Two polymorphs, with z′ = 1 and 2, of 2-amino-4-chloro-6-morpholinopyrimidine in P21/c, and 2-amino-4-chloro-6-piperidinopyrimidine, which is isomorphous and almost isostructural with the Z′ = 2 polymorph

Article abstract of DOI:10.1107/S0108270102020012

Crystallization of 2-amino-4-chloro-6-morpholinopyrimidine, C₈H₁₁ClN₄O, (I), yields two polymorphs, both with space group P2₁/c, having Z′ = 1 (from diethyl ether solution) and Z′ = 2 (from dichloromethane solution), denoted (Ia) and (Ib), respectively. In polymorph (Ia), the molecules are linked by an N-H...O and an N-H...N hydrogen bond into sheets built from alternating R₂²(8) and R₆⁶(40) rings. In polymorph (Ib), one molecule acts as a triple acceptor of hydrogen bonds and the other acts as a single acceptor; one N-H...O and three N-H...N hydrogen bonds link the molecules in a complex chain containing two types of R₂²(8) and one type of R₂²(18) ring. 2-Amino-4-chloro-6-piperidinopyrimidine, C₉H₁₃ClN₄, (II), which is isomorphous with polymorph (Ib), also has Z′ = 2 in P2₁/c, and the molecules are linked by three N-H...N hydrogen bonds into a centrosymmetric four-molecule aggregate containing three R₂²(8) rings.

Full text of DOI:10.1107/S0108270102020012

organic compounds
in vitro inhibitors of the human DNA-repair protein O6-
alkylguanine±DNA±transferase (Chae et al., 1995; Quesada,
Marchal, Melguizo et al., 2002). We report here the molecular
and supramolecular structures of two examples of this class of
pyrimidine, namely 2-amino-4-chloro-6-morpholinopyrimi-
dine, (I), which crystallizes in two polymorphic forms, and
2-amino-4-chloro-6-piperidinopyrimidine, (II), which is iso-
morphous and almost isostructural with one of the poly-
morphs of (I).
Acta Crystallographica Section C
Crystal Structure
Communications
ISSN 0108-2701
Two polymorphs, with Z0 = 1 and 2,
of 2-amino-4-chloro-6-morpholino-
pyrimidine in P21/c, and 2-amino-
4-chloro-6-piperidinopyrimidine,
which is isomorphous and almost
isostructural with the Z0 = 2
polymorph
Katharine F. Bowes,a Christopher Glidewell,a* John N.
Low,b² Manuel Melguizoc and Antonio Quesadad³
The simpler polymorph of compound (I), i.e. (Ia) (Fig. 1),
has Z0 = 1 in P21/c, and the molecules are linked into sheets by
two hydrogen bonds, one each of the NÐHÁ Á ÁO and NÐ
HÁ Á ÁN types (Table 1). Amino atom N2 in the molecule at (x,
y, z) acts as hydrogen-bond donor, via H1, to the morpholine
atom O64 in the molecule at (2 x, 12 + y, 21 z), while atom
N2 at (2 x, 12 + y, 12 z) likewise acts as donor to O64 at (x,
1 + y, z), so producing a C(9) chain running parallel to the
[010] direction and generated by the 21 screw axis along (1, y, 14)
(Fig. 2). Adjacent [010] chains are linked into (102) sheets by
means of NÐHÁ Á ÁN hydrogen bonding. Amine atom N2 in the
molecule at (x, y, z), which lies in the [010] chain along (1, y, 14),
acts as hydrogen-bond donor, via H2, to atom N3 in the
molecule at (1 x, 1 y, 1 z), which lies in the antiparallel
aSchool of Chemistry, University of St Andrews, St Andrews, Fife KY16 9ST,
Scotland, bDepartment of Chemistry, University of Aberdeen, Meston Walk,
c
Â
Â
Old Aberdeen AB24 3UE, Scotland, Departamento de Quõmica Inorganica y
Organica, Universidad de Jaen, 23071 Jaen, Spain, and dSchool of Engineering,
Â
Â
Â
University of Dundee, Dundee DD1 4HN, Scotland
Correspondence e-mail: cg@st-andrews.ac.uk
Received 21 October 2002
Accepted 4 November 2002
Online 10 December 2002
Crystallization of 2-amino-4-chloro-6-morpholinopyrimidine,
C8H11ClN4O, (I), yields two polymorphs, both with space
group P21/c, having Z0 = 1 (from diethyl ether solution) and
Z0 = 2 (from dichloromethane solution), denoted (Ia) and (Ib),
respectively. In polymorph (Ia), the molecules are linked by an
NÐHÁ Á ÁO and an NÐHÁ Á ÁN hydrogen bond into sheets built
from alternating R22(8) and R66(40) rings. In polymorph (Ib),
one molecule acts as a triple acceptor of hydrogen bonds and
the other acts as a single acceptor; one NÐHÁ Á ÁO and three
NÐHÁ Á ÁN hydrogen bonds link the molecules in a complex
chain containing two types of R22(8) and one type of R44(18)
ring. 2-Amino-4-chloro-6-piperidinopyrimidine, C9H13ClN4,
(II), which is isomorphous with polymorph (Ib), also has
Z0 = 2 in P21/c, and the molecules are linked by three
NÐHÁ Á ÁN hydrogen bonds into a centrosymmetric four-
molecule aggregate containing three R22(8) rings.
3
chain along (0, y, ), so forming an R22(8) ring centred at
4
(21, 12, 21). In a similar way, the molecule at (2 x, 21 + y, 12 z),
a component of the (1, y, 41) chain, is linked via the R22(8) motif
to the molecule at (1 + x, 21 y, 12 + z), which lies in the chain
along (2, y, 14). In this manner, [010] chains are linked into a
(102) sheet, built from alternating R22(8) and R66(40) rings
(Fig. 2). If the individual molecules are regarded as the nodes
of the net de®ning the (102) sheet, then this net is of the (6,3)-
type (Batten & Robson, 1998), while if the R22(8) dimers are
Comment
Aminochloropyrimidines are key intermediates for the
synthesis of O6-benzyloxy-5-nitrosopyrimidines, which can
themselves be converted into a wide range of alkoxy- and
amino-substituted O6-benzyloxy-5-nitrosopyrimidines (Ques-
ada, Marchal, Melguizo et al., 2002; Quesada, Marchal, Nog-
ueras et al., 2002), which are important as potential, or proven,
Figure 1
² Postal address: School of Engineering, University of Dundee, Dundee
DD1 4HN, Scotland.
  Â
³ On leave from: Departamento de Quõmica Inorganica y Organica, Univer-
The molecule of compound (I) in the Z0 = 1 polymorph, showing the
atom-labelling scheme. Displacement ellipsoids are drawn at the 30%
probability level.
 Â
sidad de Jaen, 23071 Jaen, Spain.
o4 # 2003 International Union of Crystallography
DOI: 10.1107/S0108270102020012
Acta Cryst. (2003). C59, o4±o8

organic compounds
taken to be the nodes, the net is of the (4,4)-type. There are no
direction-speci®c interactions between adjacent sheets.
In the Z0 = 2 polymorph of compound (I), viz. (Ib), the two
independent molecules (Fig. 3) exhibit different hydrogen-
bonding behaviour and the hydrogen bonds (Table 2) generate
a rather elaborate one-dimensional structure. In both mol-
ecules, the amino group acts as a double donor, but molecule 1
(containing atom N11; Fig. 3) acts as a triple acceptor of
hydrogen bonds, via atoms N11, N13 and O164, whereas in
molecule 2, there is just a single acceptor, namely N23. These
facts alone preclude any additional symmetry. Within the
selected asymmetric unit (Fig. 3), the molecules are linked by
paired NÐHÁ Á ÁN hydrogen bonds, forming an R22(8) dimer in
which the orientation of the morpholine substituents
precludes even approximate additional symmetry. In analysing
the structure of this polymorph, it is convenient to employ the
substructure approach (Gregson et al., 2000), where the key
substructural unit is the R22(8) dimer; the formation of this
dimer leaves two NÐH bonds, N12ÐH12 and N22ÐH21,
available for the formation of further hydrogen bonds. Amino
atom N12 in the type-1 molecule at (x, y, z) acts as hydrogen-
bond donor, via H12, to atom N13 in the type-1 molecule at
(2 x, 1 y, 1 z), so generating a centrosymmetric R22(8)
Figure 3
The two independent molecules of compound (I) in the Z0 = 2 polymorph,
showing the atom-labelling scheme. Displacement ellipsoids are drawn at
the 30% probability level.
formation of a chain of rings running parallel to the [110]
direction (Fig. 4), such that the type-2 morpholine groups
containing atom O264 are pendent from the chain with no
plausible hydrogen-bond donors within hydrogen-bonding
distance.
Compound (II) has so far been isolated in only one crys-
talline form and this is isomorphous with the Z0 = 2 polymorph
of compound (I); the cell dimensions of these two P21/c, Z0 = 2
phases are very similar and it is possible to re®ne either
structure starting from the coordinates of the other, paying
due regard to the difference in atom types between the
morpholine substituent in (I) and the piperidine substituent in
(II). Since, however, there can be no NÐHÁ Á ÁO hydrogen
bonds in (II) (Table 3), the supramolecular aggregation in (II)
must be different from that in either polymorph of (I); such
hydrogen bonds are present in both polymorphs of (I) and
hence the two Z0 = 2 phases are not, strictly speaking, properly
isostructural, although they are nearly so. The asymmetric unit
selected for compound (II) resembles that for the Z0 = 2
1
1
motif, centred at (1, , ), while atom N22 in the type-2 mol-
2
2
ecule at (x, y, z) acts as donor, via H21, to atom O164 in the
type-1 molecule at (1 x, y, 1 z), so generating a
centrosymmetric R44(18) motif centred at (21, 0, 12). Propagation
by inversion of these hydrogen bonds thus leads to the
Figure 2
Part of the crystal structure of the Z0 = 1 polymorph of (I), showing the
formation of a (102) sheet built from alternating R22(8) and R66(40) rings.
For the sake of clarity, H atoms bonded to C atoms have been omitted.
Figure 4
Stereoview of part of the crystal structure of the Z0 = 2 polymorph of (I),
showing the formation of a chain of edge-fused R22(8) and R44(18) rings
along [110]. For the sake of clarity, H atoms bonded to C atoms have been
omitted.
Atoms marked with an asterisk (*), hash (#) or dollar sign ($) are at the
1
symmetry positions (1 x, 1 y, 1 z), (2 x, 12 + y,
(2 x, 12 + y, 12 z), respectively.
z) and
2
ꢀ
Acta Cryst. (2003). C59, o4±o8
Katharine F. Bowes et al.
Two polymorphs of C8H11ClN4O, and C9H13ClN4 o5

organic compounds
polymorph of (I) in that it forms an R22(8) dimer which itself
exhibits no symmetry (Fig. 5). Pairs of these dimers are linked
into a centrosymmetric four-molecule aggregate centred at
(1, 12, 12); atom N12 at (x, y, z) acts as hydrogen-bond donor, via
H12, to atom N13 at (2 x, 1 y, 1 z), but the N22ÐH21
bond does not participate in the hydrogen bonding, so that this
aggregate is strictly ®nite (Fig. 6).
It is notable that even in (II), where there is an exact match
between the number of hard hydrogen-bond donors and the
number of hard hydrogen-bond acceptors, not all of these
donors and acceptors are involved in the supramolecular
aggregation. Thus, atom N21 does not accept any hydrogen
bond, nor is N22ÐH21 involved. In both (Ia) and (Ib), there is
an excess of acceptors over donors, and in (Ia), atom N1 is
unused as an acceptor, while in (Ib) both N21 and O264 are
unused. The structures of (Ia), (Ib) and (II) do not contain any
soft hydrogen bonds of the CÐHÁ Á ÁA (A = N or O) or DÐ
HÁ Á Áꢀ types (D = C or N), nor do they contain any ꢀ±ꢀ-
stacking interactions; they thus have hydrogen-bonded
supramolecular structures which are two-dimensional, one-
dimensional and (®nite) zero-dimensional, respectively.
In each of (Ib) and (II), there are rather short contacts
between pairs of Cl atoms across the inversion centre at
Figure 6
Part of the crystal structure of (II), showing the formation of a four-
molecule aggregate containing three R22(8) rings. For the sake of clarity, H
atoms bonded to C atoms and the unit-cell box have been omitted. Atoms
marked with an asterisk (*) are at the symmetry position (2 x, 1 y,
1
z).
i
1
1
(1, , ) (Figs. 4 and 6); in (Ib), the Cl14Á Á ÁCl24 distance is
2
2
N atoms in these rings almost planar; the sum of the angles at
the N atom ranges from 352.3 (2)ꢁ in (Ia) to 359.9 (2)ꢁ in
molecule 1 of (Ib). Consistent with this planarity, the asso-
ciated CÐN distances for the bond linking the two rings
Ê
3.375 (2) A [symmetry code: (i) 2 x, 1 y, 1 z], asso-
ciated with a CÐClÁ Á ÁCl angle of 145.14 (5)ꢁ, while in (II), the
ꢁ
Ê
corresponding values are 3.425 (2) A and 150.08 (5) . For Cl
bonded to neutral arene groups, Bondi (1964) recommended a
Ê
within each molecule range from 1.352 (2) to 1.372 (2) A, with
Ê
van der Waals radius of 1.77 A; the ClÁ Á ÁCl contact distances
Ê
a mean of 1.361 (2) A, typical of ArÐN(C)2 distances invol-
ving di-N-substituted anilines with a planar N atom (mean
Ê
value 1.371 A; Allen et al., 1987). By way of comparison, the
Ê
CÐNH2 distances here range from 1.338 (2) to 1.358 (2) A,
Ê
with a mean value of 1.348 (2) A, again typical of ArÐNH2
distances in simple anilines with an unsubstituted planar N
in (Ib) and (II) are thus signi®cantly shorter than twice the van
der Waals radius. Database analysis of such halogen±halogen
contacts has shown that when the XÁ Á ÁX contact distance (X =
halogen) is signi®cantly less than the van der Waals sum, the
associated CÐXÁ Á ÁX angles are clustered around 180 and 90ꢁ
(Ramasabbu et al., 1986); the CÐClÁ Á ÁCl angles observed are
not particularly close to either cluster, and almost certainly
their values are constrained by the formation of the centro-
symmetric hydrogen-bonded R22(8) rings.
Ê
atom (mean value 1.355 A; Allen et al., 1987). The other
intramolecular distances and angles show no unusual features;
in particular, there is no evidence for any bond ®xation within
the pyrimidine rings.
Within the molecules, the morpholine and piperidine
substituents adopt the usual chair conformations, but with the
The very different cell dimensions of polymorphs (Ia) and
(Ib) effectively rule out any possibility of a simple displacive
phase transition between these two polymorphs. While we
have not investigated the relative stability of these two poly-
morphs, we note that (Ia) has a signi®cantly higher density
than (Ib) and thus is probably the thermodynamically more
stable form (Burger & Ramberger, 1979). Although poly-
morph (Ib) was the ®rst to be isolated, both forms can readily
be reproduced by crystallization from the appropriate solvent;
hence there is no question here of any disappearing poly-
morph (Dunitz & Bernstein, 1995).
Experimental
Samples of (I) and (II) were synthesized from 2-amino-4,6-di-
chloropyrimidine (purchased from Aldrich). Morpholine [for (I)] or
piperidine [for (II)] (3.1 mmol) and triethylamine (6.5 mmol) were
added to a stirred solution of 2-amino-4,6-dichloropyrimidine
(3.0 mmol) in 3-methylbutan-2-ol (20 ml). The mixtures were heated
Figure 5
The two independent molecules of compound (II), showing the atom-
labelling scheme. Displacement ellipsoids are drawn at the 30%
probability level.
ꢀ
o6 Katharine F. Bowes et al.
Two polymorphs of C8H11ClN4O, and C9H13ClN4
Acta Cryst. (2003). C59, o4±o8

organic compounds
under re¯ux, with stirring, for 2 h. After cooling, the solvent was
removed, excess distilled water was added, and the resulting solid was
®ltered off and washed with water. After drying, the solid products
were recrystallized ®rst from ethyl acetate and then from dichloro-
Data collection
Nonius KappaCCD diffractometer
' scans, and ! scans with ꢆ offsets
Absorption correction: multi-scan
(DENZO±SMN; Otwinowski &
Minor, 1997)
4346 independent re¯ections
3612 re¯ections with I > 2ꢇ(I)
Rint = 0.046
1
ꢄmax = 27.5ꢁ
methane; m.p. 485 K [for (I)] and 431 K [for (II)]. NMR for (I), H
h = 18 ! 16
(CHCl3, p.p.m.): ꢁ 3.55 (4H, t, J = 5.2 Hz, N±CH2), 3.74 (4H, t, J =
5.2 Hz, O±CH2), 4.90 (2H, br, NH2, exchanges with D2O), 5.94 (1H, s,
C5±H); 13C (CDCl3, p.p.m.): ꢁ 45.3, 66.6, 82.4, 160.2, 162.3, 163.3;
NMR for (II), 1H (CHCl3, p.p.m.): ꢁ 1.58±1.68 (6H, m, 3 Â CH2), 3.54
(4H, t, J = 45.0 Hz, N±CH2), 4.76 (2H, br, NH2, exchanges with D2O),
5.96 (1H, s, C5±H); 13C (CDCl3, p.p.m.): ꢁ 24.6, 25.5, 45.3, 82.4, 160.2,
162.2, 163.3. Crystals suitable for single-crystal X-ray diffraction were
grown by slow evaporation of solutions in diethyl ether [for (Ia)] or
dichloromethane [for (Ib) and (II)].
Tmin = 0.961, Tmax = 0.995
17 326 measured re¯ections
k = 10 ! 10
l = 25 ! 25
Re®nement
Re®nement on F2
R[F2 > 2ꢇ(F2)] = 0.037
wR(F2) = 0.097
S = 1.04
4346 re¯ections
w = 1/[ꢇ2(Fo2) + (0.0488P)2
+ 0.6560P]
where P = (Fo2 + 2Fc2)/3
(Á/ꢇ)max = 0.001
3
Ê
Áꢈmax = 0.32 e A
3
Ê
0.40 e A
253 parameters
H-atom parameters constrained
Áꢈmin
=
Polymorph (Ia)
Table 2
Hydrogen-bonding geometry (A, ) for (Ib).
ꢁ
Crystal data
Ê
3
C8H11ClN4O
Mr = 214.66
Monoclinic, P21/c
Ê
a = 9.4143 (4) A
Ê
b = 13.0869 (4) A
Dx = 1.537 Mg m
Mo Kꢃ radiation
Cell parameters from 2102
re¯ections
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
N12ÐH11Á Á ÁN23
N22ÐH22Á Á ÁN11
N12ÐH12Á Á ÁN13i
N22ÐH21Á Á ÁO164ii
0.88
0.88
0.88
0.88
2.19
2.35
2.18
2.23
3.061 (2)
3.211 (2)
3.036 (2)
3.051 (2)
174
165
165
155
ꢄ = 3.1±27.5ꢁ
ꢅ = 0.38 mm
T = 120 (2) K
1
Ê
c = 7.5774 (2) A
ꢂ = 96.4260 (13)ꢁ
3
Ê
V = 927.70 (5) A
Z = 4
Block, colourless
0.42 Â 0.22 Â 0.16 mm
Symmetry codes: (i) 2 x; 1 y; 1 z; (ii) 1 x; y; 1 z.
Data collection
Compound (II)
Nonius KappaCCD diffractometer
' scans, and ! scans with ꢆ offsets
Absorption correction: multi-scan
(DENZO±SMN; Otwinowski &
Minor, 1997)
2102 independent re¯ections
1705 re¯ections with I > 2ꢇ(I)
Rint = 0.046
Crystal data
ꢄ
max = 27.5ꢁ
3
C9H13ClN4
Mr = 212.68
Monoclinic, P21/c
Dx = 1.414 Mg m
Mo Kꢃ radiation
h = 12 ! 12
k = 16 ! 12
l = 8 ! 9
Tmin = 0.856, Tmax = 0.941
6031 measured re¯ections
Cell parameters from 4533
re¯ections
Ê
a = 13.8139 (2) A
ꢄ = 3.0±27.5ꢁ
ꢅ = 0.35 mm
T = 120 (1) K
Ê
b = 8.48040 (10) A
Re®nement
1
Ê
c = 19.8379 (3) A
Re®nement on F2
R[F2 > 2ꢇ(F2)] = 0.036
wR(F2) = 0.092
S = 1.05
2102 re¯ections
w = 1/[ꢇ2(Fo2) + (0.0371P)2
+ 0.4718P]
ꢂ = 120.6770 (9)ꢁ
Ê
V = 1998.74 (5) A
Z = 8
3
Block, colourless
0.45 Â 0.35 Â 0.25 mm
where P = (Fo2 + 2Fc2)/3
(Á/ꢇ)max < 0.001
3
Ê
Data collection
Áꢈmax = 0.35 e A
3
Ê
0.38 e A
127 parameters
H-atom parameters constrained
Áꢈmin
=
Nonius KappaCCD diffractometer
' scans, and ! scans with ꢆ offsets
Absorption correction: multi-scan
(DENZO±SMN; Otwinowski &
Minor, 1997)
4533 independent re¯ections
3726 re¯ections with I > 2ꢇ(I)
Rint = 0.042
ꢄmax = 27.5ꢁ
Table 1
Hydrogen-bonding geometry (A, ) for (Ia).
h = 17 ! 16
ꢁ
Ê
Tmin = 0.826, Tmax = 0.911
15 693 measured re¯ections
k = 10 ! 11
l = 25 ! 25
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
Re®nement
N2ÐH1Á Á ÁO64i
0.88
0.88
2.44
2.36
3.288 (2)
3.119 (2)
163
145
Re®nement on F2
R[F2 > 2ꢇ(F2)] = 0.033
wR(F2) = 0.087
S = 1.04
4533 re¯ections
w = 1/[ꢇ2(Fo2) + (0.0419P)2
+ 0.5327P]
N2ÐH2Á Á ÁN3ii
where P = (Fo2 + 2Fc2)/3
(Á/ꢇ)max = 0.001
1
2
1
2
Symmetry codes: (i) 2 x; y
;
z; (ii) 1 x; 1 y; 1 z.
3
Ê
Áꢈmax = 0.25 e A
3
Ê
0.28 e A
250 parameters
H-atom parameters constrained
Áꢈmin
=
Polymorph (Ib)
Crystal data
Table 3
Hydrogen-bonding geometry (A, ) for (II).
3
C8H11ClN4O
Mr = 214.66
Monoclinic, P21/c
Dx = 1.497 Mg m
Mo Kꢃ radiation
ꢁ
Ê
Cell parameters from 4346
re¯ections
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
Ê
a = 13.9215 (3) A
b = 8.1803 (2) A
ꢄ = 3.0±27.5ꢁ
Ê
N12ÐH11Á Á ÁN23
N22ÐH22Á Á ÁN11
N12ÐH12Á Á ÁN13i
0.88
0.88
0.88
2.19
2.40
2.12
3.0479 (16)
3.2601 (17)
2.9862 (16)
163
167
169
1
Ê
c = 19.6390 (4) A
ꢅ = 0.37 mm
T = 150 (1) K
ꢂ = 121.6130 (12)ꢁ
3
Ê
V = 1904.65 (8) A
Z = 8
Plate, colourless
0.30 Â 0.25 Â 0.02 mm
Symmetry code: (i) 2 x; 1 y; 1 z.
ꢀ
Acta Cryst. (2003). C59, o4±o8
Katharine F. Bowes et al.
Two polymorphs of C8H11ClN4O, and C9H13ClN4 o7

organic compounds
Polymorphs (Ia) and (Ib) and compound (II) all crystallize in space
group P21/c, the space group being uniquely assigned from the
systematic absences in each case. All H atoms were treated as riding
References
Allen, F. H., Kennard, O., Watson, D. G., Brammer, L., Orpen, A. G. & Taylor,
R. (1987). J. Chem. Soc. Perkin Trans. 2, pp. S1±19.
Ê
atoms, with CÐH distances of 0.95 (heteroaromatic) or 0.99 A (CH2),
Ê
and NÐH distances of 0.88 A.
Batten, S. R. & Robson, R. (1998). Angew. Chem. Int. Ed. 37, 1460±1494.
Bondi, A. (1964). J. Phys. Chem. 68, 441±451.
Burger, A. & Ramberger, R. (1979). Mikrochim. Acta, 2, 259±271.
Chae, M. Y., McDougall, M. G., Dolan, M. E., Swenn, K., Pegg, A. E. &
Moschel, R. C. (1995). J. Med. Chem. 38, 359±365.
Dunitz, J. D. & Bernstein, J. (1995). Acc. Chem. Res. 28, 193±200.
Ferguson, G. (1999). PRPKAPPA. University of Guelph, Canada.
Gregson, R. M., Glidewell, C., Ferguson, G. & Lough, A. J. (2000). Acta Cryst.
B56, 39±57.
Nonius (1997). KappaCCD Server Software. Windows 3.11 Version. Nonius
BV, Delft, The Netherlands.
For polymorphs (Ia) and (Ib) and compound (II), data collection:
KappaCCD Server Software (Nonius, 1997); cell re®nement:
DENZO±SMN (Otwinowski & Minor, 1997); data reduction:
DENZO±SMN; program(s) used to solve structure: SHELXS97
(Sheldrick, 1997); program(s) used to re®ne structure: SHELXL97
(Sheldrick, 1997); molecular graphics: PLATON (Spek, 2002); soft-
ware used to prepare material for publication: SHELXL97 and
PRPKAPPA (Ferguson, 1999).
Otwinowski, Z. & Minor, W. (1997). Methods in Enzymology, Vol. 276,
Macromolecular Crystallography, Part A, edited by C. W. Carter Jr & R. M.
Sweet, pp. 307±326. New York: Academic Press.
X-ray data were collected at the EPSRC X-ray Crystal-
lographic Service, University of Southampton, England; the
authors thank the staff for all their help and advice. JNL
thanks NCR Self-Service, Dundee, for grants which have
provided computing facilities for this work.
Â
Quesada, A., Marchal, A., Melguizo, M., Nogueras, M., Sanchez, A., Low, J. N.,
Cannon, D., Farrell, D. M. M. & Glidewell, C. (2002). Acta Cryst. B58, 300±
315.
Quesada, A., Marchal, A., Nogueras, M., Sanchez, A., Melguizo, M. & Low,
Â
J. N. (2002). XXth European Colloquium on Heterocyclic Chemistry,
Stockholm, Sweden. Abstracts, p. 135.
Ramasabbu, N., Parthasarathy, R. & Murray-Rust, P. (1986). J. Am. Chem.
Soc. 108, 4308±4314.
Sheldrick, G. M. (1997). SHELXS97 and SHELXL97. University of
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: SK1595). Services for accessing these data are
described at the back of the journal.
È
Gottingen, Germany.
Spek, A. L. (2002). PLATON. Version of June 2002. Utrecht University, The
Netherlands.
ꢀ
o8 Katharine F. Bowes et al.
Two polymorphs of C8H11ClN4O, and C9H13ClN4
Acta Cryst. (2003). C59, o4±o8