Microporous materials derived from two- and three-dimensional hyperbranched conjugated polymers by thermal elimination of substituents

Article abstract of DOI:10.1039/b707925g

Symmetrical and multivalent monomers having thermally degradable alkoxyl groups such as 1,3,5-triethynylbenzene and tetra(4-bromophenyl)silane derivatives were polymerised by certain appropriate methods, i.e. Hay or Sonogashira coupling, which gave hyperb

Full text of DOI:10.1039/b707925g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Microporous materials derived from two- and three-dimensional
hyperbranched conjugated polymers by thermal elimination of substituents
Norifumi Kobayashi and Masashi Kijima*
Received 24th May 2007, Accepted 23rd July 2007
First published as an Advance Article on the web 10th August 2007
DOI: 10.1039/b707925g
Symmetrical and multivalent monomers having thermally degradable alkoxyl groups such as
1,3,5-triethynylbenzene and tetra(4-bromophenyl)silane derivatives were polymerised by certain
appropriate methods, i.e. Hay or Sonogashira coupling, which gave hyperbranched
poly(phenylenebutadiynylene)s and poly(phenyleneethynylene)s having rigid macrocyclic
frameworks in high yields. By heat treating the polymers at about 400 uC, the degradable side
chains were selectively eliminated resulting in microporous pyrolytic polymers that showed
surface areas of ca. 800 m2 g21. This result suggests that the spaces resulting from the elimination
of the side chains contribute to the generation of the pores of the pyrolytic polymers. In addition,
most of the pyrolytic polymers retained their microporosity after further heat treatments at
900 uC, affording microporous carbon.
porosity, and further development of accurate methods to
Introduction
construct nano-structured materials is an interesting and
Porous carbons are receiving scientific and technological
inexhaustible research subject.
interest because of their large variety of potential and existing
Conjugated polymers consisting of rigid components such as
commercial applications such as gas separation, gas adsor-
benzene rings and ethynyl linkages are thermally stable and
bents, catalyst supports, electric double layer capacitors, etc.1
thermosetting, showing a high carbon fixation ability in the
carbonisation process.8–10 In contrast, alkyl chains consisting
Recently, development of tailored synthesis of nano-structured
porous carbons has become desirable to meet social demands
of carbon–carbon single bonds thermally degrade to volatile
of effective utilisation of resources and energy.2 Although
fragments at a relatively low temperature. Considering these
activated carbons3 are a typical and available porous carbon
thermal properties of organic materials, pore generation is
having high surface areas, it is usually difficult to control the
possible when the leaving group is selectively eliminated from
pore distribution by the surface activation processes. The
the rigid framework. However, the space left by the leaving
template synthesis of carbons using inorganic silicates4 has
group should be closed up by the thermal motion of the
been a useful method to synthesise nano-structured carbons
polymer chains during the heat treatment. In fact, it was
having regulated pore structures, but it is necessary to dissolve
out the template materials. Carbon synthesis by defluorinating
PTFE with alkaline metals5 is a method to selectively prepare
mesoporous carbons, which also requires elimination of the
included metal fluorides. As a simpler example, it has been
reported that the carbonisation of miscible polymer blends
affords meso- to macroporous materials only upon heating at
about 500 uC.6 In this case, the space left by the eliminated
materials results in porosity. Another report says that polymer
gels dried under supercritical or freeze drying conditions can
be converted to mesoporous carbons having a narrow pore size
distribution.7 Polymer gels are unique materials consisting of
submicron particles whose interspaces could characterise the
mesoporosity. On the other hand, pyrolysis of conjugated
polymers affords microporous carbons.8,9 The generation of
the micropores is due to nano-scale gaps between the aggre-
gated basic structural units of the carbonaceous layers caused
during the carbonisation process. Indeed, these methods are
useful to prepare porous carbons having a characteristic
suggested that almost the same pre-carbon materials were
formed from certain poly(phenylenebutadiynylene)s irrespec-
tive of them having leaving groups or not.9 In order to
minimise the closing up of the pore space, we adopted hyper-
branched rigid conjugated polymers with symmetrical struc-
tures as thermosetting rigid frameworks.
In this paper, we constructed hyperbranched polymer
networks using rigid phenyleneethynylene and phenylene-
butadiynylene linkers, and degradable alkyl groups were
introduced onto each structure unit of the polymers. Ideally,
the polymer network must involve cyclic frameworks as shown
in Fig. 1. Eliminating the alkyl group from the rigid polymer
network selectively through pyrolysis at low temperature,
Institute of Materials Science, Graduate School of Pure and Applied
Sciences, University of Tsukuba, Tsukuba, Ibaraki, 305-8573, Japan.
E-mail: kijima@ims.tsukuba.ac.jp; Fax: +81 29 853 4490;
Tel: +81 29 853 6905
Fig. 1 Formation of microporous pyrolytic polymers from rigid
polymeric frameworks having thermally degradable portions.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4289–4296 | 4289

View Article Online
below 450 uC, should form microporous materials. The
pyrolytic polymers have the advantage of high thermal stabi-
lity compared to usual organic materials, and have a greater
flexibility of modifications compared to carbon materials.
In addition, the porosity of the pyrolytic polymers should be
retained even after carbonisation through heat treatment at
about 900 uC.
60 uC for 60 h. The reaction mixture was concentrated and
extracted with CHCl3. The organic layer was washed with 4 M
HCl, H2O, and 4 M NH3. Reprecipitation from CHCl3–
MeOH–H2O gave P1 (779 mg) as a yellow powder. IR (KBr)
3432, 2956, 2869, 2208, 1497, 1413, 1277, 1213, 1026, 835 cm21
;
Anal. Calcd for (C24H24O2Br0.3)n: C, 78.24; H, 6.57; N,
0.00; Br, 6.51. Found: C, 75.82; H, 6.57; N, 0.05; Br,
6.14%. GPC: Mw = 11700, Mn = 5800, and DP (degree of
polymerisation) = 17.
Experimental
General method and instrumentation
General procedure for the preparation of 2a–2c: 1,3,5-
tribromo-2-octadecyloxybenzene (2a). A mixture of 1-bromo-
octadecane (3.00 g, 9.00 mmol), K2CO3 (6.22 g, 45.0 mmol),
2,4,6-tribromophenol (5.95 g, 18.0 mmol), and DMF (20 mL)
was stirred at 70 uC for 17 h. After cooling to room tempera-
ture, hexane was added to the reaction mixture. The organic
layer was washed with H2O, 2 M NaOH, and brine and dried
over Na2SO4 to give 2a (5.17 g, 98.5% yield) as a white powder.
1H-NMR (CDCl3) d 7.63 (s, 2H), 3.97 (t, J = 6.7 Hz, 2H), 1.85
(m, 2H), 1.51–1.24 (m, 30H), 0.88 (t, J = 6.9 Hz, 3H);
13C-NMR (CDCl3) d 152.9, 134.9, 119.0, 117.0, 73.8, 32.0,
30.1, 29.8, 29.75, 29.70, 29.66, 29.50, 29.45, 25.9, 22.8, 14.2; IR
Synthetic manipulations were performed using standard
Schlenk glassware under an argon atmosphere except for the
synthesis of P5a–P5e. The disk-pellet sample (w = 13 mm,
50 mg) was prepared under a pressure of 3 MPa using a press
unit (Hitachi). The pellet on a quartz boat was heat-treated in
a quartz tube under flowing argon with a furnace EKRO-12 K
(Isuzu).
IR and NMR spectra were recorded on a JASCO FT/IR 550
spectrophotometer and a JEOL EX-270 (270 MHz) spectro-
meter, respectively. Elemental analyses were carried out with a
Perkin-Elmer type 2400. Thermal analyses were carried out by
an Extra 6000 TG/DTA (Seiko) thermogravimetric analyser at
a rate of 10 uC min21 under an argon atmosphere.
(KBr) 2918, 2848, 1561, 1446, 1387, 1252, 978, 859, 736 cm21
;
Anal. Calcd for C24H39Br3O: C, 49.42; H, 6.74; N, 0.00.
Found: C, 49.50; H, 6.61; N, 0.00%.
N2 adsorption analyses were performed with a SA 3100
(Coulter) at 77 K. The pore characterisations were carried out
by the as-plot method11 for micropores and the DH method12
for mesopores. The subtracting pore effect (SPE) analysis of
the as-plots provides the total surface area (Stotal) from the
slope of the narrow linear region around as = 0.5, the external
surface area (Aext) from the slope of the linear region between
as = 0.7 and 1.1, using eqn (1). The total micropore volume
(Vmicro) is also obtained from the intercept of the latter
straight line. Further, the average micropore width (wa) can be
evaluated from Stotal, Aext, and Vmicro according to eqn (2),
assuming that the pore shape is slit-like.
1,3,5-Tribromo-2-phenoxybutoxybenzene (2b). Colourless oil
1
(4.12 g, 98.5% yield); H-NMR (CDCl3) d 7.65 (s, 2H), 7.31–
7.25 (m, 2H), 6.97–6.90 (m, 3H), 4.08 (t, J = 3.8 Hz, 2H), 4.07
(t, J = 3.8 Hz, 2H), 2.08–2.05 (m, 4H); 13C-NMR (CDCl3) d
158.8, 152.7, 134.9, 129.3, 120.5, 119.0, 117.2, 114.4, 73.1, 67.4,
26.8, 26.0; IR (KBr) n 3067, 2950, 2874, 1600, 1496, 1442,
1372, 1038, 962, 857, 753 cm21; Anal. Calcd for C16H15Br3O2:
C, 40.12; H, 3.16; N, 0.00. Found: C, 40.43; H, 3.41; N, 0.00%.
1,3,5-Tribromo-2-cyclohexylmethoxybenzene (2c). Colourless
oil (3.28 g, 85.4% yield); 1H-NMR (CDCl3) d 7.63 (s, 2H), 3.77
(d, J = 5.9 Hz, 2H), 1.96–1.72 (m, 6H), 1.35–1.11 (m, 5H);
13C-NMR (CDCl3) d 152.8, 134.9, 119.0, 117.0, 78.6, 38.6,
29.8, 26.6, 25.9; IR (KBr) n 2924, 2850, 1538, 1444, 1371, 1248,
1076, 986, 857, 739, 683 cm21; Anal. Calcd for C13H15Br3O: C,
36.57; H, 3.54; N, 0.00. Found: C, 36.65; H, 3.75; N, 0.00%.
Stotal = 2.14 6 (slope) [m2 g21
]
(1)
(2)
wa = 2Vmicro/(Atotal 2 Aext) [nm]
Materials
General procedure for the preparation of 3a–3d: 1-phenoxy-
butoxy-2,4,6-tris(trimethylsilylethynyl)benzene (3b). A mixture
of CuI (30.5 mg, 0.160 mmol), PdCl2(PPh3)2 (84.2 mg,
0.120 mmol), PPh3 (62.9 mg, 0.240 mmol), Et3N (30 mL), 2b
(1.92 g, 4.00 mmol), and trimethylsilylacetylene (2.83 mL,
20.0 mmol) was stirred at 55 uC for 36 h. After evaporation of
Et3N, the residue was triturated with hexane and filtered. The
filtrate was subjected to column chromatography on silica gel
eluted with hexane–CH2Cl2 (7 : 1) to give 3b (1.99 g, 93.7%
yield) as a colourless oil. 1H-NMR (CDCl3) d 7.48 (s, 2H),
7.30–7.24 (m, 2H), 6.95–6.88 (m, 3H), 4.30 (t, J = 5.8 Hz, 2H),
4.03 (t, J = 6.1 Hz, 2H), 2.08–1.97 (m, 4H), 0.24 (s, 18H),
0.22 (s, 9H); 13C-NMR (CDCl3) d 161.4, 158.9, 137.3,
129.3, 120.4, 118.3, 117.6, 114.4, 102.9, 99.9, 99.7, 94.5, 73.6,
67.4, 27.1, 26.3, 0.0, 20.1; IR (KBr) n 2958, 2158, 1599,
1443, 1378, 1249, 1005, 859, 834, 758 cm21; Anal. Calcd for
DMF, THF, and Et3N were dried and distilled before being
used for reaction solvents. 1,4-Dibutoxy-2,5-diethynylbenzene,13
1,3,5-tribromo-2-methoxymethoxybenzene,14 1,3,5-tribromo-2-
(tert-butyldimethylsilyloxy)benzene,15 1,3,5-triethynylbenzene,16
tetra(4-bromophenyl)silane,17 and 1,4-didodecyl-2,5-diethynyl-
benzene18 were synthesised according to literature procedures.
Other chemicals were obtained from commercial sources and
used without further purification.
Synthesis
Poly[(1,4-dibromobenzene)-alt-(1,4-dibutoxy-3,5-diethynyl-
benzene)] (P1). A mixture of CuI (22.9 mg, 0.120 mmol),
1,4-dibutoxy-2,5-diethynylbenzene (562 mg, 2.08 mmol), 1,4-
dibromobenzene (472 mg, 2.00 mmol), Pd(PPh3)4 (139 mg,
0.120 mmol), Et3N (2.0 mL) and THF (25 mL) was stirred at
4290 | J. Mater. Chem., 2007, 17, 4289–4296
This journal is ß The Royal Society of Chemistry 2007

View Article Online
C31H42O2Si3: C, 70.13; H, 7.97; N, 0.00. Found: C, 69.84; H,
7.93; N, 0.00%.
1384, 1238, 1044, 947, 855, 749, 648 cm21; Anal. Calcd for
C22H18O2: C, 84.05; H, 5.77; N, 0.00. Found: C, 84.17; H, 6.03;
N, 0.00%.
1-Octadecyloxy-2,4,6-tris(trimethylsilylethynyl)benzene (3a).
1
Pale yellow oil (2.27 g, 89.4% yield); H-NMR (CDCl3) d 7.47
1-Cyclohexylmethoxy-2,4,6-triethynylbenzene (4c). White
powder (789 mg, 100% yield); 1H-NMR (CDCl3) d 7.54 (s,
2H), 4.05 (d, J = 6.1 Hz, 2H), 3.26 (s, 2H), 3.03 (s, 1H), 1.94–
1.73 (m, 6H), 1.33–1.09 (m, 5H); 13C-NMR (CDCl3) d 162.6,
138.0, 117.1, 116.8, 82.4, 81.4, 79.8, 78.7, 77.6, 38.7, 29.7, 26.6,
25.9; IR (KBr) n 3296, 2925, 2852, 2113, 1448, 1380, 1241, 991,
891, 656, 612 cm21; Anal. Calcd for C19H18O: C, 86.99; H,
6.92; N, 0.00. Found: C, 86.77; H, 7.15; N, 0.02%.
(s, 2H), 4.21 (t, J = 6.5 Hz, 2H), 1.79 (m, 2H), 1.50–1.24 (m,
30H), 0.88 (t, J = 6.8 Hz, 3H), 0.23 (s, 18H), 0.21 (s, 9H);
13C-NMR (CDCl3) d 161.8, 137.2, 118.1, 117.6, 103.0, 100.0,
99.5, 94.3, 74.4, 32.1, 30.5, 29.8, 29.7, 29.50, 26.2, 22.8, 14.3,
0.1, 0.0; IR (KBr) n 2925, 2854, 2158, 1444, 1379, 1250, 1006,
851, 760 cm21; Anal. Calcd for C39H66OSi3: C, 73.74; H,
10.47; N, 0.00. Found: C, 73.85; H, 10.18; N, 0.00%.
1-Cyclohexylmethoxy-2,4,6-tris(trimethylsilylethynyl)benzene
1
(3c). White powder (1.83 g, 95.6% yield); H-NMR (CDCl3) d
1,3,5-Triethynyl-2-methoxymethoxybenzene (4d). White
powder (883 mg, 95.4% yield); 1H-NMR (CDCl3) d 7.58
(s, 2H), 5.39 (s, 2H), 3.64 (s, 3H), 3.30 (s, 2H), 3.05 (s, 1H);
13C-NMR (CDCl3) d 160.0, 138.1, 118.0, 117.1, 99.2, 82.5,
81.2, 78.7, 77.9, 57.9; IR (KBr) n 3276, 2963, 2829, 2109, 1788,
1462, 1400, 1207, 1157, 1093, 940, 888, 675, 618 cm21; Anal.
Calcd for C14H10O2: C, 79.98; H, 4.79; N, 0.00. Found: C,
80.03; H, 4.95; N, 0.03%.
7.47 (s, 2H), 4.07 (d, J = 6.1 Hz, 2H), 1.97–1.74 (m, 6H), 1.33–
1.08 (m, 5H), 0.25 (s, 18H), 0.22 (s, 9H); 13C-NMR (CDCl3) d
161.8, 137.3, 117.9, 117.2, 103.0, 100.0, 99.5, 94.2, 79.4, 38.8,
29.9, 26.6, 26.0, 0.0, 20.1; IR (KBr) n 2926, 2923, 2852,
2159, 1446, 1384, 1249, 989, 842, 759 cm21; Anal. Calcd for
C28H42OSi3: C, 70.22; H, 8.84; N, 0.00. Found: C, 70.39; H,
8.80; N, 0.00%.
General procedure for the preparation of P5a–P5e:
poly(1,3,5-triethynyl-2-octadecyloxybenzene) (P5b). A mixture
of CuCl (42.6 mg, 0.430 mmol), THF (5 mL), N,N,N9,N9-
tetramethylethylenediamine (TMEDA) (75.7 mL, 0.502 mmol),
and 4a (600 mg, 1.43 mmol) in THF (20 mL) was stirred under
oxygen at room temperature for 48 h. The reaction mixture
was poured into MeOH–conc. HCl (10 : 1) and stirred for 24 h.
After filtration, the solid was poured into MeOH–NH3 aq.
(10 : 1) and stirred for 12 h. After filtration, the solid was
washed with MeOH to give P5b (600 mg) as a pale yellow gel.
This polymer is insoluble in common organic solvents such as
CHCl3 and THF. IR (KBr) n 3306, 2922, 2851, 2220, 1773,
1559, 1442, 1377, 1241, 956, 886 cm21; Anal. Calcd for
(C30H39O)n: C, 86.69; H, 9.46; N, 0.00. Found: C, 85.50; H,
9.24; N, 0.08%.
1-Methoxymethoxy-2,4,6-tris(trimethylsilylethynyl)benzene
(3d). Hexane–CH2Cl2 (3 : 1) was used as the eluent for silica gel
column chromatography and gave a colourless oil (2.65 g,
88.9% yield). 1H-NMR (CDCl3) d 7.43 (s, 2H), 5.31(s,
2H), 3.60 (s, 3H), 0.17 (s, 18H), 0.15 (s, 9H); 13C-NMR
(CDCl3) d 159.3, 137.3, 118.8, 117.8, 102.7, 100.0, 99.7, 98.9,
94.7, 57.7, 20.0, 20.1; IR (KBr) n 2959, 2899, 2159, 1435,
1393, 1250, 1161, 1007, 946, 848, 760 cm21; Anal. Calcd for
C23H34O2Si3: C, 64.73; H, 8.03; N, 0.00. Found: C, 64.91; H,
7.93; N, 0.18%.
General procedure for the preparation of 4a–4d: 1,3,5-
triethynyl-2-octadecyloxybenzene (4a).
A mixture of 3a
(1.59 g, 2.50 mmol), THF (10 mL), MeOH (3.5 mL), and
K2CO3 (51.8 mg, 0.375 mmol) in H2O (0.9 mL) was stirred at
room temperature for 10 h. The reaction mixture was diluted
with H2O (5 mL). After evaporation of organic solvents, the
aqueous residue was extracted with CH2Cl2. The organic layer
was washed with H2O and brine and dried over Na2SO4 to give
4a (1.01 g, 96.8% yield) as a white powder. 1H-NMR (CDCl3)
d 7.55 (s, 2H), 4.24 (t, J = 6.6 Hz, 2H), 3.26 (s, 2H), 3.02 (s,
1H), 1.80 (m, 2H), 1.52–1.24 (m, 30H), 0.88 (t, J = 6.9 Hz, 3H);
13C-NMR (CDCl3) d 162.5, 138.0, 117.2, 116.9, 82.3, 81.4,
78.6, 77.6, 74.7, 32.0, 30.3, 29.79, 29.75, 29.72, 29.68, 29.49,
29.45, 25.9, 22.8, 14.2; IR (KBr) n 3314, 3275, 2914, 2849,
2109, 1471, 1389, 1240, 1137, 1066, 885 cm21; Anal. Calcd for
C30H42O: C, 86.07; H, 10.11; N, 0.00. Found: C, 86.12; H,
9.90; N, 0.00%.
Poly(1,3,5-triethynylbenzene) (P5a). Explosive brown gel
(372 mg) was obtained from triethynylbenzene (350 mg). IR
(KBr) n 3421, 2189, 1671, 1572, 1406, 1212, 877 cm21; Anal.
Calcd for (C12H5)n: C, 96.62; H, 3.38; N, 0.00. Found: C,
86.79; H, 4.14; N, 0.50%.
Poly(1,3,5-triethynyl-2-phenoxybutoxybenzene) (P5c). Pale
brown gel (461 mg) was obtained from 4b (469 mg). IR
(KBr) n 3286, 2945, 2871, 2215, 1599, 1496, 1440, 1241, 883,
753, 690 cm21; Anal. Calcd for (C22H15O2)n: C, 84.87; H, 4.86;
N, 0.00. Found: C, 82.96; H, 5.26; N, 0.27%.
Poly(1-cyclohexylmethoxy-2,4,6-triethynylbenzene) (P5d).
Pale brown gel (550 mg) was obtained from 4c (550 mg). IR
(KBr) n 3296, 2921, 2849, 2208, 1745, 1558, 1437, 1375, 1241,
984, 884 cm21; Anal. Calcd for (C19H15O)n: C, 88.00; H, 5.83;
N, 0.00. Found: C, 86.16; H, 6.11; N, 0.10%.
1,3,5-Triethynyl-2-phenoxybutoxybenzene
(4b).
White
powder (600 mg, 83.4% yield); 1H-NMR (CDCl3) d 7.55 (s,
2H), 7.30–7.23 (m, 2H), 6.95–6.87 (m, 3H), 4.32 (t, J = 5.9 Hz,
2H), 4.04 (t, J = 6.0 Hz, 2H), 3.25 (s, 2H), 3.03 (s, 1H), 2.08–
1.95 (m, 4H); 13C-NMR (CDCl3) d 162.2, 158.8, 138.0, 129.3,
120.5, 117.4, 116.8, 114.4, 82.5, 81.3, 78.5, 77.7, 74.0, 67.4,
27.0, 25.9; IR (KBr) n 3267, 2958, 2117, 1601, 1497, 1445,
Poly(1,3,5-triethynyl-2-methoxymethoxybenzene) (P5e).
Yellow gel (584 mg) was obtained from 4d (600 mg). IR
(KBr) n 3294, 2925, 2827, 2197, 1578, 1434, 1389, 1237, 1200,
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4289–4296 | 4291

View Article Online
1158, 1081, 925 cm21; Anal. Calcd for (C14H7O2)n: C, 81.15;
H, 3.41; N, 0.00. Found: C, 75.68; H, 3.80; N, 0.60%.
(21.3 mg, 0.0184 mmol), Et3N (1.5 mL) and THF (25 mL) was
stirred at 65 uC for 48 h. The same purification procedure was
used as for the synthesis of P5a–P5e to give P7b (480 mg) as a
yellow gel. This polymer is insoluble in common organic
solvents such as CHCl3 and THF. IR (KBr) 3413, 2956, 2870,
1592, 1503, 1412, 1378, 1277, 1211, 1096, 1066, 1010, 822, 732,
678 cm21; Anal. Calcd for (C60H56O4SiBr)n: C, 75.93; H, 5.95;
N, 0.00; Br, 8.42. Found: C, 73.65; H, 6.17; N, 0.00; Br, 8.18%.
General procedure for the preparation of P6a–P6f:
poly[(1,3,5-tribromo-2-octadecyloxybenzene)-co-(1,3,5-triethy-
nylbenzene)] (P6b). A mixture of CuI (32.7 mg, 0.172 mmol),
1,3,5-triethynylbenzene (447 mg, 2.98 mmol), 2a (1.67 g,
2.86 mmol), Pd(PPh3)4 (198 mg, 0.172 mmol), Et3N (2.5 mL)
and THF (20 mL) was stirred at 65 uC for 72 h. The same
purification procedure was used as for the synthesis of
P5a–P5e to give P6b (1.70 g) as a brown gel. This polymer is
insoluble in common organic solvents such as CHCl3 and
THF. IR (KBr) n 3310, 2922, 2850, 2207, 1577, 1455,
1377, 1238, 984, 877, 744, 690 cm21; Anal. Calcd for
(C36H42OBr1.38)n: C, 71.95; H, 7.04; N, 0.00; Br, 18.35.
Found: C, 70.03; H, 7.10; N, 0.00; Br, 17.98%.
Poly{[tetra(4-bromophenyl)silane]-co-(1,4-diethynylbenzene)}
(P7a). Yellow gel (347 mg) was obtained from tetra(4-
bromophenyl)silane (300 mg). IR (KBr) 3420, 3064, 2209,
1591, 1571, 1509, 1377, 1306, 1259, 1188, 1097, 1010, 822, 730,
687 cm21; Anal. Calcd for (C44H24SiBr0.9)n: C, 80.97; H, 3.71;
N, 0.00; Br, 11.02. Found: C, 77.02; H, 4.51; N, 0.20; Br,
11.09%.
Poly{[tetra(4-bromophenyl)silane]-co-(1,4-didodecyl-2,5-
diethynylbenzene)} (P7c). Yellow gel (366 mg) was obtained
from tetra(4-bromophenyl)silane (200 mg). IR (KBr) 3413,
2923, 2851, 2209, 1593, 1572, 1465, 1377, 1260, 1187, 1102,
1067, 1011, 897, 822, 732, 682 cm21; Anal. Calcd for
(C92H120SiBr1.6)n: C, 79.96; H, 8.75; N, 0.00; Br, 9.25.
Found: C, 78.39; H, 7.74; N, 0.00; Br, 9.24%.
Poly[(1,3,5-tribromobenzne)-co-(1,3,5-triethynylbenzene)]
(P6a). Brown gel (676 mg) was obtained from 1,3,5-tribromo-
benzene (605 mg). IR (KBr) 3300, 3144, 2212, 1586, 1433, 1407,
1098, 1018, 876, 676 cm21; Anal. Calcd for (C18H6Br0.8)n: C,
75.55; H, 2.11; N, 0.00; Br, 22.34. Found: C, 68.20; H, 3.48; N,
1.72; Br, 22.62%.
Poly[(1,3,5-tribromo-2-phenoxybutoxybenzene)-co-(1,3,5-
triethynylbenzene)] (P6c). Yellow gel (1.29 g) was obtained
from 2b (1.20 g). IR (KBr) n 3295, 3059, 2944, 2871, 2215,
1585, 1496, 1470, 1454, 1379, 1242, 877, 751, 690 cm21; Anal.
Calcd for (C28H18O2Br1.23)n: C, 69.38; H, 3.74; N, 0.00; Br,
20.28. Found: C, 67.65; H, 4.32; N, 0.00; Br, 20.68%.
Results and discussion
Syntheses and thermal properties
In order to prepare the microporous pyrolytic polymers from
rigid polymeric frameworks having thermally degradable
groups as shown in Fig. 1, we attempted to synthesise some
kinds of polymers which are one-dimensional, two-dimen-
sional, and three-dimensional polymers as shown in Schemes
1–4. In all cases, the introduced side chains play the role of
efficiently progressing the polymerisation reactions by increas-
ing the solubility of the materials, in addition to the primary
role as the moiety for elimination.
Poly[(1,3,5-tribromo-2-cyclohexylmethoxybenzene)-co-(1,3,5-
triethynylbenzene)] (P6d). Brown gel (807 mg) was obtained
from 2c (1.07 g). IR (KBr) n 3299, 2920, 2848, 2209, 1577, 1469,
1237, 984, 876, 743 cm21; Anal. Calcd for (C25H18OBr1.41)n: C,
67.16; H, 4.06; N, 0.00; Br, 25.20. Found: C, 65.83; H, 4.73; N,
0.00; Br, 25.50%.
First, the linear chain polymer (P1) having phenylene-
ethynylene linkages was synthesised by Sonogashira coupling
reaction19 of 1,4-dibutoxy-2,5-diethynylbenzene with p-dibro-
mobenzene in the presence of Pd(PPh3)4 and CuI (Scheme 1).
Second, the hyperbranched two-dimensional polymers having
phenylenebutadiynylene linkages (P5a–P5e) were synthesised
by Hay polycondensation20 of the corresponding 1,3,5-
triethynyl-2-substituted benzenes in the presence of CuCl and
TMEDA under an oxygen atmosphere (Scheme 2). Last, the
Poly[(1,3,5-tribromo-2-methoxymethoxybenzene)-co-(1,3,5-
triethynylbenzene)] (P6e). Brown gel (1.26 g) was obtained
from 1,3,5-tribromo-2-methoxymethoxybenzene (1.20 g). IR
(KBr) n 3294, 2951, 2825, 2205, 1578, 1457, 1436, 1159, 1090,
927, 876, 742, 692 cm21; Anal. Calcd for (C20H10O2Br1.23)n: C,
63.12; H, 2.65; N, 0.00; Br, 25.82. Found: C, 61.36; H, 3.31; N,
0.19; Br, 26.19%.
Poly{[1,3,5-tribromo-2-(tert-butyldimethylsilyloxy)benzene]-
co-(1,3,5-triethynylbenzene)} (P6f). Brown gel (1.62 g) was
obtained from 1,3,5-tribromo-2-(tert-butyldimethylsilyloxy)-
benzene (1.42 g). IR (KBr) n 3308, 2953, 2857, 2198, 1576,
1460, 1303, 1255, 908, 841, 785, 691 cm21; Anal. Calcd for
(C24H20OSiBr1.55)n: C, 60.51; H, 4.23; N, 0.00; Br, 26.00.
Found: C, 58.32; H, 4.35; N, 0.06; Br, 25.74%.
General procedure for the preparation of P7a–P7c: poly-
{[tetra(4-bromophenyl)silane]-co-(1,4-dibutoxy-2,5-diethynyl-
benzene)} (P7b). A mixture of CuI (5.26 mg, 0.0276 mmol),
1,4-dibutoxy-2,5-diethynylbenzene (255 mg, 0.943 mmol),
tetra(4-bromophenyl)silane (300 mg, 0.460 mmol), Pd(PPh3)4
Scheme 1 Synthesis of P1.
4292 | J. Mater. Chem., 2007, 17, 4289–4296
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Scheme 4 Synthesis of P7a–c.
Scheme 2 Synthesis route to P5a–e.
hyperbranched two-dimensional polymers
(P6a–P6f)
(Scheme 3) and the three-dimensional polymers (P7a–P7c)
(Scheme 4) having phenyleneethynylene linkages were also
synthesised by Sonogashira coupling. All polymers were
obtained in high yields.
The polymers P5–P7 are obtained in a gelatinous form, and
are insoluble in most organic solvents such as CHCl3, THF,
and DMF. It has been reported that the hyperbranched P5
having hexyloxy groups is completely soluble in common
organic solvents in spite of having a high weight-average
molecular weight (Mw = 30 000).21 The low solubility of
P5b–P5d despite having effective solubilising groups suggests
that the cyclic network skeletons are sufficiently built by the
polymerisation.
Fig. 2 TGA curves of P1, P5c, P6d, and P7b.
500 uC, which is due to the thermal decomposition of the alkyl
side chains. After the gradual mass loss in the higher tem-
perature region up to 900 uC, the carbonised materials retain
about 50% mass, which means the carbon fixation ability of
the frameworks is adequately high.
To investigate the elimination behaviour of the alkyl chains,
thermogravimetric analyses (TGA) of P1, P5c, P6d, and P7b
were performed (Fig. 2). In all cases, the mass loss of the
polymers begins at about 300 uC and reaches ca. 40% loss at
Surface analysis
The mass % yields and the surface parameters of all the
heat-treated samples at several temperatures are summarised
in Table 1. Fig. 3 shows N2 adsorption isotherms of the
linear chain polymer heat-treated at 350, 400, and 900 uC,
abbreviated as P1(350), P1(400), and P1(900), respectively.
The fact that P1(400) yielded 69% mass suggests that the butyl
side chains were mostly eliminated at 400 uC, but its total
surface area (Stotal) turned out to be small (2 m2 g21). This
means that the linear polymer chains are not rigid enough to
retain the space of the eliminated groups under the heating
conditions. Thus, the high Stotal (1040 m2 g21) observed in the
case of P1(900) is considered to be due to the formation of
microporous gaps caused by the aggregation of the basic
structural units of carbonaceous layers during the carbonisa-
tion as well as the poly(phenyleneethynylene) analogues.9
Fig. 4 shows isotherms of the hyperbranched polymers
without elimination groups (P5a, P6a, and P7a) which were
Scheme 3 Synthesis of P6a–f.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4289–4296 | 4293

View Article Online
Table 1 Surface analysis data of the heat-treated polymersa
Temp.b/ Yield Stotal
Vmicro Vtotal Vmeso
/
/
/
/
wa/
Polymer uC
(%) m2 g21 mL g21 mL g21 Vtotal (%) nm
P1
350
73
69
52
92
87
77
76
85
79
54
48
64
56
60
54
66
60
63
59
43
39
69
59
69
67
78
74
76
65
57
45
2
2
N/Ac
N/Ac
0.31
0.14
0.28
0.097
0.24
0.037
0.23
0.27
N/Ac
0.22
0.24
0.25
0.29
0.21
0.25
0.22
0.20
0.099
N/Ac
0.26
0.23
0.21
0.16
0.13
0.20
0.16
0.18
0.15
0.18
0.026
0.016
0.39
0.34
0.42
0.29
0.31
0.21
0.33
0.37
0.090
0.43
0.36
0.45
0.36
0.39
0.31
0.42
0.28
0.18
0.048
0.51
0.40
0.55
0.28
0.35
0.25
0.52
0.27
0.35
0.31
49
86
9.0
N/Ac
N/Ac
0.62
0.83
0.68
0.92
0.62
1.4
400
900
450
900
450
900
450
900
350d
900
400
900
450
900
450
900
450
900
450
900
450
900
450
900
450
900
350
900
400
900
1040
520
960
360
960
180
760
840
13
770
850
910
990
820
850
750
650
260
2
940
820
820
560
550
630
570
630
540
650
P5a
P6a
P7a
P6b
P6c
P6d
P6e
P6f
P5b
P5c
P5d
P5e
P7b
P7c
a
21
14
21
22
24
14
10
27
23
15
17
5.9
20
2.3
21
10
0.68
0.73
N/Ac
0.70
0.63
0.70
0.63
0.64
0.65
0.66
0.67
0.99
N/Ac
0.67
0.63
0.75
0.70
0.75
0.68
0.85
0.66
0.81
0.63
Fig. 4 N2 adsorption isotherms of the heat-treated samples of P5a,
P6a, and P7a.
3.1
84
32
22
26
17
23
2.4
36
14
19
24
Total surface area (Stotal), micropore volume (Vmicro), and average
micropore width (wa) were determined by the as plots and the SPE
analysis. The total pore volume (Vtotal) was estimated from the
maximum adsorbed volume, and the mesopore ratio (Vmeso/Vtotal
)
b
was estimated by the DH method. In the case of 900 uC, the
samples were carbonised at a heating rate of 5 uC min21. In the case
of 350–450 uC, the samples were heat-treated at a heating rate of
10 uC min21 and annealed at the temperature for 3 h, except for
c
d
P6b. Not applicable. Annealed at 350 uC for 9 h.
Fig. 5 N2 adsorption isotherms of the heat-treated samples of P6b–f.
heat-treated at 450 uC and 900 uC. P5a(450) shows the Stotal
of 520 m2 g21 due to the cross-linking reactions at the
butadiyne and terminal acetylene groups,8,9,22 while P6a(450)
and P7a(450) show the Stotal of 360 and 180 m2 g21, respec-
tively, mainly due to the elimination of terminal Br. These
Stotal are about half of those of P5a(900), P6a(900), and
P7a(900), which means that P5a, P6a, and P7a require heat
treatment at high temperature (about 900 uC) to develop
similar porous structures as P1.
Fig. 5, 6, and 7 show the isotherms of the heat-treated
P6b–f, P5b–e, and P7b,c which have both hyperbranched and
cyclised rigid frameworks and degradable alkyl side chains. In
these cases, the samples heat-treated below 450 uC adsorb
more N2 gaseous molecules than those heat-treated at 900 uC.
In addition, the average micropore widths (wa) of the samples
heat-treated at 900 uC became narrower and their mesopore
ratios (Vmeso/Vtotal) became smaller than those heat-treated
below 450 uC. These results suggest that the elimination spaces
of the alkyl chains generated by the heat-treatment below
450 uC are reflected in the porous structures of the materials
showing high Stotal, and the higher-temperature treatment at
900 uC ends in shrinkage of the materials. In fact, in the cases
of 6b and 5b having long string-like octadecyl side chains,
the Stotal observed in P6b(350) and P5b(450) considerably
Fig. 3 N2 adsorption isotherms of the heat-treated samples of P1.
4294 | J. Mater. Chem., 2007, 17, 4289–4296
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Fig. 8 Schematic outline for the pore formation from hyperbranched
rigid polymers having leaving groups under pyrolytic conditions.
to microporous materials having Stotal of about 800 m2 g21 by
heat treatments around 400 uC. These results suggest that the
elimination spaces of the leaving groups of the hyperbranched
rigid polymers could serve as permanent micropores. As a
result, we successfully synthesised porous pyrolytic polymers
by the presented method. The schematic outline of this result is
summarised in Fig. 8. Actually, the hyperbranched polymers
are more disordered in two and three dimensions. At the first
step, leaving groups were selectively eliminated from the rigid
polymer frameworks to form the porous pyrolytic polymer by
heating below 450 uC. The elimination groups decomposed
leading to sublimation of gaseous substances, leaving pores:
the elimination spaces. Some should remain though others
might close or connect with each other. Additionally, we
experimented with further heat treatments at 900 uC. Though
we found shrinkage of the materials to some degree, the
pores of the pyrolytic polymer were retained, affording
microporous carbon having narrower spaces than the results
of the first step.
Fig. 6 N2 adsorption isotherms of the heat-treated samples of P5b–e.
Acknowledgements
This work was supported by a JSPS Research Fellowships
(No. 18?3824). We thank the Research Facility Centre for
Science and Technology, Chemical Analysis Division,
University of Tsukuba, for NMR, elemental, thermal, and
surface analysis data.
Fig. 7 N2 adsorption isotherms of the heat-treated samples of P7b,c.
decreased by heat-treating at 900 uC from 840 and 260 m2 g21
to 13 and 2 m2 g21, respectively. This means that the narrow
elimination space derived from the octadecyl chains closed up
more easily than the other wide spaces derived from the bulky
side chains under the heating conditions.
References
1 W. Han, H. Liu and H. Zhu, Catal. Commun., 2007, 8, 351;
Z. Yang, Y. Xia and R. Mokaya, J. Am. Chem. Soc., 2007, 129,
1673; H. Yamada, H. Nakamura, F. Nakahara, I. Moriguchi and
T. Kubo, J. Phys. Chem. C, 2007, 111, 227; Y. Wang, H. C. Zeng
and J. Y. Lee, Adv. Mater., 2006, 18, 645; S.-I. Kim, T. Yamamoto,
A. Endo, T. Ohmori and M. Nakaiwa, Microporous Mesoporous
Mater., 2006, 96, 191; P. S. Tin, T.-S. Chung, L. Jiang and
S. Kulprathipanja, Carbon, 2005, 43, 2025.
Conclusion
For the purpose of obtaining novel porous materials utilising
elimination spaces of the leaving groups as the pores, we
synthesised 2- and 3-dimensional hyperbranched rigid poly-
mers having leaving alkyl substituents (P5, P6, and P7) in high
yields. Since the polymers obtained in this study are insoluble
in most organic solvents, the polymers are assumed to be very
hyperbranched and partially cyclic framework materials. The
leaving groups basically consisting of carbon–carbon single
bonds thermally decompose around 400 uC. By heat treat-
ments below 450 uC, the one-dimensional polymer (P1) and
hyperbranched polymers without leaving groups (P5a, P6a,
and P7a) produced low porous materials. On the other hand,
the two- and three-dimensional hyperbranched polymers
having leaving groups (P6b–f, P5b–e, and P7b,c) converted
2 Z. Lei, Y. Zhang, H. Wang, Y. Ke, J. Li, F. Li and J. Xing,
J. Mater. Chem., 2001, 11, 1975; A. A. Zakhidov, R. H. Baughman,
Z. Iqbal, C. Cui, I. Khayrullin, S. O. Dantas, J. Marti and
V. G. Ralchenko, Science, 1998, 282, 897; T. Kyotani, N. Sonobe
and A. Tomita, Nature, 1988, 331, 331.
3 Z. Hu, M. P. Srinivasan and Y. Ni, Adv. Mater., 2000, 12, 62.
4 B. Sakintuna and Y. Yu¨ru¨m, Ind. Eng. Chem. Res., 2005, 44, 2893.
5 O. Tanaike, H. Hatori, Y. Yamada, S. Shiraishi and A. Oya,
Carbon, 2003, 41, 1759.
6 M. Yamazaki, M. Kayama, K. Ikeda, T. Alii and S. Ichihara,
Carbon, 2004, 42, 1641.
7 N. Tonanon, A. Siyasukh, W. Tanthapanichakoon, H. Nishihara,
S. R. Mukai and H. Tamon, Carbon, 2005, 43, 525; R. W. Pekala,
J. Mater. Sci., 1989, 24, 3221.
8 M. Kijima, T. Oda, T. Yamazaki, Y. Tazaki and J. Nakamura,
Chem. Lett., 2006, 35, 844; M. Kijima, H. Tanimoto, H. Shirakawa,
A. Oya, T.-T. Liang and Y. Yamada, Carbon, 2001, 39, 297.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4289–4296 | 4295

View Article Online
9 M. Kijima, H. Tanimoto, K. Takakura, D. Fujiya, Y. Ayuta and
K. Matsuishi, Carbon, 2007, 45, 594; M. Kijima, D. Fujiya, T. Oda
and M. Ito, J. Therm. Anal. Calorim., 2005, 81, 549.
10 M. Kijima, H. Tanimoto and H. Shirakawa, Synth. Met., 2001,
119, 353.
15 T. Heverer and H.-J. Stan, Anal. Chim. Acta, 1997, 341, 21.
16 D. L. Trumbo and C. S. Marvel, J. Polym. Sci., 1986, A24,
2311.
17 X.-M. Liu, C. He, X.-T. Hao, L.-W. Tan, Y. Li and K. S. Ong,
Macromolecules, 2004, 37, 5965.
11 N. Setoyama, T. Suzuki and K. Kaneko, Carbon, 1998, 36, 1459;
K. Kaneko, C. Ishii, H. Kanoh, Y. Hanazawa, N. Setoyama and
T. Suzuki, Adv. Colloid Interface Sci., 1998, 76–77, 295.
12 D. Dollimore and G. R. Heal, J. Colloid Interface Sci., 1970, 33, 508.
13 S. Takahashi, Y. Kuroyama, K. Sonogashira and N. Hagihara,
Synthesis, 1980, 627.
18 M. Noroni and J. L. Moigne, Macromolecules, 1994, 27, 562.
19 S. Takahashi, Y. Kuroyama, K. Sonogashira and N. Hagihara,
Synthesis, 1980, 627.
20 A. Hay, J. Org. Chem., 1960, 25, 1275.
21 M. Ha¨ussler, R. Zheng, J. W. Y. Lam, H. Tong, H. Dong and
B. Z. Tang, J. Phys. Chem. B, 2004, 108, 10645.
22 J. Ozaki, M. Ito, H. Fukazawa, T. Yamanobe, M. Hanaya and
A. Oya, J. Anal. Appl. Pyrolysis, 2006, 77, 56.
14 Y. Morita, A. Kashiwagi and K. Nakasuji, J. Org. Chem., 1997, 62,
7464.
4296 | J. Mater. Chem., 2007, 17, 4289–4296
This journal is ß The Royal Society of Chemistry 2007