2,6-Dimeth­oxy-7,9-dimethyl­purinium iodide hemihydrate

In the title compound, C₉H₁₃N₄O₂⁺·I^-·0.5H₂O, the non-H atoms of the ionic components lie on a mirror plane in Cmca, with the O atom of the partial water mol­ecule lying on a twofold rotation axis. Whereas one of the meth­oxy methyl groups is directed away from the adjacent N-methyl group, the other methoxy methyl group is directed towards its adjacent N-methyl group. The conformation of the methoxy methyl groups provides an explanation for the outcomes of intra­molecular thermal rearrangements of 2,6-dialk­oxy-7,9-di­methyl­purinium salts.

S1. Comment

Purinium salts exhibit interesting biological activity against adenocarcinoma and cytomegalovirus (Fujii & Itaya, 1999b) and affect pig brain ATPase (Fujii & Itaya, 1999a). Some substituted 6-methoxy-7,9-dimethylpurinium salts (isolated from Heterostemma brownii Hay – Asclepiadaceae) exhibit cytotoxic activity towards several tumor cell lines (Lin et al., 1996, 1997), and some 2,6-dimethoxy-7,9-dimethylpurinium salts exhibit cytotoxic activity against alga Chlorella vulgaris (Kowalska & Sochacka, 2003). Purinium salts also play an important role in the synthesis of purine compounds via the Hilbert–Johnson reaction (Pliml & Prystas, 1967) or transglycosylation (Boryski, 1996). We found that quaternization of 2,6-dialkoxy- and 2(6)-chloroalkoxy-7-methylpurines with alkyl halides in aprotic solvents led mostly to 2,6-disubstituted-7,9-dialkylpurinium halides and 6-alkoxyhypoxanthines instead of 1,3,7-trialkylxanthines (Kowalska et al., 1993; Kowalska & Maślankiewicz, 2001), although reaction of 2,6-dimethoxy-7-methylpurine, (I), with methyl iodide in a sealed tube at 373 K gave caffeine (Bergman & Heimbold, 1935). The absence of N1-alkylation was postulated to result from the conformations of the two alkoxy groups, causing steric hindrance at N1 (Kowalska et al., 1993).

Recently, the structures of a few 7,9-dialkylpurinium salts (where the anion is iodide, bromide or perchlorate) have been analyzed (Sigel et al., 2002; Nasiri et al., 2005; Hocek et al., 2005; Fu & Lam, 2005; Torii et al., 2006). We have now determined the crystal structure of 2,6-dimethoxy-7,9-dimethylpurinium iodide, (II), as its hemihydrate, to obtain information about the positions and conformations of the methoxy groups and the geometry of the imidazole ring in comparison with (I) (Kowalska et al., 1999), in an attempt to explain some results of the internal thermal O–N and N–N migrations of the alkyl groups (Kowalska & Maślankiewicz, 1997).

The non-H atoms of the cation of (II) (Fig. 1) all lie on a mirror plane in Cmca. For the O-methoxy groups, this permits complete overlap between the non-bonding p-type orbitals of the O atoms and π-orbitals of the pyrimidine ring. Alkylation at atom N9 does not influence the conformation of the O2/CH₃ group as compared with that in (I). Whereas the bond lengths and bond angles in the pyrimidine ring of (II) are very close to those found in (I), the geometric details of the imidazole rings are different. The essential difference concerns the bonds and bond angles connected with atoms N7, C8 and N9. Whereas in (I) the N7—C8 bond is longer than the C8—N9 bond [1.349 (2) versus 1.312 (2) Å], in (II) the C8—N9 bond is longer [1.326 (4) versus 1.343 (4) Å]. There is also a change in the bond angles C5—N7—C8 [104.6 (1) versus 107.1 (3)° in compounds (I) and (II), respectively], N7—C8—N9 [115.1 (1) versus 110.6 (3)°] and C4—N9—C8 [103.6 (1) versus 107.3 (3)°]. The methyl atoms C71 and C91 are equally directed towards atom C8.

All these geometric details make this fragment of the imidazole ring more regular in (II) than in (I). The N7—C8 and N9—C8 bond lengths are also unequal in the recently reported 7,9-dimethylpurinium salts (Sigel et al., 2002; Nasiri et al., 2005; Hocek et al., 2005; Fu & Lam, 2005; Torii et al., 2006) represented by structures (III)–(V), respectively [it would be preferable to specify exactly which authors reported which compounds, since there are 3 structures and 5 references] (the differences in N—C bond lengths are in the range 0.016–0.025 Å). As the N7—C8 bond is shorter by 0.017 Å in salt (II), double-bond character can be assigned to this bond and the resonance structure is represented by formula (IIB). The presence of the second methyl group on the imidazole ring has some consequences in physicochemical properties. We found atom H8 to be much more acidic in the purinium salt (II) than in the purine (I). We observed the signal of atom H8 to be strongly shifted downfield by 2.89 p.p.m. in ¹H NMR spectroscopy [7.75 p.p.m. and 10.64 p.p.m. in compounds (I) and (II), respectively; Kowalska et al., 1993; Kowalska, 2007). Both N-methyl groups have very close chemical shifts (4.03 p.p.m. for N7/CH₃ and 4.07 p.p.m. for N9/CH₃) and the chemical shift of N7/CH₃ is different from that in (I) (3.93 p.p.m.).

There is a close contact of 3.101 (4) Å between the peri-substituents (O6···C71H₃), which is less than the sum of their van der Waals radii (3.40 Å; Pauling, 1960), meaning that the methoxy group is directed towards atom N1. We have reported that 2,6-dialkoxy-7,9-dimethylpurinium salts undergo O–N and N–N thermal rearrangement to give all four possible 1,3-dialkylxanthines via inter- and intramolecular alkyl group migration (Kowalska, & Maślankiewicz, 1997). The very close intramomolecular contacts, i.e. C61···N1 of 2.714 (4) Å, C21···N3 of 2.668 (4) Å and C91···N3 of 3.055 (4) Å, well explain the observed O6–N1, O2–N3 and N9–N1 methyl-group migration as intramolecular rearrangement.

As the crystals of (II) were grown from aqueous solution, we found water molecules in the crystal structure, with the O atom located οn a twofold rotation axis with an occupancy of 0.5. There is a weak C—H···O hydrogen bond between the cation and the partial water molecule (Table 2). There are also hydrogen bonds between the disordered iodide anions and water molecules (Table 2). The cations and anions are arranged in layers separated by 3.492 (2) Å (Fig. 2), while the water molecules are located in discrete cavities between the layers (Fig. 3).

Related literature

For related literature, see: Bergman & Heimbold (1935); Boryski (1996); Fu & Lam (2005); Fujii & Itaya (1999a, 1999b); Hocek et al. (2005); Kowalska (2007); Kowalska & Maślankiewicz (1997, 2001); Kowalska & Sochacka (2003); Kowalska et al. (1993, 1999); Lin et al. (1996, 1997); Nardelli (1999); Nasiri et al. (2005); Pauling (1960); Pliml & Prystas (1967); Sigel et al. (2002); Torii et al. (2006).

S2. Experimental

The title compound was obtained in 71% yield from the reaction of (I) with methyl iodide in acetonitrile at room temperature following the procedure described by Kowalska et al. (1993). Single crystals were grown from aqueous solution at room temperature.

S3. Refinement

H atoms in the cation were treated as riding atoms in geometrically idealized positions, with C—H distances of 0.95 Å (ring) or 0.98 Å (CH₃), and with U_iso(H) = kU_eq(C), where k = 1.5 for the methyl groups and k = 1.2 otherwise. The H atoms of the water component could not be located in difference maps, but their positions were calculated as described by Nardelli (1999), and the water molecule was then refined as a rigid body [O—H = 0.85 Å and U_iso(H) = 1.5U_eq(O)]. Even applying an analytical absorption correction, there was a large (> 6 e Å^-3) peak in the Fourier electron density maps approx 0.5 Å from the iodide anion. Applying a disorder of the I^- anion (4:1) gave better convergence of the refinement. The large thermal displacement parameter of the water molecule, when compared with the other atoms, may be explained by possible disorder-induced formation of hydrogen bonds to the disordered I^- anion.

Computing details

Data collection: COLLECT (Nonius, 1998); cell refinement: DENZO and SCALEPACK (Otwinowski & Minor, 1997); data reduction: DENZO and SCALEPACK (Otwinowski & Minor, 1997); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: ORTEPIII (Burnett & Johnson, 1996) and X-SEED (Barbour, 2001); software used to prepare material for publication: publCIF (Westrip, 2008).

Figures

	Fig. 1. The molecular structure of (II), with the atom labelling; the anion is disordered over two sites, and the water molecule has 0.5 occupancy. Displacement ellipsoids are drawn at the 30% probability level.
	Fig. 2. The packing of (II) along the b axis, showing the layer-type structure.
	Fig. 3. The interlayer cavities that accommodate the water molecules.

2,6-dimethoxy-7,9-dimethylpurinium iodide hemihydrate

Crystal data

C9H13N4O2^+·I^−·0.5H2O	F(000) = 1352
Mr = 345.14	Dx = 1.767 Mg m^−3
Orthorhombic, Cmca	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -C 2bc 2	Cell parameters from 15312 reflections
a = 6.9841 (2) Å	θ = 2.6–35.0°
b = 13.0977 (4) Å	µ = 2.47 mm^−1
c = 28.3695 (8) Å	T = 100 K
V = 2595.12 (13) Å^3	Block, colourless
Z = 8	0.21 × 0.15 × 0.08 mm

Data collection

Bruker KappaAPEXII diffractometer	2983 independent reflections
Radiation source: Enraf-Nonius FR590	2699 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.064
Detector resolution: 8.3 pixels mm^-1	θmax = 35.0°, θmin = 2.9°
CCD rotation images scans	h = −11→11
Absorption correction: analytical (Alcock, 1970)	k = −21→16
Tmin = 0.641, Tmax = 0.828	l = −45→45
15312 measured reflections

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.041	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.082	H-atom parameters constrained
S = 1.19	w = 1/[σ^2(Fo^2) + (0.P)^2 + 10.7633P] where P = (Fo^2 + 2Fc^2)/3
2983 reflections	(Δ/σ)max < 0.001
115 parameters	Δρmax = 0.74 e Å^−3
0 restraints	Δρmin = −0.99 e Å^−3

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq	Occ. (<1)
C2	0.0000	0.2494 (2)	0.46713 (10)	0.0220 (6)
C4	0.0000	0.2970 (2)	0.39276 (10)	0.0193 (5)
C5	0.0000	0.1980 (2)	0.37747 (10)	0.0178 (5)
C6	0.0000	0.1214 (2)	0.41236 (10)	0.0191 (5)
C8	0.0000	0.2955 (3)	0.31546 (11)	0.0253 (6)
H8	0.0000	0.3178	0.2836	0.030*
C21	0.0000	0.3702 (2)	0.53004 (11)	0.0244 (6)
H21A	−0.1061	0.4072	0.5152	0.037*	0.50
H21B	0.1215	0.4027	0.5214	0.037*	0.50
H21C	−0.0154	0.3717	0.5644	0.037*	0.50
C61	0.0000	−0.0529 (2)	0.43593 (11)	0.0248 (6)
H61A	−0.0146	−0.1206	0.4217	0.037*	0.50
H61B	−0.1066	−0.0403	0.4576	0.037*	0.50
H61C	0.1212	−0.0500	0.4533	0.037*	0.50
C71	0.0000	0.1102 (3)	0.29729 (12)	0.0391 (11)
H71A	0.0892	0.0585	0.3092	0.059*	0.50
H71B	0.0400	0.1316	0.2657	0.059*	0.50
H71C	−0.1292	0.0812	0.2958	0.059*	0.50
C91	0.0000	0.4692 (3)	0.35194 (13)	0.0302 (8)
H91A	−0.1234	0.4947	0.3634	0.045*	0.50
H91B	0.0203	0.4926	0.3195	0.045*	0.50
H91C	0.1031	0.4951	0.3721	0.045*	0.50
N1	0.0000	0.1489 (2)	0.45736 (9)	0.0203 (5)
N3	0.0000	0.32885 (19)	0.43801 (9)	0.0183 (4)
N7	0.0000	0.1986 (2)	0.32894 (9)	0.0221 (5)
N9	0.0000	0.3578 (2)	0.35306 (9)	0.0204 (5)
O2	0.0000	0.26578 (18)	0.51398 (8)	0.0252 (5)
O6	0.0000	0.02437 (17)	0.39915 (8)	0.0229 (5)
O1W	0.2500	0.4344 (5)	0.2500	0.0463 (16)	0.50
H1W	0.2900	0.3832	0.2657	0.069*	0.25
H2W	0.2135	0.4128	0.2232	0.069*	0.25
I1A	0.0000	0.26071 (6)	0.17445 (4)	0.02465 (15)	0.80
I1B	0.0000	0.2855 (2)	0.17975 (14)	0.0218 (5)	0.20

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C2	0.0341 (16)	0.0177 (13)	0.0141 (11)	0.000	0.000	−0.0017 (9)
C4	0.0216 (13)	0.0178 (12)	0.0186 (12)	0.000	0.000	0.0013 (9)
C5	0.0186 (12)	0.0210 (13)	0.0140 (10)	0.000	0.000	0.0007 (9)
C6	0.0218 (13)	0.0185 (12)	0.0171 (11)	0.000	0.000	0.0020 (9)
C8	0.0384 (18)	0.0228 (14)	0.0149 (12)	0.000	0.000	0.0021 (10)
C21	0.0334 (17)	0.0208 (14)	0.0191 (13)	0.000	0.000	−0.0052 (10)
C61	0.0395 (19)	0.0172 (13)	0.0176 (12)	0.000	0.000	0.0027 (10)
C71	0.080 (3)	0.0215 (15)	0.0162 (13)	0.000	0.000	−0.0025 (11)
C91	0.048 (2)	0.0168 (13)	0.0253 (15)	0.000	0.000	0.0027 (11)
N1	0.0252 (13)	0.0199 (11)	0.0157 (10)	0.000	0.000	0.0009 (8)
N3	0.0207 (12)	0.0178 (11)	0.0165 (10)	0.000	0.000	0.0006 (8)
N7	0.0328 (14)	0.0199 (11)	0.0135 (10)	0.000	0.000	0.0012 (8)
N9	0.0286 (13)	0.0167 (11)	0.0160 (10)	0.000	0.000	0.0022 (8)
O2	0.0421 (15)	0.0183 (10)	0.0153 (9)	0.000	0.000	−0.0007 (7)
O6	0.0366 (14)	0.0160 (9)	0.0162 (9)	0.000	0.000	0.0005 (7)
O1W	0.056 (5)	0.042 (3)	0.041 (3)	0.000	0.001 (3)	0.000
I1A	0.02620 (18)	0.0259 (3)	0.0218 (2)	0.000	0.000	0.0048 (2)
I1B	0.0147 (5)	0.0324 (15)	0.0184 (9)	0.000	0.000	0.0076 (9)

Geometric parameters (Å, º)

C2—N3	1.329 (4)	C21—H21C	0.9800
C2—N1	1.344 (4)	C61—O6	1.454 (4)
C2—O2	1.346 (4)	C61—H61A	0.9800
C4—N3	1.350 (4)	C61—H61B	0.9800
C4—C5	1.368 (4)	C61—H61C	0.9800
C4—N9	1.380 (4)	C71—N7	1.466 (4)
C5—N7	1.377 (4)	C71—H71A	0.9800
C5—C6	1.409 (4)	C71—H71B	0.9800
C6—O6	1.325 (4)	C71—H71C	0.9800
C6—N1	1.326 (4)	C91—N9	1.459 (4)
C8—N7	1.326 (4)	C91—H91A	0.9800
C8—N9	1.343 (4)	C91—H91B	0.9800
C8—H8	0.9500	C91—H91C	0.9800
C21—O2	1.442 (4)	O1W—H1W	0.85
C21—H21A	0.9800	O1W—H2W	0.85
C21—H21B	0.9800
N3—C2—N1	129.7 (3)	N7—C71—H71A	109.5
N3—C2—O2	119.2 (3)	N7—C71—H71B	109.5
N1—C2—O2	111.1 (3)	H71A—C71—H71B	109.5
N3—C4—C5	126.5 (3)	N7—C71—H71C	109.5
N3—C4—N9	126.7 (3)	H71A—C71—H71C	109.5
C5—C4—N9	106.8 (3)	H71B—C71—H71C	109.5
C4—C5—N7	108.1 (3)	N9—C91—H91A	109.5
C4—C5—C6	116.9 (3)	N9—C91—H91B	109.5
N7—C5—C6	135.0 (3)	H91A—C91—H91B	109.5
O6—C6—N1	122.2 (3)	N9—C91—H91C	109.5
O6—C6—C5	118.9 (3)	H91A—C91—H91C	109.5
N1—C6—C5	118.9 (3)	H91B—C91—H91C	109.5
N7—C8—N9	110.6 (3)	C6—N1—C2	117.7 (3)
N7—C8—H8	124.7	C2—N3—C4	110.4 (3)
N9—C8—H8	124.7	C8—N7—C5	107.1 (3)
O2—C21—H21A	109.5	C8—N7—C71	125.5 (3)
O2—C21—H21B	109.5	C5—N7—C71	127.4 (3)
H21A—C21—H21B	109.5	C8—N9—C4	107.3 (3)
O2—C21—H21C	109.5	C8—N9—C91	126.2 (3)
H21A—C21—H21C	109.5	C4—N9—C91	126.5 (3)
H21B—C21—H21C	109.5	C2—O2—C21	117.6 (2)
O6—C61—H61A	109.5	C6—O6—C61	117.7 (2)
O6—C61—H61B	109.5	H1W—O1W—H2W	107.7
H61A—C61—H61B	109.5	I1A—O1W—I1A^i	101.2 (2)
O6—C61—H61C	109.5	I1B—O1W—I1B^i	107.2 (2)
H61A—C61—H61C	109.5	I1A—O1W—I1B^i	104.2 (2)
H61B—C61—H61C	109.5
N3—C4—C5—N7	180.0	N9—C8—N7—C5	0.0
N9—C4—C5—N7	0.0	N9—C8—N7—C71	180.0
N3—C4—C5—C6	0.0	C4—C5—N7—C8	0.0
N9—C4—C5—C6	180.0	C6—C5—N7—C8	180.0
C4—C5—C6—O6	180.0	C4—C5—N7—C71	180.0
N7—C5—C6—O6	0.0	C6—C5—N7—C71	0.0
C4—C5—C6—N1	0.0	N7—C8—N9—C4	0.0
N7—C5—C6—N1	180.0	N7—C8—N9—C91	180.0
O6—C6—N1—C2	180.0	N3—C4—N9—C8	180.0
C5—C6—N1—C2	0.0	C5—C4—N9—C8	0.0
N3—C2—N1—C6	0.0	N3—C4—N9—C91	0.0
O2—C2—N1—C6	180.0	C5—C4—N9—C91	180.0
N1—C2—N3—C4	0.0	N3—C2—O2—C21	0.0
O2—C2—N3—C4	180.0	N1—C2—O2—C21	180.0
C5—C4—N3—C2	0.0	N1—C6—O6—C61	0.0
N9—C4—N3—C2	180.0	C5—C6—O6—C61	180.0

Symmetry code: (i) −x+1/2, y, −z+1/2.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
C8—H8···O1W	0.95	2.51	3.131 (5)	123
C8—H8···I1B	0.95	2.98	3.852 (5)	154
O1W—H2W···I1A	0.85	2.85	3.580 (4)	146
O1W—H2W···I1B	0.85	2.55	3.290 (5)	145
O1W—H1W···I1A^i	0.85	2.76	3.580 (4)	163
O1W—H1W···I1B^i	0.85	2.49	3.290 (5)	157

Symmetry code: (i) −x+1/2, y, −z+1/2.