroxy-amide)
3
as
a
poly(benzoxazole) precursor was successfully
prepared by the polycondensation of 1 and 2.
The resist system was directly formulated
from the polymerization solution with
S-DNQ.
This resist showed a high sensitivity 80
mJ/cm 2 and a contrast of 2.2 with 436 nm
light. The positive image obtained was
converted to the corresponding PBO pattern
by thermal treatment.
References
1. T. Omote, “Polyimides”, M. Ghosh, and K.
L. Miital, Eds., Dekker, New York, 1996,
p.121.
2. S. Hayase, K. Takano, Y. Mikogami, and Y.
Nakano, J. Electronchem. Soc., 138
(1979) 3625.
3. T. Omote, K. Koseki, and T. Yamaoka,
Macromolecules, 23 (1990) 4788.
4. T. Omote, S. Hayashi, K. Ishii, K. Naito,
and T. Yamaoka, Adv. Technol., 4 (1993)
277.
5. D. N. Khanna and W. H. Mueller, Polym.
Eng. Sci, 29 (1989) 954.
6. H. Ahne, R. Rubner, and E. Kuhn,
EP0023626, to Siemens, 1980.
7. R. Rubner, Adv. Mater, 2 (1990) 452.
8. T. Yamaoka, N. Nakajima, K. Koseki, and
Y. Maruyama, J. Polym. Sci. Polym.
Chem., 28 (1990) 2517.
9. H. Ahne, E. Kuhn, and R. Ruber. U.S.
patent No. 4,395,482 (1985).
10. M. Ueda, K. Seki, and Y. Imai,
Macromolecules, 15 (1982) 17.