, 2003, 13(3), 148–149
Spontaneous rearrangement of hydrogen bonding in a crystalline state
Reiko Kuroda* and Yoshitane Imai
Department of Life Sciences, Graduate School of Arts and Sciences, The University of Tokyo, 3-8-1 Komaba, Meguro-ku, Tokyo 153-
8902; Kuroda Chiromorphology Project, ERATO, Japan Science and Technology Corporation, 4-7-6 Komaba, Tokyo 153-0041,
Japan. Fax: +81 3 5454 6600; e-mail: ckuroda@mail.ecc.u-tokyo.ac.jp
10.1070/MC2003v013n03ABEH001805
Spontaneous rearrangement of hydrogen bonding in a crystalline phase was observed in molecular crystals, which included
sublimable benzoquinones.
Chirality is a fundamental aspect of nature and is expressed at
all levels, either microscopic or macroscopic, and either animate
or inanimate. It is particularly important in life as the biological
world is homochiral. We have been studying molecular recogni-
tion,1 chirospectroscopy2 and supramolecular chirogenesis3 in a
crystalline state, where chiral discrimination energy is orders of
magnitudes larger as compared with a solution state. In many
cases, hydrogen bonding plays a key role in determining whether
a chiral or racemic crystal will form. For example, structures of
the historically famous Pasteur’s chiral and Scacchi’s racemic
sodium ammonium tartrate crystals that we have determined4
contain different numbers of water of crystallization and sub-
stantial hydrogen bonding among tartrate ions, water of crystal-
lization and ammonium ions. Hydroxyl hydrogen atoms of a
tartrate ion adopt different orientations in the chiral and racemic
crystals in order to form two intramolecular hydrogen bonds in
the chiral crystal but one intra- and one intermolecular hydro-
gen bonding in the racemic crystal.4 Here we report an interest-
ing case where this important hydrogen bonding is spontaneously
rearranged in a solid state.
O
OH
OH
2 rac-
2
,
,
O
BN
BQ
toluene
Crystal I
exposed to air
spontaneously
OH
OH
rac-
(BN)
Crystal of rac-BN
Scheme 1
Previously, we reported5 the formation of black three-com-
ponent inclusion crystals (crystal I) with a 2:1:2 stoichiometry
of rac-BN: BQ: toluene, where rac-BN = racemic bis-β-naphthol
and BQ = benzoquinone. In a crystal of I, which has the space
group P1 (#2), there is substantial hydrogen bonding involving
hydroxyl groups, as in the case of sodium ammonium tartrate.
R-BN, BQ and S-BN form a triplet structure, where BQ is
sandwiched in parallel with one of the two naphthyl ring planes
of R-BN and S-BN, along the crystal axis c. The π–π stacking
gives rise to the black colour. The two carbonyl oxygen atoms
of BQ each form two hydrogen bonds with the OH of the non-
parallel naphthyl ring of R- and S-BN within the triplet. An
additional hydrogen bond is formed between the carbonyl oxygen
of BQ and the OH of a near-parallel naphthyl ring, which
belongs to the neighbouring triplet. In this manner, a continuous
hydrogen-bonded column along the crystal axis a is formed. Two
toluene molecules, which are related by an inversion centre, are
included in a continuous channel-like cavity along the axis c
and fixed in place by two weak C–H···π interactions per mole-
cule (see Figures 1 and 2 in ref. 5).
When the black crystals are exposed to air, the crystal colour
changes from black to white within one day (in the case of
microcrystallites) or a few days (single crystals). The colour
change was monitored by X-ray powder diffractometry (Figure 1).
Samples were softly ground, and 20 mg of samples were mounted
on a sample plate. Diffraction data were recorded on a Rigaku
Multi Flex X-ray diffractometer with graphite-monochromated
CuKα1 radiation (l = 1.5418 Å), repeatedly after an appropriate
time of exposure of the sample to air. The scan was made in the
range 3° £ 2q ³ 35° in 0.02° steps and recorded over a total of
11 min. It might be expected that the crystals would deteriorate
to become amorphous; however, as is seen in Figure 1, the
peaks corresponding to the black inclusion crystals (black arrows)
decreased in intensity, while new peaks (white arrows) appeared
[Figure 1(b) and (c)]. Eventually, the peaks corresponding to the
black crystals disappeared completely [Figure 1(d)]. Surprisingly,
the final powder diffraction pattern is identical to that of crys-
talline rac-BN. No intermediate peaks are observed (Scheme 1).
rac-BN crystallises in the space group Iba2; hydrogen bonds
are formed between the hydroxyl groups of 21-screw related
BN molecules, thus forming a homochiral column along the
axis c.6,7 During the process of toluene and BQ loss, hydrogen
bonds between BN and BQ are severed and BN molecules spon-
taneously rearrange themselves to form new hydrogen bonds
among themselves. When the crystals are kept under both toluene
and BQ vapour, they retain the black colour. Intermediate dif-
fraction patterns are compared for the two cases where a crystal
of I is placed under only toluene vapour or only BQ vapour.
When placed under BQ vapour, the crystals rapidly decomposed,
as is expected from their channel-like cavity structure. Interest-
ingly, three peaks at 2q = 9.2, 13.1 and 18.3° disappeared rather
quickly and the peaks around 2q = 12.5~13.8° remained for a
long time. The order is opposite under an atmosphere of toluene;
peaks around 2q = 12.5~13.8° disappeared quickly, whereas the
three peaks (2q = 9.2, 13.1 and 18.3°) disappeared more slowly.
Finally, the crystals turned to rac-BN in both cases. This may
give some clue as to the mechanism of hydrogen bond re-
arrangement.
This interesting spontaneous rearrangement of a hydrogen
bond network initiated by the loss of volatile compounds from
crystals was first reported for the two-component inclusion crys-
tals consisting of 1:1.5 rac-BN: BQ (Crystal II) or 1:1 rac-BN:
BQ (Crystal III).8 When crystals II or III were exposed to air,
they lost their strong red colour and became white after a while.
We monitored the colour change of II by X-ray diffraction as
shown in Figure 1. The X-ray powder diffraction unambiguously
demonstrated that these two-component crystals also reverted to
the starting rac-BN crystals without going through an intermediate
amorphous phase.
These results suggest that the diffusion of molecules in a solid
state can be considerable, and hydroxyl groups can easily change
hydrogen-bonding partners in solid systems, where volatile or
sublimable compounds are involved. The detailed mechanism
of the diffusion process and hydrogen bond rearrangement is
currently being studied.
– 148 –
Delaby et al.
