Microscope and DSC study of polymorphism of 1,1′-(1,12-Dodecanediyl)bisthymine

Article abstract of DOI:10.1246/cl.2002.1122

1,1′-(1,12-Dodecanediyl)bisthymine exhibited two distinct types of crystals. One of them was spontaneously transformed into the other under heat treatment. This phenomenon was monitored by DSC and observed under polarizing microscope.

Full text of DOI:10.1246/cl.2002.1122

1122
Chemistry Letters 2002
Microscope and DSC Study of Polymorphism of 1,10-(1,12-Dodecanediyl)bisthymine
Toshio Itahara,ꢀ Junpei Kanda, Shuji Ikeda, and Takehiko Ueda
Department of Bioengineering, Faculty of Engineering, Kagoshima University, Korimoto, Kagoshima 890-0065
(Received July 4, 2002; CL-020554)
1,10-(1,12-Dodecanediyl)bisthymine exhibited two distinct
types of crystals. One of them was spontaneously transformed
into the other under heat treatment. This phenomenon was
monitored by DSCand observed under polarizing microscope.
the crystals were performed.5 Figure 1 shows the DSC
thermogram of 1a and 1b. The crystal 1a had a melting transition
at 189 ꢁC (ÁH 105 J/g) as shown in Figure 1-(A). On the other
hand, Figure 1-(B) showed three endothermic transitions at
125 ꢁC (ÁH 12 J/g), 150 ꢁC (ÁH 56 J/g), and 189 ꢁC (ÁH 52 J/g)
upon the heating of the crystal 1b at rate of 5 ꢁC/min to 200 ꢁC.
Furthermore, a broad exothermic phase transition was found in
150–170 ꢁC (ÁH À48 J/g). The values of ÁH at 189 ꢁC were 42 J/
g at rate of 10 ꢁC/min and 28 J/g at rate of 15 ꢁC/min, and therefore
were dependent on the heating rate. A polarizing microscopy
observation revealed that 1b once melted at 150 ꢁC, and then
experienced a slow transformation into a new crystal (1c). When
1b was heated to 175 ꢁC (Figure 1-(C)) and then cooled to room
temperature, the new crystal (1c) was isolated. Figure 2 shows the
polarizing microscope picture at the formation of 1c at 160 ꢁC
after the melting of 1b. The crystal 1c had a simple thermal
behavior with only single melting transition temperature at
188 ꢁC (ÁH 103 J/g) (Figure 1-(D)), and the IR spectrum of 1c in
KBr was essentially similar to that of 1a. Thus it was concluded
that the crystal 1c was identical to 1a. Furthermore, on the basis of
the NMR spectra of 1a and 1b in CDCl3, it was confirmed that
these crystals did not include the solvents used for recrystalliza-
tion. This suggests that the different behavior of 1a and 1b was not
due to solvent inclusion in the crystal.
Increasing interest is being shown in the role of hydrogen
bond in relation to the development of the crystal engineering that
aims at an intended molecular network formation in crystal.1
Hydrogen bond also plays an important role in the structure
formation of molecular assemblies in nature. It is well known that
nucleobase forms the hydrogen-bonded assemblies.2 A dissocia-
tion and recombination of the hydrogen bond is of interest in
connection with not only the crystal engineering1 but also a
diversity of biological function. However, the detailed research
has little been carried out about the dissociation and recombina-
tion of the hydrogen bond in solid phase. This paper describes a
polymorphism of 1,10-(1,12-dodecanediyl)bisthymine (1) that
may be related to the dissociation and recombination of the
hydrogen bond.
A mixture of thymine and 1,12-dibromododecane in the
presence of t-BuOK in DMF was stirred at room temperature for
15 h. The resulting mixture was evaporated to give a residue,
which was submitted to chromatography over silica gel. Elution
with ethyl acetate gave 1-(12-bromododecyl)thymine (21%).3
Further elution with a mixture of ethyl acetate and methanol
(9 : 1) afforded a solid mass, which was triturated with ethyl
acetate to give the compound (1) (mp 148–150 ꢁC) in 13% yield.
When 1 was recrystallized from methanol, it precipitated as a
crystal (1a), which melted at 188–189 ꢁC. On the other hand, a
solution of 1 in benzene was allowed to stand at room temperature
to give a crystal (1b), which melted at 149–150 ꢁC.4 Even after the
formation of different crystals, molecular structure of them was
identical because the crystals 1a and 1b dissolved together in
CDCl3 showed exactly the same NMR chemical shifts. The
different crystal formation was also supported by IR spectra of the
crystals in KBr. The IR spectrum of 1a showed two broad bands at
3170 and 3040 cmÀ1 in the range of 3000–3600 cmÀ1 andÀ11b in
KBr also displayed two broad bands at 3162 and 3032 cm . In
Figure 1. DSCthermogram of 1, obtained upon heating
at rate of 5 ꢁC/min. (A) The crystal 1a upon heating to
200 ꢁC. (B) The crystal 1b upon heating to 200 ꢁC. (C )
The crystal 1b upon the heating to 175 ꢁC, giving rise to
the crystal 1c. (D) The crystal 1c upon heating to 200 ꢁC.
contrast, 1a displayed two carbonyl bands at1691 and 1666 cmÀ1
,
while four carbonyl bands at 1709, 1689, 1662, and 1641 cmÀ1
were observed for 1b.
In an effort to determine the effect of the polymethylene
chains upon the thermal behaviors of bisthymines, the DSC
measurements of 2 and 36 were performed. The compounds 2 and
To estimate the difference in energy for intermolecular
interaction, differential scanning calorimeter (DSC) analyses of
Copyright Ó 2002 The Chemical Society of Japan

Chemistry Letters 2002
1123
crystals between 1a and 1b was not only monitored by DSCbut
also observed under polarizing microscope. The reports that
described such melt and recrystallization of crystals in solid phase
under heat treatment are little known.
The thermal transformation of the crystals from 1b to 1a may
be explained in terms of the dissociation and recombination of the
hydrogen bond in solid phase. Based on the difference in the
melting point, we supposed that 1b was a crystal in which most of
thymine moiety formed intramolecular hydrogen bond. Intramo-
lecular hydrogen bond is expected to dissociate at lower
temperature than the intermolecular hydrogen bond because of
less favorable bond length and/or bondangles.3;10 After the
dissociation of the intramolecular hydrogen bond, the inter-
molecular hydrogen bond might have formed at temperatures
raging from 150 ꢁC to 170 ꢁC where we observed the broad
endothermic transition. These speculations will be supported by
further investigation that is now in progress.
(A)
(B)
Figure 2. Optical polarizing micrograph of the formation of
the crystal 1c at 160 ꢁC after the melting of 1b. (A) Beginning
of the formation of 1c at 2 min after the melting of 1b. (B)
Further growing of 1c at 10 min after the melting of 1b.
3 displayed almost single melting transitions at 182 ꢁC
(ÁH 109 J/g) and 251 ꢁC (ÁH 110 J/g), respectively. These
thermal behaviors were similar to that of the crystal 1a. However,
attempts to prepare crystals of 2 and 3 similar to the crystal 1b
were unsuccessful. The reason why only 1 can exhibit the crystal
transformation was of our next interest, especially in the
viewpoint that the formation of the crystal 1b was possibly
correlated with the length in the polymethylene chains.
The authors thank to Professor Isao Saito and Dr. Akimitsu
Okamoto (Kyoto University, Faculty of Engineering) for mass
spectrometric analysis.
References and Notes
1
‘‘Comprehensive Supramolecular Chemistry,’’ ed. by D. D.
MacNicol, F. Toda, and R. Bishop, Elsevier Science, Oxford
(1996), Vol. 6.
The X-ray analysis of 1,10-trimethylenebisthymine (4) has
demonstrated the formation of an intramolecular stacking
between the thymine rings and intermolecular hydrogen bonds.7
A similar intermolecular hydrogen bond structure has been
reported on the X-ray analysis of 1-methylthymine.8 Further-
more, the X-ray analyses of 1-alkylthymines having long alkyl
chains have made clear the molecular packing structures.9
Therefore, it seems reasonable to assume that the crystals 1a
and/or 1b can hold a certain intermolecular hydrogen bond
structure, similar to those of 47 and 1-alkylthymines.8;9 On the
other hand, Kuroda et al. have reported the formation of an
intramolecular hydrogen bond structure of bisuridines connected
together through long alkyl spacers in chloroform.10 We also have
reported a similar result showing the intramolecular hydrogen
bond of 9-[12-(thymin-1-yl)dodecyl]adenine.3 Taking these facts
into consideration, the crystal 1a or 1b is reasonably expected to
have intramolecular hydrogen bond structure.
2
W. Saenger, ‘‘Principles of Nucleic Acid Structure,’’
Springer, New York (1984).
3
4
T. Itahara, Bull.Chem.Soc.Jpn. , 75, 285 (2002).
The crystal 1b: 1H NMR (CDCl3) ꢀ 8.86 (s, 2H, NH), 6.97 (s,
2H, Thy-6), 3.69 (t, 4H, J ¼ 7:2 Hz), 1.92 (s, 6H, Me), 1.67
(broad quintet, 4H, J ¼ 7 Hz), 1.35–1.2 (m, 16H); 13CNMR
(CDCl3) ꢀ 164.20, 150.86, 140.38, 110.56, 48.46, 29.34,
29.33, 29.07, 29.06, 26.33, 12.35. HRFABMS m/z Calcd for
C22H34N4O4 (M+1) 419.2658, found 419.2657. Found: C,
63.39; H, 8.29; N, 13.21%. Calcd for C22H34N4O4: C, 63.13;
H, 8.19; N, 13.39%.
DSCmeasurements were carried out with a Shimadzu DSC-
60. Polarizing microscopy observations were performed
under a Nikon Eclipse E600 POL equipped with a hot stage
(Tokai Hit ThermoPlate).
T. Itahara, Bull.Chem.Soc.Jpn. , 70, 2239 (1997).
J. K. Frank and I. C. Paul, J.Am.Chem.Soc. , 95, 2324 (1973).
K. Hoogsteen, Acta Crystallogr., 16, 28 (1963).
a) N. Tohnai, Y. Inaki, M. Miyata, N. Yasui, E. Mochizuki,
and Y. Kai, Bull.Chem.Soc.Jpn. , 72, 851 (1999). b) N.
Tohnai, Y. Inaki, M. Miyata, N. Yasui, E. Mochizuki, and Y.
Kai, Bull.Chem.Soc.Jpn. , 72, 1143 (1999). c) E. Mochizuki,
N. Yasui, Y. Kai, Y. Inaki, N. Tohnai, and M. Miyata, Bull.
Chem.Soc.Jpn. , 73, 1035 (2000).
5
6
7
8
9
10 Y. Kuroda, J. M. Lintuluoto, and H. Ogoshi, J.Chem.Soc,.
Perkin Trans.2 , 1997, 333.
11 The hydrogen bond between the carbonyl group at 2-position
and the hydrogen atom at 6-position is shown on the X-ray
analyses of 4 and 1-methylthymine.7;8
In earlier papers9;12 dealing with the thermal analyses of 1-
alkylthymines having long alkyl chains, a pair of endo- and
exothermic transitions similar to the thermal behavior of 1b was
found and was discussed from the standpoint of the molecular
packing.9 On the other hand, the thermal transformation of the
12 J. Michas, C. M. Paleos, A. Skoulios, and P. Weber, Mol.
Cryst.Liq.Cryst. , 239, 245 (1994).