Single-crystal structure analysis of a novel aryl phosphinate diglycidyl ether

Article abstract of DOI:10.1107/S010876819900018X

The crystal structure of 10-[2,5-bis(2,3-epoxy-1-propoxy)phenyl]-9-oxa-10-phosphaphenanthren-10-one has been studied by single-crystal X-ray diffraction. The unit cell of C₂₄H₂₁O₆P, M_r = 436.4, is triclinic, P1, with a = 8.507 (3), b = 10.613 (4), c = 12.457 (3) A, α = 80.05 (3), β = 71.38 (2), γ = 76.69 (3)°, V = 1031.1 (6) A³, Z = 2, D_x = 1.406 Mg m^-3 and μ(Mo Kα) = 0.17 mm^-1. The final R (wR) is 0.063 (0.057) {w = 1/[σ²(F) + 0.0004F²]} for 3619 unique reflections measured at 295 K. The aryl phosphinate group bonded to the central phenyl ring comes close to one of the two glycidyl ether groups, the epoxide ring of which is ordered. The epoxide ring far from the aryl phosphinate group is disordered. The NMR chemical shifts of the protons of the glycidyl ether group close to the aryl phosphinate group are reduced by the 'ring-current effect'.

Full text of DOI:10.1107/S010876819900018X

525
Acta Cryst. (1999). B55, 525±529
Single-crystal structure analysis of a novel aryl phosphinate diglycidyl ether
Ching-Sheng Cho,a* Wen-Bin Liaub and Leo-Wang Chena
aDepartment of Chemical Engineering, National Taiwan University, Taiwan, and bInstitute of Materials Science and
Engineering, National Taiwan University, Taiwan. E-mail: chlo@ms.cc.ntu.edu.tw
(Received 4 March 1998; accepted 5 January 1999)
Abstract
pendent groups to improve the ¯ame resistance are
scarce.
The crystal structure of 10-[2,5-bis(2,3-epoxy-1-prop-
oxy)phenyl]-9-oxa-10-phosphaphenanthren-10-one has
been studied by single-crystal X-ray diffraction. The
unit cell of C24H21O6P, Mr = 436.4, is triclinic, P1, with
a = 8.507 (3), b = 10.613 (4), c = 12.457 (3) A, ꢀ =
Recently, a novel aryl phosphinate diglycidyl ether,
10-[2,5-bis(2,3-epoxy-1-propoxy)phenyl]-9-oxa-10-phos-
phaphenanthren-10-one (DHQEP; Fig. 1), was synthe-
sized in which a cyclic organophosphorus compound was
incorporated into the molecular structure of the epoxy
ether and grafted onto the main chain. This aryl phos-
phinate epoxy ether is expected to be highly effective in
increasing the ¯ame resistance of the corresponding
epoxy resin. During the determination of the structure
of DHQEP by 1H- and 13C-NMR, it was found that the
chemical shifts of the protons of the glycidyl ether group
at one end of DHQEP were similar to those observed in
the glycidyl ether group of the diglycidyl ether of
bisphenol A (DGEBA, Fig. 1). However, the chemical
shifts of the protons of the glycidyl ether group at the
other end of DHQEP were quite different from those
Å
Ê
3
ꢀ
Ê
80.05 (3), ꢁ = 71.38 (2), ꢂ = 76.69 (3) , V = 1031.1 (6) A ,
Z = 2, Dx = 1.406 Mg m 3 and ꢃ(Mo Kꢀ) = 0.17 mm
.
1
The ®nal R (wR) is 0.063 (0.057) {w = 1/[ꢄ2(F) +
0.0004F2]} for 3619 unique re¯ections measured at
295 K. The aryl phosphinate group bonded to the
central phenyl ring comes close to one of the two
glycidyl ether groups, the epoxide ring of which is
ordered. The epoxide ring far from the aryl phosphinate
group is disordered. The NMR chemical shifts of the
protons of the glycidyl ether group close to the aryl
phosphinate group are reduced by the `ring-current
effect'.
1. Introduction
Despite their comparatively high cost, epoxy resins have
achieved signi®cant importance as industrial materials
with applications as surface coatings, adhesives, tooling
compounds, matrices in composites and in the encap-
sulation of electronic components (Lee & Neville, 1972;
Lubin, 1982). Epoxy resins can be cured with amines or
anhydrides and the cross linking of epoxy resins may be
carried out either through the epoxy groups or the
hydroxyl groups. Although epoxy resins have many
attractive properties, such as low shrinkage on curing,
good alkali resistance, and weather resistance, the
¯ammability of epoxy resins is a disadvantage in appli-
cations requiring high ¯ame resistance. Thus, modifying
the structures of epoxy resins to improve their ¯ame
resistance has recently received increasing attention.
Introduction of certain phosphorus groups, such as
phosphate (Flury et al., 1996), phosphonate (Liu et al.,
1996), phosphinate (Welch & Paxton, 1968; Vogt et al.,
1968), cyclic phosphine oxide (Shau & Wang, 1996),
phosphine oxide and thioxophosphine oxide (Gentzkow
et al., 1991), into the backbone of epoxy resins to
enhance the ¯ame resistance has been reported.
However, reports of the modi®cation of the structures of
epoxy resins by introducing phosphorus groups as
Fig. 1. The synthesis of DHQEP and the schematic structure of
DGEBA.
# 1999 International Union of Crystallography
Printed in Great Britain ± all rights reserved
Acta Crystallographica Section B
ISSN 0108-7681 # 1999

526
A NOVEL ARYL PHOSPHINATE DIGLYCIDYL ETHER
observed in the glycidyl ether group of DGEBA. One
possible cause of the unusual chemical shifts is the cyclic
organophosphorus group grafted onto the main chain.
Table 1. Experimental details
Crystal data
Chemical formula
C24H21O6P
436.4
The relative positions of the glycidyl ether groups and Chemical formula weight
Cell setting
Space group
Triclinic
P1
the pendent aryl phosphinate group is the key point.
Thus, a single-crystal structure analysis was performed
in order to determine the conformation of DHQEP.
Ê
a (A)
8.507 (3)
10.613 (4)
12.457 (3)
80.05 (3)
71.38 (2)
76.69 (3)
1031.1 (6)
2
Ê
b (A)
Ê
c (A)
ꢀ ꢁꢀ†
ꢁ ꢁꢀ†
ꢂ ꢁꢀ†
2. Experimental
3
Ê
V (A )
Z
Dx (Mg m
DHQEP was synthesized by the reaction of a cyclic
organophosphorus compound with 1,4-benzoquinone,
followed by glycidyl etheri®cation with excess
epichlorohydrin using sodium methoxide as a catalyst
(Fig. 1). The chemical structures of the aryl phosphinate
diol (DHQ) and DHQEP were con®rmed by elemental
3
)
Radiation type
1.406
Mo Kꢀ
0.71073
Ê
Wavelength (A)
No. of re¯ections for cell para-
meters
25
ꢅ range (ꢀ)
7.43±10.94
0.17
295
1
ꢃ (mm
)
1
analysis and mass, FTIR and H-, 13C- and 31P-NMR
Temperature (K)
Crystal form
Crystal size (mm)
Crystal color
spectroscopy. The 31P-NMR spectrum of DHQEP
showed only one sharp peak at 19.1286 p.p.m. Differ-
ential scanning calorimetry (DSC) thermograms were
recorded using a Perkin±Elmer DSC-7 calorimeter at a
Block
0.50 Â 0.38 Â 0.20
Colorless
Data collection
Diffractometer
Data collection method
Absorption correction
Tmin
1
heating rate of 10 K min under a nitrogen atmos-
Enraf±Nonius CAD-4
ꢅ/2ꢅ scans
Empirical
0.9178
0.9528
3632
phere. The DSC results showed only one sharp melting
peak at 430.6 K for the ®nal product, DHQEP, in the
temperature range 323 to 473 K. IR spectra were
obtained using a Bio-Rad FTS-40 Fourier transform
infrared spectrometer and KBr pellets. The FTIR
spectrum of the ®nal product showed the peak of the
epoxy group at 916 cm 1. Elemental analyses were
determined by digestion treatment and calorimetry
Tmax
No. of measured re¯ections
No. of independent re¯ections
No. of observed re¯ections
3619
2634
Criterion for observed re¯ections Inet > 2ꢄ(Inet
)
Rint
ꢅmax (ꢀ)
Range of h, k, l
0.010
24.92
9 ! h ! 10
0 ! k ! 12
14 ! l ! 14
3
using a Carlo±Erba EA-106 analyzer with the following
results for DHQEP: C24H21O6P (436.4 g mol
1
)
observed: C 65.85, H 4.80, P 6.87. These values are close
to the calculated values: C 66.05, H 4.85, P 7.10. The
elemental analysis results for DHQ are C18H13O4P
(234.3 g mol 1) observed: C 66.60, H 4.02, P 9.54, close
to the calculated values: C 66.67, H 4.04, P 9.56. Mass
spectra were collected by the electrospray technique
using a VG Platform mass spectrometer. The base peak
No. of standard re¯ections
Frequency of standard re¯ections Every 60 min
Intensity decay (%)
0
Re®nement
Re®nement on
R
wR
F
0.063
0.057
1.48
1
of DHQEP was M+H+ = 437.1 m/e. H-, 13C-, and 31P-
S
No. of re¯ections used in re®ne-
ment
No. of parameters used
Weighting scheme
NMR spectra were recorded using a Bruker Avance-300
spectrometer with DMSO-d6 or CDCl3 as solvents,
working at 300, 75, and 121 MHz, respectively.
3619
299
1/[ꢄ2(F) + 0.0004F2]
0.010
4.30
The single crystal used for X-ray analysis was
prepared by recrystallization from a very dilute
DHQEP/acetone solution in a tube exposed in a closed
vessel to a saturated vapor of ethyl ether at room
temperature for 1 d. Details of the X-ray experiment are
given in Table 1.² Atomic scattering factors used for the
least-squares re®nement were those listed in Interna-
tional Tables for X-ray Crystallography (1974). All
calculations were performed on a VAX workstation
ꢁÁ=ꢄ†
max
3
Ê
Ê
Áꢆmax (e A
)
3
Áꢆmin (e A
)
2.50
Larson (1970)
62 (21) Â 101
Extinction method
Extinction coef®cient
Computer programs
Data collection and cell re®ne-
ment
CAD-4-EXPRESS (Enraf±
Nonius, 1994; Gabe et al., 1989)
using the program NRCSDP (Gabe et al., 1989). The
re®ned coordinates and Ueq values for the heavy atoms
are listed in Table 2. The molecular structure of DHQEP
is shown in Fig. 2.
² Supplementary data for this paper are available from the IUCr
electronic archives (Reference: OA0018). Services for accessing these
data are described at the back of the journal.

CHING-SHENG CHO, WEN-BIN LIAU AND LEO-WANG CHEN
527
3. Results and discussion
3.1. X-ray diffraction
Table 2. Fractional atomic coordinates and equivalent
Ê 2
isotropic displacement parameters (A )
Ueq ˆ ꢁ1=3†ÆiÆjUijaiajai:aj:
The X-ray analysis revealed disorder at one end of the
molecule (see Fig. 3). Similar disorder of one epoxide
ring of DGEBA has been reported (Flippen-Anderson
& Gilardi, 1981). The disorder affects the epoxide ring
furthest from the pendent aryl phosphinate group and
consists of two alternative orientations for the epoxide
ring connected to C22. However, both alternatives have
one atomic position in common, O6. One possible
position of the epoxide group, labeled E, consists of
C23ÐC24ÐO6 while the other, labeled E0, consists of
C230ÐC240ÐO6 (see Fig. 3). The occupancies for the
C23 and C24 atomic sites re®ned to 0.6. The atoms at the
common position are different from those reported by
Flippen-Anderson & Gilardi (1981). For DGEBA, the
common positions for the two alternatives are occupied
partially by C and partially by O and are approximately
equally favored. Thus, several different possible disor-
dered epoxide-ring structures were tested. However, the
re®nements converged at larger R and wR. Bond lengths
and angles are listed in Table 3. Distances and angles in
and around the epoxide rings are similar to those
observed in DGEBA (Flippen-Anderson & Gilardi,
1981). From a conformation point of view E0 is more
crowded than E and has shorter bond distances and
intramolecular interatomic distances. The ordered
epoxide ring below the pendent aryl phosphinate group
is more crowded than E and less crowded than E0.
However, the difference is not signi®cant.
x
y
z
Ueq
P
O1
O2
O3
O4
O5
O6
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C11
C12
C13
C14
C15
C16
C17
C18
C19
C20
C21
C22
C23
C24
C230
C240
0.91830 (7)
0.9139 (2)
1.01965 (18)
0.62925 (19)
0.4335 (2)
0.6501 (3)
0.3526 (3)
0.9910 (2)
0.9979 (3)
1.0522 (4)
1.0977 (3)
1.0911 (3)
1.0400 (2)
1.0431 (2)
1.1068 (3)
1.1138 (3)
1.0608 (3)
0.9971 (3)
0.9875 (2)
0.7009 (3)
0.5772 (3)
0.4131 (3)
0.3723 (3)
0.4916 (3)
0.6568 (3)
0.5078 (3)
0.5989 (4)
0.5268 (4)
0.5480 (4)
0.4298 (5)
0.4840 (6)
0.4999 (8)
0.3475 (8)
0.72952 (5)
0.88184 (13)
0.65441 (14)
0.88473 (16)
0.50935 (19)
1.0762 (2)
0.3511 (2)
0.69429 (19)
0.5676 (2)
0.5342 (2)
0.6271 (2)
0.7526 (2)
0.78933 (19)
0.92132 (19)
1.0145 (2)
1.1359 (2)
1.1689 (2)
1.0809 (2)
0.96034 (19)
0.7147 (2)
0.7927 (2)
0.7708 (3)
0.6749 (3)
0.5992 (2)
0.6177 (2)
0.9919 (3)
1.0917 (3)
1.1872 (3)
0.4195 (3)
0.3256 (4)
0.2380 (6)
0.3175 (7)
0.2776 (6)
0.42866 (4)
0.41924 (11)
0.50183 (12)
0.34941 (13)
0.70245 (17)
0.13093 (17)
0.93426 (15)
0.28446 (17)
0.26422 (19)
0.1545 (2)
0.0383 (3)
0.0509 (10)
0.0500 (9)
0.0628 (10)
0.0863 (13)
0.0982 (17)
0.0861 (14)
0.0388 (11)
0.0539 (15)
0.0658 (18)
0.0621 (17)
0.0512 (14)
0.0390 (11)
0.0387 (11)
0.0537 (14)
0.0637 (16)
0.0555 (14)
0.0453 (12)
0.0379 (11)
0.0417 (12)
0.0481 (12)
0.0609 (16)
0.0655 (16)
0.0573 (14)
0.0492 (13)
0.0695 (16)
0.0769 (19)
0.097 (23)
0.084 (2)
0.0647 (2)
0.08321 (18)
0.19331 (17)
0.21477 (17)
0.12615 (18)
0.1470 (2)
0.2571 (2)
0.34569 (18)
0.32373 (17)
0.48413 (17)
0.43766 (18)
0.4793 (2)
0.5663 (2)
0.6148 (2)
0.57315 (19)
0.3251 (2)
0.2456 (2)
0.1737 (3)
0.7485 (3)
0.8195 (3)
0.9056 (4)
0.8488 (6)
0.8532 (5)
0.061 (3)
0.092 (4)
0.072 (4)
0.063 (4)
3.2. NMR chemical shifts
at the other end of DHQEP are quite different from
those observed in the glycidyl ether group of DGEBA
(see Table 4). Because of the signi®cant difference in the
chemical shifts of the atoms of the two glycidyl ether
groups two-dimensional NMR techniques were used to
con®rm the assignment of the NMR spectra. Figs. 4 and
The motivation for the single-crystal X-ray analysis of
DHQEP was the observation that the NMR chemical
shifts of the protons of the glycidyl ether group at one
end of DHQEP are similar to those observed for the
glycidyl ether group of DGEBA (Han et al., 1993).
However, the chemical shifts of the glycidyl ether group
Fig. 2. An ORTEP (Johnson, 1976)
drawing of the structure of
DHQEP. Displacement ellipsoids
are shown at the 30% probability
level for non-H atoms.

528
A NOVEL ARYL PHOSPHINATE DIGLYCIDYL ETHER
ꢀ
Ê
Table 3. Selected geometric parameters (A, )
5 show the two-dimensional NMR spectra of DHQEP
from CÐH correlation and COSY experiments,
respectively, and con®rm the assignment of the NMR
spectrum of DHQEP. In Table 4 the NMR chemical
shifts of the protons in DHQEP and DGEBA are
compared. It is obvious that the chemical shifts of H10,
H20, H30, H40, and H50 of DHQEP are signi®cantly
reduced. A possible reason for this is that these atoms
are positioned close to the aryl phosphinate group. An
aromatic molecule can be visualized as a current loop in
which the ꢇ electrons are free to move on a circle
formed by the ꢄ framework. A diamagnetic ring current
is induced when these compounds are subjected to an
external magnetic ®eld Bo and a dipole opposed to Bo
and centered in the middle of the ring is induced by the
diamagnetic ring current. Therefore, protons in the
molecular plane and outside the ring are deshielded.
Conversely, protons in the region above or below the
PÐO1
PÐO2
1.5932 (16)
1.4641 (15)
1.779 (2)
1.790 (2)
1.387 (2)
1.370 (3)
1.412 (3)
1.363 (3)
1.389 (4)
1.381 (3)
1.437 (4)
1.415 (4)
1.447 (6)
1.380 (7)
1.395 (7)
1.395 (3)
1.406 (3)
1.376 (3)
1.375 (4)
1.377 (3)
1.395 (3)
1.478 (3)
C7ÐC8
C7ÐC12
C8ÐC9
C9ÐC10
1.406 (3)
1.390 (3)
1.376 (3)
1.381 (4)
1.371 (3)
1.379 (3)
1.396 (3)
1.396 (3)
1.384 (3)
1.369 (4)
1.375 (4)
1.382 (3)
1.494 (4)
1.399 (4)
1.541 (5)
1.520 (7)
1.415 (6)
0.776 (8)
0.910 (7)
1.008 (8)
1.455 (8)
1.436 (9)
PÐC1
PÐC13
O1ÐC12
O3ÐC14
O3ÐC19
O4ÐC17
O4ÐC22
O5ÐC20
O5ÐC21
O6ÐC23
O6ÐC24
O6ÐC230
O6ÐC240
C1ÐC2
C1ÐC6
C2ÐC3
C3ÐC4
C4ÐC5
C5ÐC6
C6ÐC7
C10ÐC11
C11ÐC12
C13ÐC14
C13ÐC18
C14ÐC15
C15ÐC16
C16ÐC17
C17ÐC18
C19ÐC20
C20ÐC21
C22ÐC23
C22ÐC230
C23ÐC24
C23ÐC230
C23ÐC240
C24ÐC230
C24ÐC240
C230ÐC240
È
plane of the ring are strongly shielded (Gunther, 1980).
Thus, the chemical shifts of H10, H20, H30, H40, and H50
(the atomic labels of the protons are show in Fig. 1) of
DHQEP are signi®cantly reduced by the `ring-current
effect' of the pendent aryl phosphinate group.
O1ÐPÐO2
O1ÐPÐC1
O1ÐPÐC13
O2ÐPÐC1
O2ÐPÐC13
C1ÐPÐC13
PÐO1ÐC12
C14ÐO3ÐC19
C17ÐO4ÐC22
C20ÐO5ÐC21
C23ÐO6ÐC24
C23ÐO6ÐC230
C23ÐO6ÐC240
C24ÐO6ÐC230
C24ÐO6ÐC240
C230ÐO6ÐC240
PÐC1ÐC2
111.96 (9)
103.87 (9)
104.14 (9)
115.46 (9)
112.41 (9)
108.07 (10)
127.02 (13)
118.50 (18)
119.2 (2)
59.49 (19)
59.3 (3)
32.2 (3)
37.8 (3)
41.7 (3)
61.5 (4)
62.3 (4)
117.50 (16)
122.00 (15)
120.51 (19)
120.4 (2)
119.6 (2)
120.8 (2)
121.2 (2)
117.53 (18)
120.62 (18)
121.81 (18)
121.96 (18)
122.35 (17)
115.67 (18)
121.7 (2)
O4ÐC17ÐC16
O4ÐC17ÐC18
C16ÐC17ÐC18
C13ÐC18ÐC17
O3ÐC19ÐC20
O5ÐC20ÐC19
O5ÐC20ÐC21
C19ÐC20ÐC21
O5ÐC21ÐC20
O4ÐC22ÐC23
O4ÐC22ÐC230
C23ÐC22ÐC230
O6ÐC23ÐC22
O6ÐC23ÐC24
O6ÐC23ÐC230
O6ÐC23ÐC240
C22ÐC23ÐC24
C22ÐC23ÐC230
C22ÐC23ÐC240
C24ÐC23ÐC230
C24ÐC23ÐC240
C230ÐC23ÐC240
O6ÐC24ÐC23
O6ÐC24ÐC230
O6ÐC24ÐC240
C23ÐC24ÐC230
C23ÐC24ÐC240
C230ÐC24ÐC240
O6ÐC230ÐC22
O6ÐC230ÐC23
O6ÐC230ÐC24
O6ÐC230ÐC240
C22ÐC230ÐC23
C22ÐC230ÐC24
C22ÐC230ÐC240
C23ÐC230ÐC24
C23ÐC230ÐC240
C24ÐC230ÐC240
O6ÐC240ÐC23
O6ÐC240ÐC24
O6ÐC240ÐC230
C23ÐC240ÐC24
C23ÐC240ÐC230
C24ÐC240ÐC230
115.3 (2)
125.2 (2)
119.5 (2)
120.0 (2)
108.1 (2)
116.8 (3)
62.3 (2)
123.4 (3)
58.26 (19)
98.6 (3)
124.6 (3)
29.3 (3)
114.2 (3)
61.5 (3)
71.4 (6)
69.9 (5)
116.7 (4)
73.9 (6)
169.4 (6)
43.7 (6)
73.9 (5)
116.6 (8)
59.2 (3)
65.6 (5)
57.5 (3)
32.1 (4)
37.0 (3)
68.6 (5)
117.7 (5)
76.4 (6)
72.7 (5)
59.4 (4)
76.8 (6)
169.2 (8)
111.3 (5)
104.3 (8)
34.5 (5)
70.6 (6)
72.3 (5)
61.0 (4)
58.3 (4)
69.2 (5)
28.9 (4)
40.8 (4)
PÐC1ÐC6
C2ÐC1ÐC6
C1ÐC2ÐC3
C2ÐC3ÐC4
C3ÐC4ÐC5
C4ÐC5ÐC6
C1ÐC6ÐC5
C1ÐC6ÐC7
C5ÐC6ÐC7
C6ÐC7ÐC8
C6ÐC7ÐC12
C8ÐC7ÐC12
C7ÐC8ÐC9
C8ÐC9ÐC10
C9ÐC10ÐC11
C10ÐC11ÐC12
O1ÐC12ÐC7
O1ÐC12ÐC11
C7ÐC12ÐC11
PÐC13ÐC14
PÐC13ÐC18
C14ÐC13ÐC18
O3ÐC14ÐC13
O3ÐC14ÐC15
C13ÐC14ÐC15
C14ÐC15ÐC16
C15ÐC16ÐC17
Fig. 3. The two alternative structures of the epoxide ring (solid line: E;
dashed line: E0).
120.4 (2)
119.6 (2)
119.6 (2)
122.51 (17)
114.45 (18)
123.02 (18)
122.03 (16)
118.26 (16)
119.64 (19)
116.71 (19)
123.9 (2)
119.4 (2)
120.2 (2)
121.2 (2)
Fig. 4. The two-dimensional NMR spectrum of DHQEP (CÐH).

CHING-SHENG CHO, WEN-BIN LIAU AND LEO-WANG CHEN
529
1
Table 4. Chemical shifts (p.p.m.) of the glycidyl ether groups in the H-NMR spectra of DHQEP and DGEBA in
CDCl3
H1
H2
H3
H4
H5
H10
H20
H30
H40
H50
DHQEP
DGEBA
4.40
4.16
4.02
4.02
3.43
3.31
2.99
2.87
2.84
2.73
3.73
3.63
2.43
2.45
2.19
University, Taiwan) for collecting the single-crystal
X-ray diffraction data.
References
Enraf±Nonius (1994). CAD-4-EXPRESS. Version 5.1. Enraf±
Nonius, Delft, The Netherlands.
Flippen-Anderson, J. L. & Gilardi, R. (1981). Acta Cryst. B37,
1433±1435.
Flury, P., Mayer, C. W., Scharf, W. & Vanoli, E. (1996). US
Patent 5 506 313.
Gabe, E. J., Le Page, Y., Charland, J. P., Lee, F. L. & White, P. S.
(1989). J. Appl. Cryst. 22, 384±387.
Gentzkow, W. V., Huber, J., Rogler, W. & Wilhelm, D. (1991).
US Patent 5 036 135.
È
Gunther, H. (1980). NMR Spectroscopy. New York: John
Wiley and Sons.
Han, J. L., Hsieh, K. H. & Chiu, W. Y. (1993). J. Appl. Polym.
Sci. 50, 1099±1106.
Johnson, C. K. (1976). ORTEPII. Report ORNL-5138. Oak
Ridge National Laboratory, Tennessee, USA.
Larson, A. C. (1970). Crystallographic Computing, edited by
F. R. Ahmed, S. R. Hall & C. P. Huber, pp. 291±294.
Copenhagen: Munksgaard.
Fig. 5. The two-dimensional NMR spectrum of DHQEP (HÐH).
Lee, H. & Neville, K. (1972). Editors. Handbook of Epoxy
Resins. New York: McGraw-Hill.
Liu, Y. C., Hsiue, G. H., Chiu, Y. S., Jeng, R. J. & Perng, L. H.
(1996). J. Appl. Polym. Sci. 61, 613±621.
Lubin, G. (1982). Editor. Handbook of Composites. New York:
Van Nostrand Reinhold.
Shau, M. D. & Wang, T. S. (1996). J. Polym. Sci. Polym. Chem.
Ed. 34, 387±396.
The authors are indebted to Professor Dr M. K.
Leung (Department of Chemistry, National Taiwan
University, Taiwan) for helpful discussions. We also
thank Dr T. R. Wu (Chung-Shan Institute of Science and
Technology, Taiwan) for collecting the NMR spectra and
Mr G. H. Lee (Taipei Regional Analytical Instrument
Vogt, W., Janssen, P. & Richtzenhain, H. (1968). US Patent 3
399 171.
Center, National Science Council, National Taiwan Welch, F. J. & Paxton, H. J. (1968). US Patent 3 405 075.