Preparation of polyimide-cellulose composite using oligoimide with ethynyl terminals

Article abstract of DOI:10.1246/cl.140072

A soluble imide oligomer (IO) (repeating number; 2 and 4) having N-phenylated melamine units and phenylethynyl terminals was prepared, and the application to polyimide composites with cellulose (CEL) crystals derived from pulp (CEL_p) or Halocyn

Full text of DOI:10.1246/cl.140072

Received: January 30, 2014 | Accepted: February 18, 2014 | Web Released: February 26, 2014
CL-140072
Preparation of Polyimide­Cellulose Composite Using Oligoimide with Ethynyl Terminals
Tomoya Shirata, Tatsuki Kon, Keiko Sasaki, Yoshiyuki Oishi, and Yuji Shibasaki*
Department of Chemistry and Bioengineering, Iwate University, 4-3-5 Ueda, Morioka, Iwate 020-8551
(
E-mail: yshibasa@iwate-u.ac.jp)
A soluble imide oligomer (IO) (repeating number; 2 and 4)
having N-phenylated melamine units and phenylethynyl termi-
nals was prepared, and the application to polyimide composites
with cellulose (CEL) crystals derived from pulp (CELp) or
Halocynthia (CEL ) with or without a modifier was investigated.
h
The CELh additive up to 0.5 wt % was easily dispersed in
IO matrix in solution, and thermostable (glass-transition tem-
perature, ca. 285 °C) transparent tough film (tensile strength,
ca. 169 MPa; elongation at break, ca. 8.9%; tensile modulus,
ca. 4.6 GPa) was obtained, while the addition of a modifier such
as cetyltrimethylammonium bromide and poly(ethylene glycol)
derivatives was necessary to obtain a similar tough film for the
CELp additive.
Figure 1. Fabrication of IO­CEL composites. The SEM images
show (a) pulp-derived CEL (CEL , 1000 nm scale bar) and (b)
p
Halocynthia-derived CEL (CELh, 100 nm scale bar).
Polyimides (PIs) have been widely adopted in a variety of
fields due to excellent thermal and mechanical properties.
They have also developed a highly soluble imide oligomer (IO)
based on 2-phenyl-4,4¤-diaminodiphenyl ether. The oligomer
shows excellent solubility and was thus applied to a composite
1
Generally speaking, the molding of this type of polymer is
difficult because of the high melting point and insolubility in
organic solvents arising from the rigid rod structures. Thus,
many kinds of processable PIs have been developed so far by
1
0
with carbon fiber. We have recently reported the synthesis and
properties of triazine-containing PIs, which have unusual high
solubility in organic solvents along with good mechanical
2
3
4
11
introducing amide, ether, and sulfide functions. These PIs,
however, have poor miscibility with carbon or glass fibers, so
that the addition of fillers to make higher performance PI
composites is relatively difficult. With this view, addition-type
PIs have been developed; they are easily processable in the form
of short chain oligomers end-capped with latent crosslinking
properties. Thus, we expected that the IO with a triazine
moiety would show adequate solubility in order to fabricate
imide composites. As for the additives to fabricate the
composite, we chose two types of cellulose crystals, CEL and
p
CELh, derived from pulp and Halocynthia, respectively. The
average size and shape of CELp are about 50 ¯m in plate-like
5
6
sites. The TRW’s PI resin (P-13N), that is an imide oligomer
end-capped with norbornene moiety, was first introduced.
crystals, and those of CEL are 20 nm wide and 1 mm long
h
nanofibers (Figure 1). Thus, we here report the synthesis of
soluble IO with N-phenylated melamine units and phenylethynyl
terminals, and the application to PI composites with CEL
crystals, which are common bio-macromolecules, and frequently
7
LARC-13 was then developed, which is a thermosetting resin
based on 4,4¤-diphthalic acid dianhydride, diamine, and norbor-
nene-5,6-dicarboxylic anhydride. The critical issues of these
resins are 1) the lower thermostability due to the aliphatic
end-group, 2) the need of high-pressure processing in order to
suppress retro Diels­Alder reaction, and 3) the use of high-
temperature aprotic solvents such as N-methylpyrrolidinone
1
2
applied as reinforcements for polyolefins. We also expected
that the hydroxy groups in CEL could strongly interact with
triazine units in IO to afford well-dispersed PI composites.
The phenylethynyl-terminated amic acid oligomer (AAO)
was prepared by conventional low-temperature polyaddition
(
NMP) to fabricate the polyimide composites. Polymerization
8
13
of monomeric reactant (PMR) has been developed, in which
low-temperature aprotic solvents such as methanol can be used
to immerse monomeric anhydrides and diamines into carbon or
glass fibers. The drawback of these PI composites is the need
of harsh process conditions and low toughness, which causes
abrasion and cracks on the surface of the composites. Yokota
and his co-workers developed a high-temperature, processable
of 2,4-bis(4-aminoanilino)-6-anilino-1,3,5-triazine (ATDA),
ODA, and BPDA in NMP at 10 °C for 1 h, and 60 °C for 4 h,
followed by reaction with 4-(phenylethynyl)phthalic anhydride
(PEPA) at 20 °C for 12 h (Scheme 1). The resulting AAO
solution was cast on a glass plate, and heated at elevated
temperature up to 250 °C for 1 h to give the corresponding IO.
Two types of IO samples, IO2 and IO4, were prepared by
changing the feed of anhydride and amine monomers of 2:3 and
4:5, respectively. Table 1 summarizes the results of the oligo-
merization. Both of the oligomers were successfully obtained in
high yields with comparable molecular weights (determined by
GPC) to the calculated values, which indicates the oligomers
have the expected composition of each monomer. The structures
thermosetting imide resin with good mechanical properties
9
(
TriA-PI), composed of asymmetric imide oligomers and
phenylethynyl terminals. The oligo(amic acid) of TriA-PI is
dissolved in aprotic solvent, in which carbon or glass fibers are
immersed. Thermal curing of this material gives high-perform-
ance PI composites, but careful process conditions must be taken
in order to obtain homogeneous composites free from an air-
voids arising from the water generated during the imidization.
of the IO (x = 2 and 4) were characterized by IR (Figure 2)
x
¹
1
and the elemental analysis. The absorption at 1680 cm
Chem. Lett. 2014, 43, 787–789 | doi:10.1246/cl.140072
© 2014 The Chemical Society of Japan | 787

O
O
O
H
N
H
N
O
O
O
N
O
O
O
N
N
BPDA
10°C, 1 h, 60°C, 4 h in NMP
PEPA
H2N
NH2
+
H2N
O
NH2
HN
r.t., 12 h
1) 100 °C, 1h
ATDA
ODA
2
) 150 °C, 1h
) 200 °C, 1h
3
O
H
N
H
O
O
H
H
N
O
4) 250 °C, 6h
Ar
H
N
N
Ar
H
H
O
H
O
O
H
H
O
N
N
N
O
O
O
O
x
N
N
AAOx
Ar =
HN
O
O
N
O
O
O
N
Ar
N
Ar
N
O
O
O
or
O
x
IOx
Scheme 1. Synthesis of the phenylethynyl-terminated amic acid
oligomer AAOx, and the corresponding IOx.
Table 1. Synthesis of IOx
Yielda
Mn © 10¹3
Feed/mol %
ATDA ODA BPDA PEPA /% Calcd Foundb
Sample
Mw/Mnb
Figure 3. IR spectra of (a) as-made IO, (b) hot-pressed IO, (c) hot-
pressed mixture of IO with CELh, and (d) hot-pressed mixture of IO
with CELp.
IO2
IO4
1.5
2.5
1.5
2.5
2
4
2
2
87
77
1.8
2.9
1.6
3.5
3.1
2.1
a
b
Methanol-insoluble part. Determined by GPC (NMP, PSt).
into the solution, and transferred onto a petri dish, followed by
the removal of the solvents in a vacuum oven at 250 °C for 12 h.
The resulting brown solid was crashed and ground well to afford
a IO­CEL mixture. This mixture was hot-pressed at elevated
temperature up to 370 °C for 1 h under 3 MPa to afford a brown
transparent composite film. The characterization of the compo-
sites was performed by IR spectroscopy, in which the charac-
teristic absorption corresponding to ethynyl stretching at
¹
1
2
210 cm almost disappeared after the hot-press process as
shown in Figures 3c­3f. The hot-pressed film does not show any
solubility in an organic solvent. These results indicated that the
ethynyl group at the terminus of the IO sample in the mixture
quantitatively reacted to give the composite with the crosslinked
structure. In order to characterize the structure of CEL crystals
in the composite after hot pressing, the CELp crystals were
hot-pressed at 370 °C for 1 h under 3 MPa pressure (exactly the
same conditions for making the composite film) between two
μ
polyimide (Upilex ) films. The white powdery CELp sample
turned into black powder with reducing the weight by 81% from
the original. In the IR spectrum of this sample (Figure 3b), the
Figure 2. IR spectra of cured amic acid oligomer (AAO) at the set
conditions.
¹
1
characteristic OH absorption at 3346 cm and aliphatic ether
¹
1
corresponding to the amide stretching observed in AAO samples
completely disappeared after heating at 250 °C for 1 h, and a new
absorption corresponding to imide carbonyl stretching was
absorption at 1060 cm (Figure 3a) almost disappeared. The
elemental analysis of the hot-pressed CEL sample shows the
p
content of C, H, and O of 73.63, 3.59, and 22.78%, which is
quite far from the original contents of 42.74, 6.17, and 51.09,
respectively. Therefore, the mixed CEL components should be
within the crosslinked imide matrix as the carbonized structure.
Table 2 shows the thermal properties of the composite films.
¹
1
observed at 1760 and 1710 cm , the intensity of which was
¹
1
saturated at 250 °C for 1 h. At 2210 cm , the terminal phenyl-
ethynyl absorption was observed, whose intensity was almost the
¹
1
same (vs. the intensity of ether absorption at 1235 cm ) during
the imidization reaction. These results supported the satisfactory
formation of the IOs with expected structure.
The glass-transition temperature (T ) of the crosslinked IO (X-
g
IO) by hot pressing are 282 and 281 °C for X-IO2 and X-IO4,
respectively. While the composites with the CELp additive show
similar Tg, CELh was effective to provide slightly higher Tg. The
small nanofibrous structure was able to disperse well in the
polyimide matrix, and thus the mobility of the polymer chain
should be restricted as depicted in Figure 1. The formation of
composites did not affect the thermal degradation process a lot,
but the char yields were somehow increased.
Generally speaking, IOs show poor solubility in organic
solvents, however, the prepared IO samples described above are
soluble in aprotic solvents such as NMP and DMAc; the
oligomers were completely soluble in the solvent at 0.2 wt % at
20 °C, and even at 30 wt % on heating. These results clearly
indicated that the anilinotriazine pendent should effectively
disturb the strong aggregation between aromatic imide groups,
resulting in excellent solubility.
The IO sample was then dissolved in NMP, and 0.5­2 wt %
CEL crystals (0.25­1 wt % to IO) dispersed in water was added
Table 3 summarizes the mechanical properties of the
composite films. The brittle property of X-IO2 film was
improved with the addition of CELh (Runs 1 and 10), while
788 | Chem. Lett. 2014, 43, 787–789 | doi:10.1246/cl.140072
© 2014 The Chemical Society of Japan

Table 2. Thermal properties of hot-pressed IOx and the composites
In conclusion, we successfully prepared an IO sample
bearing N-phenylated melamine units and phenylethynyl termi-
nals, and achieved excellent solubility in aprotic solvents. The
imidization was performed in a conventional thermal process ate
elevated temperature up to 250 °C. The composite formation was
performed by mixing IO with CEL in NMP/water medium, and
the dried solid was well ground in a mortar, and hot pressed at
370 °C for 1 h under 3 MPa pressure. The quantitative reaction
of the terminal ethynyl group was confirmed by IR, and the
introduced CEL was found to change into the carbonized state
by IR and elemental analysis. The composites show slightly
higher Tg (µ 285 °C), char yields (ca. 65%), and much better
mechanical properties (T µ 169 MPa, E µ 8.9%). When the
Cellulose Tga
/
Td5/°Cb
Td10/°Cb
Char/%
Run Polymer
wt % /°C in N2 in air in N2 in air (in N2 800 °C)
1
2
X-IO2
X-IO4
®
®
282 473 482 511 531
281 495 500 526 551
62
61
3
4
5
6
X-IO2-CELp
X-IO4-CELp
X-IO4-CELp
X-IO4-CELp
0.5
0.25
0.5
1
280 486 484 520 532
279 492 501 524 545
280 490 500 521 547
279 493 504 524 548
63
63
63
65
7
8
9
0
X-IO2-CELh
X-IO4-CELh
X-IO4-CELh
X-IO4-CELh
0.5
0.25
0.5
1
285 475 480 513 531
283 488 498 519 538
281 495 503 525 544
285 493 506 525 550
65
62
65
62
1
s
b
aDSC (N2) at a heating rate of 20 °C min¹1. b5% and 10% weight
two CEL additives were compared, CELh from Halocynthia
showed better results in view of the thermal and mechanical
properties. This can be explained by the size and the shape of the
nanofibrous CEL characters. CEL , a more common material,
loss temperatures by TGA in air or nitrogen at a heating rate of
¹
1
1
0 °C min .
h
p
can show similar improvement for IO resin when the additional
modifier is applied for composite formation.
1
4
Table 3. Mechanical properties of IOx and the composites
Cellulose
Tsa
Ebb
TMc
Run Polymer
Modifier
/wt %
/MPa /% /GPa
References and Notes
1
a) R. D. McCullough, R. D. Lowe, J. Chem. Soc., Chem. Commun.
992, 70. b) Polyimides: Fundamentals and Applications, ed. by M. K.
Ghosh, K. L. Mittal, Marcel Decker, New York, 1996. c) Polyimides,
ed. by D. Wilson, H. D. Stenzenberger, P. M. Hergenrother, Blackie,
New York, 1990. doi:10.1007/978-94-010-9661-4.
a) D.-J. Liaw, B.-Y. Liaw, Y.-S. Chen, Polymer 1999, 40, 4041.
b) H.-H. Wang, C.-C. Su, J. Appl. Polym. Sci. 1996, 61, 1087.
W. G. Kim, A. S. Hay, Macromolecules 1993, 26, 5275.
T. Takekoshi, Polym. J. 1987, 19, 191.
J. W. Connell, J. G. Smith, Jr., P. M. Hergenrother, J. Macromol. Sci.,
Rev. Macromol. Chem. Phys. 2000, C40, 207.
a) H. R. Lubowitz, U.S. Patent 3,528,950, 1970. b) R. W. Vaughan,
J. F. Jones, H. R. Lubowitz, E. A. Bums, SAMPE Proc. 1969, 15, 59.
c) H. R. Lubowitz, Polym. Prepr. (Am. Chem. Soc., Div. Polym.
Chem.) 1971, 12, 329. d) S. L. Kaplan, Plast. Des. Process. 1972,
1
2
X-IO2
X-IO4
®
®
®
®
61
147
3.0
4.5
4.5
4.8
1
3
4
5
6
7
8
9
X-IO2-CELp
X-IO4-CELp
X-IO4-CELp
X-IO4-CELp
X-IO4-CELp
X-IO4-CELp
®
®
0.5
0.25
0.5
1
0.5
0.5
0.5
48
130
158
110
146
104
124
2.7
4.2
4.3
4.2
6.8
4.1
8.9
3.5
4.3
4.6
3.0
4.6
3.9
1.9
®
®
CTABd
SDSe
2
3
4
5
f
X-IO4-CELp TRITONX-100
10
X-IO2-CELh
X-IO4-CELh
X-IO4-CELh
X-IO4-CELh
®
®
®
®
0.5
0.25
0.5
1
79
129
169
106
4.8
4.8
5.1
3.1
4.0
4.6
4.2
4.4
6
1
1
1
1
2
3
a
b
c
d
Tensile strength. Elongation at break. Tensile modulus. Cetyl-
1
0, 5.
trimethylammonium bromide. eSodium dodecyl sulfate. Poly-
f
7
8
9
a) A. K. St. Clair, T. L. St. Clair, Polym. Eng. Sci. 1982, 22, 9. b) T. L.
St. Clair, D. J. Progar, SAMPE Proc. 1979, 24, 1081.
(
ethylene glycol) (1,1,3,3-tetrabutyl)phenyl ether.
a) H. D. Stenzenberger, Adv. Polym. Sci. 1994, 117, 165. b) G. D.
Roberts, R. W. Lauver, J. Appl. Polym. Sci. 1987, 33, 2893.
a) R. Yokota, S. Yamamoto, S. Yano, T. Sawaguchi, M. Hasegawa, H.
Yamaguchi, H. Ozawa, R. Sato, High Perform. Polym. 2001, 13, S61.
b) T. Ogasawara, Y. Ishida, R. Yokota, T. Watanabe, T. Aoi, J. Goto,
Composites, Part A 2007, 38, 1296. c) M. Miyauchi, Y. Ishida, T.
Ogasawara, R. Yokota, Polym. J. 2012, 44, 959.
no significant improvement was observed for CEL introduced
composites (Run 3). In the case of X-IO4, the tensile strength
p
(
Ts) of the composite film increased with the amount of CEL
from 0.25 to 0.5 wt %, but decreased after that. When an over
.5 wt % of CEL sample was added to the IO solution, the
appearance of the finally obtained film became inhomogeneous.
This could be attributed to the large phase separation between
IO solution in NMP and CEL dispersion in water. However,
it is clear that CELh should be the better choice as an additive
0
1
0
M. Miyauchi, Y. Ishida, T. Ogasawara, R. Yokota, Polym. J. 2013, 45,
5
94.
T. Kudo, Y. Oishi, J. Oravec, K. Mori, J. Photopolym. Sci. Technol.
004, 17, 259.
11
2
1
2
a) A. K. Bledzki, J. Gassan, Prog. Polym. Sci. 1999, 24, 221. b) Q. Wu,
M. Henriksson, X. Liu, L. A. Berglund, Biomacromolecules 2007, 8,
3687. c) G. Siqueira, J. Bras, A. Dufresne, Polymers 2010, 2, 728.
d) M. M. Alam, T. Ahmed, M. M. Haque, M. A. Gafur, A. N. M. H.
Kabir, Polym.-Plast. Technol. Eng. 2008, 48, 110. e) H.-P. Fink, J.
Ganster, Macromol. Symp. 2006, 244, 107. f) T. Paunikallio, M.
Suvanto, T. T. Pakkanen, J. Appl. Polym. Sci. 2004, 91, 2676. g) D.
Bondeson, A. Mathew, K. Oksman, Cellulose 2006, 13, 171. h) D. M.
Panaitescu, P. V. Notingher, M. Ghiurea, F. Ciuprina, H. Paven, M.
Iorga, J. Optoelectron. Adv. Mater. 2007, 9, 2524. i) T. Nishino, I.
Matsuda, K. Hirao, Macromolecules 2004, 37, 7683.
because the higher T of 169 MPa and higher elongation at break
Eb) of 5.1% were achieved. Again, this can be explained by the
s
(
size and the shape effect of the nanofibrous CELh nature.
In order to improve the mechanical properties of X-IO4-
CELp composites, we then used several surfactants as the
modifier. As shown in Table 3, among them, cetyltrimethylam-
monium bromide (CTAB) and poly(ethylene glycol) (1,1,3,3-
μ
tetramethylbutyl)phenyl ether (TRITONX-100 ) realized higher
elongation at break up to 8.9% for the X-IO4-CELp composite
film. The cationic surfactant could get good interaction with
melamine moiety, and link with aliphatic CEL molecules very
well.
1
3
W. W. Cuthbertson, J. S. Moffatt, J. Chem. Soc. 1948, 561.
14 Supporting Information is available electronically on the CSJ-Journal
Web site, http://www.csj.jp/journals/chem-lett/index.htm.
Chem. Lett. 2014, 43, 787–789 | doi:10.1246/cl.140072
© 2014 The Chemical Society of Japan | 789