Helix-helix interactions - Homochirality and heterochirality

Article abstract of DOI:10.1039/c1ce06081c

The relatively simple molecule 2-amino-4-(thiazolin-2-yl)phenol, 1, as its acetonitrile solvate crystallises such that the lattice can be considered to contain heterochiral sheets of helical, N...HO H-bonded polymers which are linked within their sheets by weaker N...HN H-bonds between adjacent helices of opposite chirality. Weaker interactions of the CH...π type can be discerned as linking helices of the same chirality between these sheets, these interactions being reinforced by weaker interactions still involving the acetonitrile solvent molecules. A similar analysis can be conducted of the helical structures found within the lattice of the related compound 4-(methoxycarbonyl)-2-aminophenol, 2, although here it is more difficult to define the hierarchy of the weak interactions observed.

Full text of DOI:10.1039/c1ce06081c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
CrystEngComm
Cite this: CrystEngComm, 2011, 13, 7207
www.rsc.org/crystengcomm
COMMUNICATION
Helix-helix interactions – homochirality and heterochirality†
ab
a
a
*
ꢀ
Andre De Cian and Jack Harrowfield
Artur R. Stefankiewicz,
Received 22nd August 2011, Accepted 14th October 2011
DOI: 10.1039/c1ce06081c
general.14 In the present instance, the molecule 1 (Fig. 1), 2-amino-4-
(thiazolin-2-yl)phenol, has been found to crystallise, solvated, in
a form where there are both right- and left-handed helical polymer
entities resulting from N/O H-bonding present in the lattice and
where these helices can be regarded as segregated into domains on the
basis of different types of interactions occurring within these
domains. It may be noted in general that the properties of chiral
materials are of increasing interest in numerous sophisticated
domains of science.15
The relatively simple molecule 2-amino-4-(thiazolin-2-yl)phenol, 1,
as its acetonitrile solvate crystallises such that the lattice can be
considered to contain heterochiral sheets of helical, N/HO H-
bonded polymers which are linked within their sheets by weaker N/
HN H-bonds between adjacent helices of opposite chirality. Weaker
interactions of the CH/p type can be discerned as linking helices of
the same chirality between these sheets, these interactions being
reinforced by weaker interactions still involving the acetonitrile
solvent molecules. A similar analysis can be conducted of the helical
structures found within the lattice of the related compound
4-(methoxycarbonyl)-2-aminophenol, 2, although here it is more
difficult to define the hierarchy of the weak interactions observed.
Fig. 1 Molecule 1, studied as its crystalline acetonitrile solvate.
Introduction
Helicity is a common aspect of molecular structure in both the liquid
and solid states, with the composition of the species involved varying
from that as simple as SiO2 to one as complex as DNA.1 While the
issue of which forces give rise to helical structures is of fundamental
importance,2–7 the interactions between helices are also critical in
determining properties as diverse as the aggregation of proteins8 and
the crystallisation of a solid either as a racemic mixture or a racemic
compound.9 They may also be important in mixed synthetic and
natural systems of medical potential.10 Since such association is
usually the result of weak interactions, its analysis must involve
a dissection of forces which are not necessarily easily placed in
a hierarchy.11 While sophisticated tools are available for the analysis
of helix-helix interactions in proteins,12 for example, such systems
involve not only numerous types of weak interaction but also
numerous variations in any particular type of interaction, as well as
being subject to the chirality bias of all natural materials. Hence,
synthetic small molecule crystals, where helical assemblies may be
identified in the lattice,13 are of interest because of the possibility of
establishing the hierarchy of a small number of separate interactions
between helices of both the same chirality and of the opposite, as well
as being of interest for the understanding of chiral interactions in
Results and discussion
Within the lattice of the mono-acetonitrile solvate of 1, the closest
intermolecular contact indicative of an H-bonding interaction is that
ꢁ
between phenolic-O and –N of 2.688(3) A. This interaction leads to
the formation of helical polymer chains where the helix axis lies
parallel to the crystallographic b axis (Fig. 2).
In viewing the lattice down b, the impression gained is that it has
a layered form in which homochiral sheets of 1n helices alternate in
planes parallel to ab (Fig. 3(a)). Acetonitrile molecules lie largely
within these sheets and not between them, although from a view of
the lattice down c (Fig. 3(b)) it is apparent that they lie in what would
otherwise be voids formed through the molecular packing. Thus,
there is clearly a degree of segregation of D and L forms of the helices
within the lattice, though this segregation is incomplete in terms of the
complete lattice since both forms are present.
Within a homochiral sheet, the shortest intermolecular contacts
other than those of the H-bonds are those between the methylene
groups of the thiazoline ring and the aromatic phenyl ring. Since these
are presumably multi-centre CH/p interactions,16 they are difficult
to represent and only the shortest aliphatic-C/aromatic-C contacts
ꢁ
of 3.367(3) A are shown in Fig. 4. Three of the four H-atoms of the
ꢁ
aInstitut de Science et d’Ingeꢀnierie Supramoleꢀculaires, Universiteꢀ de
Strasbourg, 8, alleꢀe Gaspard Monge, 67083 Strasbourg, France
bDepartment of Chemistry, University of Cambridge, Lensfield Road,
Cambridge, CB2 1EW, UK
two methylene groups lie between 3.0 and 3.1 A of aromatic carbons,
a separation which is essentially that of van der Waals contact,16 so
that the interaction energy must be small, although they are inter-
actions which occur with neighbours to each side of a given helix
within the sheet.
† CCDC reference numbers 838207 and 838884. For crystallographic
data in CIF or other electronic format see DOI: 10.1039/c1ce06081c
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 7207–7211 | 7207

View Article Online
Fig. 5 A partial view of two adjacent helices of opposite chirality with
the double H-bonding interactions between anilino-N and phenolic-O
ꢁ
(3.139(3) A) shown as green and white dashed lines.
reciprocal, that is, each can be regarded as double, thus possibly
enhancing their influence. In combination, the two forms of N/HO
H-bonding observed in the lattice can be regarded as defining an
array (Fig. 6) which is a heterochiral sheet of helices lying parallel to
the bc plane. As with the homochiral sheets, the acetonitrile molecules
can also be considered to lie within these sheets.
Fig. 2 Stick representations of one of the right-handed helices formed
(along with the same number of left-handed) by thiazoline-N/phenol-O
H-bonding (dashed lines) (a) perpendicular to the helix axis (which is
parallel to b) and (b) down this axis. H-atoms are not shown but the
phenolic H-atoms lie close to the dashed lines; N/H–O 174.7ꢀ.
A final issue to be considered in the analysis of the lattice structure
of 1$CH3CN is that of the reason for the presence of acetonitrile
molecules. It is indeed difficult to find any evidence for strong, specific
interactions, since all contacts to any of the three atoms in the solvent
ꢁ
molecule are remote. The shortest, 3.266(3) A, is that between
acetonitrile-N and anilino-N and can be described as N/HN
bonding but interestingly it is an interaction which serves to bind the
heterochiral sheets together, since the anilino-N is in the sheet adja-
cent to that which can be considered to « contain » the acetonitrile.
ꢁ
Within a bc (heterochiral) sheet, the shortest contact, of 3.470(3) A, is
that between the central acetonitrile-C and phenolic-O (incipient
Fig. 3 Partial views of the crystal lattice of 1$CH3CN (a) down b and (b)
down c. For clarity, all atoms in the acetonitrile molecules are shown in
pink and one H-bonded 1n helix is shown with all atoms in black.
Fig. 4 A partial view of two adjacent left-handed helices present in
a homochiral sheet (parallel to the ab plane) of the lattice of 1$CH3CN,
ꢁ
showing the shortest methylene-C/aromatic-C contacts (3.367(3) A) as
green and white dashed lines (and the H-bonding contacts within the
helices as black and white dashed lines).
Every helix within a given homochiral sheet has neighbours of
opposite chirality to both sides in the adjacent sheets. The shortest
ꢁ
intermolecular atomic contacts here (3.139(3) A) are between the
Fig. 6 A partial view of a heterochiral (i.e. overall achiral) sheet parallel
to the bc plane within the lattice of 1$CH3CN involving strong (black and
white) and weak (green and white) hydrogen bonds, the former being
considered to define the helices.
anilino-N and phenolic-O and are taken to be indicative of
H-bonding which must be considerably weaker than that within the
helical chains. However, as indicated in Fig. 5, these interactions are
7208 | CrystEngComm, 2011, 13, 7207–7211
This journal is ª The Royal Society of Chemistry 2011

View Article Online
presumably indicative of a CH/p interaction (and not, for example,
a dipole–dipole interaction, since the two dipoles are parallel). These
three interactions, shown in a perspective view in Fig. 7, link helices of
the same chirality in adjacent bc sheets.
The importance of the nature of the substituents on a 2-amino-
phenol ring is nicely illustrated by a comparison of the structure of
1$CH3CN with that of 4-(methoxycarbonyl)-2-aminophenol, 2
obtained by recrytallisation from DCM. The latter forms a crystal
lattice in which the dominant H-bonding motif (that is, the one
ꢁ
involving the shortest OH/N contact, here 2.735(6) A), is typical of
2-aminophenols17 in that it involves the amino and hydroxyl groups
and generates a helical polymer, although it may be noted that with
a substituent as large as a difluorobora-azaindacene (« Bodipy »),18
the H-bonded polymer is not helical. Once again, in the lattice of 2,
helical polymers of alternating chirality may be considered to lie in
sheets (parallel to the bc plane) with inter-helix interactions seemingly
mediated by stacking of the (methoxycarbonyl)phenyl groups
(Fig. 8). In the a direction these sheets stack in such a way that helices
of alternating chirality form sheets parallel to the ab plane, with
interactions involving the methoxycarbonyl substituents in CH3/O
Fig. 7 The bridging contacts, shown as green and white dashed lines, of
an acetonitrile molecule (C atoms pink, N atom blue) between two helices
of the same chirality in different heterochiral sheets parallel to the bc
plane.
ꢁ
contacts (3.488(8) A) (Fig. 9(a)). While the impression thus given is
that heterochiral contacts are the more important in this lattice, there
ꢁ
are in fact homochiral contacts (3.228(8) A), again involving
carbonyl-O but to an aromatic-CH and linking helices along
a direction parallel to the ac diagonal (Fig. 9(b)). As well, there is an
additional interaction of the amino group, apparently an NH/O
ꢁ
(carbonyl) interaction (at 3.125(8) A), linking homochiral helices. It is
difficult to place a quantitative ordering on these various observations
but since the lattice is heterochiral, the numerous close atom contacts
ꢁ
(the shortest being carbonyl-O/C(2) of 3.346(8) A, while C(4)/C
0
ꢁ
(4 ) is 3.500(8) A) involved in the stacking parallel to the bc plane
must presumably outweigh the simple pairs involved otherwise.
Fig. 8 A partial view of two adjacent helices formed by strong H-bonds
(black and white dashed lines) and of opposite chirality, found within the
lattice of unsolvated 4-(methoxycarbonyl)-2-amino phenol, 2, showing
the stacking of the methoxycarbonylphenyl units from separate helices.
Conclusions
Contrary to the visual impression gained from the view of the lattice
of 1$CH3CN down b that it might be considered as being built up
from homochiral sheets of helical H-bond polymers of 1 lying parallel
to the ab plane, if additional H-bonding interactions between the
helical chains are more important than rather long contacts indicative
of CH/p interactions, then the lattice is better regarded as being
built up from heterochiral sheets lying parallel to the bc plane. This
structure allows acetonitrile molecules to form links between the
sheets involving a third and presumably weaker form of H-bonding
along with probably extremely weak interactions, one of which might
be termed electrostatic, the other CH/p. As has long been recog-
nised in the search for weak interactions in crystal lattices,19 there are
many situations where it is difficult to establish whether or not
a particular contact is simply enforced by the existence of others and
thus it is possible than one or the other (or even both) of the non-H-
bonding interactions of the acetonitrile molecule should be ignored.
Nonetheless, at least one attractive interaction must be found to
explain the presence of the solvent!
nucleophilic addition?), although it is perhaps not significantly
ꢁ
different from the contact, 3.478(3) A, between the methyl-C and the
aromatic-C to which the phenolic-O is attached, and which is
Overall, if the simple criterion of the distance of atom/atom
separation for a given pair is taken as an index of the strength of an
attractive interaction, the nature of the lattice of 1$CH3CN can be
readily rationalised. In this case, interactions between helices of
opposite chirality dominate those between helices of the same
chirality, giving rise to a lattice which is overall achiral. While it is
Fig. 9 Partial views of adjacent helices within the lattice of 2, showing
ꢁ
(a) CH3/O contacts (3.488(8) A; pink and white dashed lines) between
helices of opposite chirality and (b) carbonyl-O/HC(aromatic) contacts
ꢁ
(3.228(8) A; pink and white dashed lines) between helices of the same
chirality (here, left-handed).
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 7207–7211 | 7209

View Article Online
Table 1 Crystal data and structure refinement details
observed that chiral (but racemic) compounds crystallise far more
frequently as racemic compounds than as racemic mixtures,1 this
applies to many systems which do not involve extended helices in
their crystal lattices. The present study is nonetheless informative in
that it suggest various ways in which hetero- versus homo-chiral
interactions might be controlled. Thus, methylation of the aromatic
amino group, for example, should serve to block the principal
interactions favoring heterochiral contacts. While numerous struc-
tures of 2-aminophenol derivatives may be found in the Cambridge
data base, nearly all have such elaborate or particular functionali-
sation that no direct comparison with the present systems is possible.
In the case of the relatively simple derivative N(2-hydroxy-5-methyl-
phenyl)-imidazole,24 for example, the lattice contains helical polymer
units due to phenolic-OH interactions with the unalkylated imid-
azole-N and contacts (CH/p) between homochiral helices appear to
be shorter than those between heterochiral, with no obvious contacts
involving the N-atom bound to the phenyl ring, but obviously the
incorporation of this N-atom into an imidazole ring must alter its
properties rather greatly.
Compound
1$CH3CN
2
Chemical formula
M/g molꢂ1
C9H11N2OS/CH3CN
235.31
C8H9NO3
167.16
Orthorhombic
Pbca
10.1980(8)
8.7310(12)
17.664(2)
90.00
90.00
90.00
1572.8(3)
8
1.412
Crystal system
Space group
Monoclinic
P21/c
8.0960(2)
9.8380(2)
15.2620(5)
90.00
103.0690(9)
90.00
1184.11(5)
2
1.320
ꢁ
a/A
ꢁ
b/A
ꢁ
c/A
a/ꢀ
b/ꢀ
g/ꢀ
3
ꢁ
V/A
Z
Dc/g cmꢂ3
m/mmꢂ1
F(000)
0.256
496
0.109
704
173(2)
3185
1774
T/K
Reflections collected
173(2)
5856
Independent reflections
‘‘Observed’’ reflections
3443
2720
868
(I > 2s(I))
Rint
Parameters refined
0.017
146
0.091
119
R1
wR2
S
0.047
0.134
1.046
0.077
0.1672
1.050
Experimental
Synthesis
Crystallography
a) molecule 1. To a solution of 3-amino-4-hydroxybenzoic acid
(1 g, 6.52 mmol) in dry DMF (20 ml) 1-hydroxybenzotriazole (HOBt,
0.88 g, 6.52 mmol) and N,N0- diisopropylcarbodiimide (DIC, 0.82 g,
6.52 mmol) were added under an N2 atmosphere. After 30 min.
stirring, 2-aminoethanethiol hydrochloride (0.74 g, 6.52 mmol) was
added and the mixture was stirred at room temperature under N2 for
24 h. After removal of the solvent, H2O (50 ml) was added to the
residue and the mixture was extracted with DCM (3 ꢁ 50 ml). The
combined organic phases were dried over Na2SO4, filtered and taken
to dryness. The colourless residue was recrystallised from acetonitrile
to give the pure product as a colourless crystals. Yield: 0.31g
(1.57 mmol, 24%). 1H NMR ([D6]DMSO, 400 MHz): d (ppm) ¼ 3.12
(m, 2H), 4.02 (s, 1H), 4.53 (s, 2H), 6.52 (d, J ¼ 8.4, 1H), 6.57 (d, J ¼
6.4 Hz, 1H), 6.71 (s, 1H), 7.58 (t, 1H) 9.46 (s, 1H). 13C NMR ([D6]
DMSO, 100 MHz): d ¼ 54.7, 75.6, 113.2, 116.4, 119.6, 132.3, 137.2,
142.0, 159.2. HR-MS (ES): calcd for C9H13N2OS: m/z ¼ 197.0739,
found m/z ¼ 197.0522 [M + H]+. Microanalysis: calcd (%) for
C9H12N2OS$2MeCN: C 56.09, H 6.52, N 20.13; found C 56.38, H
6.42, N 20.05.
X-Ray diffraction data collection was carried out at 173(2) K on
a Nonius Kappa-CCD diffractometer equipped with an Oxford
Cryosystem liquid N2 device, using Mo-Ka radiation (l ¼ 0.71073
20
ꢁ
A). COLLECT software was used for the data measurement and
DENZO-SMN21 for the processing. The structures were solved by
direct methods using the program SHELXS-97.22 The refinement and
all further calculations were carried out using SHELXL-97.23 The H-
atoms of the OH, NH and NH2 groups were located from Fourier
difference maps and refined isotropically. The other H-atoms were
included in calculated positions and treated as riding atoms using
SHELXL default parameters. The non-H atoms were refined
anisotropically, using weighted full-matrix least-squares on F2
(Table 1).
Acknowledgements
A.R.S would like to express his deep gratitude to Prof. Jean-Marie
Lehn for the opportunity to carry out these studies in his laboratory
and for his precious advice. We also thank Dr. Jack K. Clegg for his
helpful suggestions.
b) molecule 2. Commercial 3-amino-4-hydroxybenzoic acid (1 g,
6.52 mmol) was suspended in methanol (40 ml) under N2. Sulfuric
acid (0.7 ml, 13.1 mmol) was added slowly via syringe, and the
mixture was heated under reflux for 24 h. On cooling, it was
neutralized with saturated NaHCO3 solution to pH ¼ 7 and
extracted with DCM (3 ꢁ 50 ml). The combined organic phases were
dried over Na2SO4, filtered and the solvent was evaporated. A brown
solid was isolated and recrystallized from DCM. The crystals
obtained were suitable for X-ray crystallography. Yield: 0.52g
(3.11mmol, 47%). 1H-NMR (CDCl3, 400 MHz): d (ppm) ¼ 3.86 (s, 3
H, CH3), 6.79 (d, J ¼ 8 Hz, 1H, H2), 7.43 (dd, J ¼ 8 Hz and 1.6 Hz
1H, H1), 7.46 (s, 1H, H5). 13C-NMR (CDCl3, 100 MHz): 52.0, 114.5,
117.8, 122.3, 122.4, 133.9, 148.9, 167.7. HR-MS (ES): calcd for
C8H10N1O3: m/z ¼ 168.0655, found m/z ¼ 168.0661 [M + H]+.
References
1 (a) K. Mislow, Introduction to Stereochemistry, W.A. Benjamin, Inc.,
New York, 1965; Top. Stereochem. 1999, 22, pp. 1–82; (b)
S. F. Mason, Molecular Optical Activity and the Chiral
Discriminations, Cambridge University Press, Cambridge, 1982.
2 J. M. Stolz and R. L. Baldwin, Annu. Rev. Biophys. Biomol. Struct.,
1992, 21, 95–118.
3 J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995, Ch. 9.
4 (a) C. Piguet, G. Bernardinelli and G. Hopfgartner, Chem. Rev., 1997,
97, 2005–2062; (b) C. Piguet, M. Borkovec, J. Hamacek and
K. Zeckert, Coord. Chem. Rev., 2005, 249, 705–726.
5 (a) M. Albrecht, Chem. Soc. Rev., 1998, 27, 281–287; (b) Chem. Rev.
2001, 101, pp. 3457–3498; (c) Top. Curr. Chem. 2004, 248,pp. 105–139.
7210 | CrystEngComm, 2011, 13, 7207–7211
This journal is ª The Royal Society of Chemistry 2011

View Article Online
6 D. L. Caulder and K. N. Raymond, Acc. Chem. Res., 1999, 32, 975–
982.
15 D. N. Beratan, R. Naaman, D. H. Waldeck (Eds) Top. Curr. Chem.
2011, 298, pp. 1–306.
7 K. Maeda and E. Yashima, Top. Curr. Chem., 2006, 265, 47–88.
8 See, for example, (a) W. P. Russ and D. M. Engelmann, J. Mol. Biol.,
2000, 296, 911–919; (b) L. Adamian and J. Liang, J. Mol. Biol., 2001,
311, 891–907, and references therein.
9 P. A. Levkin, V. Yu. Torbeev, D. A. Lenev and R. G. Kostyanovsky,
Top. Stereochem., 2006, 25, 81–134.
10 A. Oleksi, A. G. Blanco, R. Boer, I. Uson, J. Aymami, A. Rodger,
M. J. Hannon and M. Coll, Angew. Chem., Int. Ed., 2006, 45, 1227–
1231.
11 A. Scarso and J. Rebek, Jr, Top. Curr. Chem., 2006, 265, 1–46.
12 A. E. Burba, U. Lehnert, E. Z. Yu and M. Gerstein, Bioinformatics,
2006, 22, 2735–8.
13 Examples of very simple molecules which form helical H-bonding
arrays in their crystals include (a) aziridine N. W. Mitzel, J. Riede
and C. Kiener, Angew. Chem., Int. Ed. Engl., 1997, 36, 2215–2216;
(b) 4-nitrophenol G. Wojcik and I. Mossakowska, Acta
Crystallogr., Sect. B: Struct. Sci., 2006, 62, 143–152.
14 K. Kinbara and K. Saigo, Top. Stereochem., 2003, 23, 207–265.
16 M. Nishio, M. Hirota, Y. Umezawa, The CH/p Interaction –
Evidence, Nature and Consequences, Wiley-VCH, New York, 1998
(Methods in Stereochemical Analysis, A.P. Marchand, Series Editor).
17 M. Haisa, S. Kashino and T. Kawashima, Acta Crystallogr., Sect. B:
Struct. Crystallogr. Cryst. Chem., 1980, 36, 1598–1601.
18 H. Lu, Z. Xue, J. Mack, Z. Shen, X. You and N. Kobayashi, Chem.
Commun., 2010, 46, 3565–3567.
19 R. Taylor and O. Kennard, J. Am. Chem. Soc., 1982, 104, 5063–5070.
20 R. W. W. Hooft, COLLECT Data Collection Software, Bruker AXS,
Delft, The Netherlands, 2008.
21 Z. Otwinowski and W. Minor, Methods in Enzymology, vol. 276:
Macromolecular Crystallography, Part A, Elsevier, 1997, pp. 307-326.
22 G. M. Sheldrick, ‘‘SHELXS-97 Program for Crystal Structure
Determination’’, Acta Crystallogr. 1990, A46, pp. 467–473.
€ €
23 G. M. Sheldrick, (1999) ‘‘SHELXL-97’’, Universitat Gottingen,
€
Gottingen, Germany.
24 Y. Naruta, Y. Tachi, T. Chishiro, Y. Shimazaki and F. Tani, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2001, 57, 550–552.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 7207–7211 | 7211