Self-recognition in a flexible bis(pyrrole) Schiff base derivative: formation of a one-dimensional hydrogen-bonded polymer

The title Schiff base compound, N,N′-bis­(pyrrol-2-yl­methyl­ene)­propane-1,2-di­amine, C₁₃H₁₆N₄, forms an interesting supramolecular structure (a one-dimensional ladder-like polymer) in the solid state that is based on the existence of complementary intermolecular N—H...N=C hydrogen bonds between the monomer units. The polymer axis is collinear with the c axis of the orthorhombic unit cell. Quantum-chemical AM1 calculations clearly indicate that self-recognition in this system by hydrogen bonding is favoured on electrostatic grounds, since the partial atomic charge on the H atom of the pyrrole NH group (0.274 e) complements the partial atomic charge of the N atom of the C=N group (−0.239 e) on a neighbouring mol­ecule.

S1. Comment

Although tetradentate Schiff base ligands comprising two pyrrole groups bridged by a synthetically variable di(azomethine) unit X [structure (I)] have been known for several decades (Weber, 1967), studies aimed at elucidating the chemistry (Jones & McCleverty, 1971; van Stein et al., 1984) and structures of both the free bases and metal-containing coordination complexes of these synthetically feasible compounds are quite limited. Coordination of type (I) ligands to metal ions typically occurs with concomitant deprotonation of the two pyrrole NH H atoms, to give structures of type (II), where M represents a divalent metal ion (or any other feasible oxidation state). Structurally characterized examples of systems belonging to type (II) include complexes of Ru^II (Stern et al., 2000), Pd^II (Bacchi et al., 2003), Ni^II (Bailey & Hull, 1976), Mn^II (Franceschi et al., 2001), Sm^II (Berube et al., 2003) and Co^III (Mueller-Westerhoff et al., 1996; Allen, 2002). \sch

Although mononuclear coordination complexes are generally anticipated from reactions of (I) with divalent metal ions, e.g. NiL, where L is the N,N'-bis(pyrrol-2-ylmethylene)ethane-1,2-diamine dianion (Kabuto et al., 1984), recent structural studies have shown that even a relatively rigid derivative of (I), in which fragment X is based on 1,2-diamino-3,4-dimethylbenzene, is capable of conformational twisting to form binuclear complexes with a metal:ligand stoichiometry of 1:1 and the general formula M₂L₂ (Franceschi et al., 2001).

Despite the interesting coordination possibilities offered by derivatives of (I), only two free bases belonging to this group of compounds have been structurally characterized, namely N,N'-(1,2-cyclohexylene)bis(1H-pyrrol-2-ylmethylene)amine (Bacchi et al., 2003) and 4,5-dimethyl-N,N'-bis(1H-pyrrol-2-ylmethylene)benzene-1,2-diamine (Franceschi et al., 2001). The structure of the latter derivative is quite remarkable, since it forms a hydrogen-bonded dimer in which each C-shaped molecule becomes interlocked by hydrogen-bonding as a result of self-recognition. Indeed, we have recently found that self-recognition seems to be favoured in such systems, e.g. N-(1H-pyrrol-2-ylmethylene)benzene-1,2-diamine, as a result of the combination of one or more hydrogen-bond donor (pyrrole NH group) and hydrogen-bond acceptor sites (C═N group) within the molecule (Munro et al., 2003). In this paper, we report the X-ray structural characterization of the title compound, (III), the first example of a one dimensional hydrogen-bonded homopolymeric free base derivative belonging to group (I).

The X-ray crystal structure of (III) reveals that the geometry around each azomethine or imine group is exclusively the E isomer, consistent with the fact that the Z isomer would lead to unfavourable steric interactions between the pyrrole NH groups and the adjacent methylene groups of the propyl bridge (Fig. 1). The mean pyrrole α-C—N, α-C-β-C and β-C-β-C bond lengths are 1.373 (7), 1.384 (5) and 1.413 (4) Å, respectively (Table 1). The pair of imine C═N bond lengths average 1.280 (1) Å, while the mean pyrrole α-C-(C═N) bond length is 1.446 (1) Å. The mean C—C bond length of the aliphatic chain is 1.528 (2) Å. The mean bond angles subtended at the N atoms of (III) are 109.0 (1) (α-C—N-α-C), 125.0 (5) (H—N-α-C) and 116.9 (1)° (C═N—CH₂). Collectively, these mean distances and angles compare favourably with those reported for similar group (I) derivatives (Franceschi et al., 2001; Bacchi et al., 2003).

With six torsional degrees of freedom, many potentially stable conformations are possible for (III). However, the conformation favoured in the crystal structure exhibits an all-staggered conformation for the methylene groups of the propyl chain and an anti configuration for the two (1H-pyrrol-2-ylmethylene)amine units at either end of the molecule. The reason for this particular conformational preference in (III) becomes apparent upon inspection of the supramolecular structure (see below). Furthermore, although there is no formal or crystallographically imposed symmetry on the conformation of (III), it is clear that the molecule has approximate C₂ symmetry, with the twofold axis running through atom C7 along the bisector of the C6—C7—C8 backbone angle and in the plane of these three atoms (Fig. 1 b). The latter projection of the molecular structure also illustrates the rather interesting W-shaped conformation for (III).

The unique conformational features of (III) may be attributed to the supramolecular structure of the compound. More specifically, as illustrated in Fig. 2, the pyrrole NH groups serve as a hydrogen-bond donors and the azomethine groups as hydrogen-bond acceptors (Table 2). This pair of functional groups creates a structural motif that allows for `recognition' of the complementary motif in a neighbouring molecule in the lattice. Indeed, self-recognition and dimer formation of this type have been observed previously in the structurally related system 2-(2-pyrrolyl)-1,3-benzothiazole (Davidović et al., 1999). However, in (III), there are two hydrogen-bonding motifs at opposite ends of the molecule, such that self-recognition or complementary hydrogen-bonding favours the formation of one-dimensional polymeric chains, in which the polymer axis runs collinear with the c axis of the unit cell. The hydrogen-bonding in this system is therefore characterized by the formation of stable ten-membered rings, in which the pairs of planar hydrogen-bonding motifs do not lie in the same plane but are canted at 67.8 (1)°, to produce a twist in the ten-membered ring. This twisting presumably minimizes unfavourable steric contacts between the pyrrole α-CH group on one molecule and the methylene group appended to the azomethine group of the neighbouring molecule.

The formation of complementary hydrogen bonds between neighbouring molecules in the lattice of (III) leads to a ladder-like structure, in which each step of the ladder is laterally displaced from the preceding step. The anti configuration of the two (1H-pyrrol-2-ylmethylene)amine units (hydrogen-bonding motifs) at either end of the molecule is thus clearly required to facilitate the formation of a stable polymer chain. Interestingly, in the recently reported X-ray crystal structure of N,N'-(1,2-cyclohexylene)-bis(1H-pyrrol-2-ylmethylene)amine monohydrate, a hydrogen-bonded water molecule is effectively chelated by the (1H-pyrrol-2-ylmethylene)amine units to form a polymeric supramolecular structure (Bacchi et al., 2003). Evidently, this structural unit (pyrrole NH/imine C═N) builds into such systems the intrinsic ability to form hydrogen-bonded networks. To date, therefore, two of the three structurally characterized examples of compounds belonging to group (I) are essentially one-dimensional polymeric structures. Although the hydrogen-bonding interactions are clearly complementary, the interaction angles are 12–16° narrower than the `ideal' hydrogen-bonding angle of 180°, consistent with the specific architecture of the two interacting (1H-pyrrol-2-ylmethylene)amine units and their relative non-coplanar, or slightly twisted, orientation in space.

In order to understand better the supramolecular structure of (III), we have carried out a series of gas-phase quantum-chemical AM1 calculations (Dewar et al., 1985) on the monomer structure of (III), as well as on its hydrogen-bonded trimer and pentamer supramolecular analogues. Our objectives were, firstly, to quantify the electrostatics of the intermolecular hydrogen-bonding interactions observed for (III) and, secondly, to assess how reliably one could simulate a complex hydrogen-bonded polymer using electronic structure theory methods. Since calculations on three or five interacting monomers are large-scale to say the least, ab initio or density functional theory calculations at the required level of theory (6–31G* or better) were clearly unfeasible in this case. Notwithstanding the obvious theoretical limitations of a semi-empirical valence-electron-only method like AM1, we found that qualitatively sensible partial charge distributions were calculated for all structures considered (monomer, trimer and pentamer). This is shown in Fig. 3a, which gives the partial charge distribution (i.e. Mulliken charges) for the centre molecule in the hydrogen-bonded trimer shown in Fig. 3 b. Clearly, the pyrrole NH H atom has the highest positive fractional charge, while the imine N atom has the largest negative fractional charge.

Thus in terms of simple electrostatic arguments, the AM1 calculations readily explain the N—H···N═C hydrogen-bond complementarity observed in the X-ray crystal structure of (III). Moreover, as shown by the root-mean-square fit (0.460 Å for all atoms) of the geometry of the AM1-calculated trimer to that of the X-ray structure, a reasonable simulation of the hydrogen-bonded polymer is possible at this level of theory. The complementary N—H···N═C hydrogen bonds average 2.56 (2) Å in the calculated structure, compared with 2.07 (1) Å in the X-ray structure. However, the most significant deviations between the calculated and crystal structures are for the orientations of the pyrrole rings. This reflects the fact that each pyrrole ring packs rather closely with a neighbouring molecule in the experimental structure, a factor that has not been included in the simulations because a complete simulation of the lattice is unfeasible with our presently available computer resources.

S2. Experimental

Hexane and dichloromethane (BDH) were distilled from sodium metal and calcium hydride, respectively, before use. Ethanol (96%, BDH) was used as received. Compound (III) was synthesized from propane-1,3-diamine and 1H-pyrrole-2-carbaldehyde (both from Aldrich) in refluxing ethanol, following the literature method of Jones (1994). ¹H and ¹³C NMR spectroscopic data for (III) were consistent with those reported in the literature (Jones, 1994). Single crystals of (III) suitable for X-ray diffraction studies were grown by slow diffusion of hexane into a dichloromethane solution of (III). AM1 geometry optimization calculations were carried out with the default singlet state parameters in HYPERCHEM (Hypercube, 2000).

S3. Refinement

A difference Fourier calculation, after anisotropic refinement of the C and N atoms of (III), located all of the H atoms in the molecule. We elected to refine the pyrrole NH-group H atoms isotropically without restraints; all other H atoms were calculated using the standard riding model of SHELXL97 (HFIX 23 and HFIX 43; Sheldrick, 1997).

Computing details

Data collection: CrysAlis CCD (Oxford Diffraction, Year?); cell refinement: CrysAlis CCD; data reduction: CrysAlis RED (Oxford Diffraction, Year?); program(s) used to solve structure: SHELXS97 (Sheldrick, 1997); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: WinGX (Farrugia, 1999); software used to prepare material for publication: WinGX.

Figures

	Fig. 1. (a) A labelled `top' view of (III) at 120 (2) K. (b) A partly labelled side view of (III), illustrating the W-shaped conformation of the molecule. In both diagrams, displacement ellipsoids are drawn at the 60% probability level and H atoms are shown as small spheres of arbitrary radii.
	Fig. 2. A stereoscopic view of three molecules forming part of an infinite one-dimensional hydrogen-bonded polymer in the crystal lattice of (III). The polymer axis is collinear with the c axis of the unit cell (symmetry codes are as given in Table 2).
	Fig. 3. (a) The AM1-calculated Mulliken charges for (III). (b) The root-mean-square fit (0.460 Å) of the AM1-calculated structure of three consecutive hydrogen-bonded units of (III) to the X-ray crystal structure. The calculated Mulliken charges in (a) are taken from the centre molecule in (b).

N,N'-bis(pyrrol-2-ylmethylene)propane-1,2-diamine

Crystal data

C13H16N4	F(000) = 976
Mr = 228.3	Dx = 1.187 Mg m^−3
Orthorhombic, Pccn	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -P 2ab 2ac	Cell parameters from 826 reflections
a = 15.469 (3) Å	θ = 4–32°
b = 20.644 (4) Å	µ = 0.08 mm^−1
c = 7.997 (4) Å	T = 120 K
V = 2554.0 (14) Å^3	Needle, yellow
Z = 8	0.75 × 0.25 × 0.25 mm

Data collection

Oxford Diffraction Xcalibur2 CCD area-detector diffractometer	3367 reflections with I > 2σ(I)
ω/2θ scans	Rint = 0.033
Absorption correction: multi-scan (Blessing, 1995)	θmax = 31.9°, θmin = 4.2°
Tmin = 0.879, Tmax = 0.985	h = −21→22
23402 measured reflections	k = −29→29
4144 independent reflections	l = −7→11

Refinement

Refinement on F^2	0 restraints
Least-squares matrix: full	H atoms treated by a mixture of independent and constrained refinement
R[F^2 > 2σ(F^2)] = 0.046	w = 1/[σ^2(Fo^2) + (0.0575P)^2 + 0.6559P] where P = (Fo^2 + 2Fc^2)/3
wR(F^2) = 0.123	(Δ/σ)max = 0.001
S = 1.10	Δρmax = 0.38 e Å^−3
4144 reflections	Δρmin = −0.25 e Å^−3
160 parameters

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.

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

	x	y	z	Uiso*/Ueq
C1	−0.10106 (7)	0.63907 (5)	−0.90905 (14)	0.0221 (2)
H1A	−0.1031	0.6118	−1.0046	0.026*
C2	−0.16703 (7)	0.67885 (6)	−0.85429 (15)	0.0228 (2)
H2	−0.2226	0.6833	−0.9037	0.027*
C3	−0.13650 (7)	0.71161 (5)	−0.71119 (14)	0.0197 (2)
H3	−0.1674	0.7428	−0.6475	0.024*
C4	−0.05301 (7)	0.68995 (5)	−0.68053 (13)	0.0166 (2)
C5	0.00268 (7)	0.70402 (5)	−0.53978 (13)	0.01671 (19)
H5	−0.0111	0.7396	−0.4695	0.02*
C6	0.11919 (7)	0.68795 (5)	−0.35662 (12)	0.0178 (2)
H6A	0.1031	0.7323	−0.3214	0.021*
H6B	0.1817	0.6878	−0.383	0.021*
C7	0.10141 (7)	0.64064 (5)	−0.21366 (13)	0.0197 (2)
H7A	0.0423	0.648	−0.17	0.024*
H7B	0.1043	0.5957	−0.2565	0.024*
C8	0.16671 (7)	0.64896 (5)	−0.07220 (12)	0.0181 (2)
H8A	0.2242	0.6339	−0.1104	0.022*
H8B	0.1715	0.6955	−0.0437	0.022*
C9	0.18122 (6)	0.55911 (5)	0.10468 (13)	0.0175 (2)
H9	0.2242	0.5456	0.0274	0.021*
C10	0.16340 (7)	0.51857 (5)	0.24798 (12)	0.0170 (2)
C11	0.18834 (7)	0.45494 (5)	0.27627 (14)	0.0209 (2)
H11	0.2237	0.4292	0.2054	0.025*
C12	0.15141 (7)	0.43535 (5)	0.43018 (15)	0.0223 (2)
H12	0.1564	0.3939	0.4808	0.027*
C13	0.10686 (7)	0.48781 (5)	0.49267 (14)	0.0208 (2)
H13	0.0766	0.489	0.596	0.025*
N1	−0.03241 (6)	0.64556 (4)	−0.80283 (11)	0.01793 (18)
N2	0.06989 (6)	0.67017 (4)	−0.50660 (10)	0.01746 (18)
N3	0.14145 (6)	0.61245 (4)	0.07761 (10)	0.01782 (18)
N4	0.11336 (6)	0.53797 (5)	0.38170 (11)	0.01789 (18)
H1	0.0212 (9)	0.6300 (7)	−0.8218 (18)	0.021*
H4	0.0972 (9)	0.5786 (7)	0.4009 (17)	0.021*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0233 (5)	0.0212 (5)	0.0217 (5)	−0.0031 (4)	−0.0050 (4)	0.0004 (4)
C2	0.0170 (5)	0.0253 (5)	0.0260 (5)	−0.0030 (4)	−0.0039 (4)	0.0060 (4)
C3	0.0161 (4)	0.0209 (5)	0.0219 (5)	0.0011 (4)	0.0019 (4)	0.0041 (4)
C4	0.0170 (4)	0.0165 (4)	0.0163 (4)	0.0002 (3)	0.0010 (3)	0.0017 (3)
C5	0.0173 (4)	0.0172 (4)	0.0156 (4)	0.0001 (4)	0.0015 (4)	0.0011 (3)
C6	0.0182 (4)	0.0213 (5)	0.0139 (4)	0.0006 (4)	−0.0007 (3)	0.0008 (4)
C7	0.0217 (5)	0.0221 (5)	0.0153 (4)	−0.0018 (4)	−0.0024 (4)	0.0022 (4)
C8	0.0197 (4)	0.0201 (5)	0.0145 (4)	0.0018 (4)	−0.0008 (4)	0.0011 (4)
C9	0.0163 (4)	0.0205 (5)	0.0157 (4)	0.0016 (4)	−0.0006 (3)	−0.0012 (4)
C10	0.0160 (4)	0.0178 (4)	0.0172 (5)	0.0007 (4)	−0.0014 (3)	0.0006 (4)
C11	0.0212 (5)	0.0187 (5)	0.0229 (5)	0.0024 (4)	−0.0012 (4)	−0.0009 (4)
C12	0.0215 (5)	0.0184 (5)	0.0271 (6)	−0.0011 (4)	−0.0014 (4)	0.0058 (4)
C13	0.0170 (4)	0.0245 (5)	0.0210 (5)	−0.0008 (4)	0.0003 (4)	0.0053 (4)
N1	0.0181 (4)	0.0180 (4)	0.0177 (4)	0.0017 (3)	−0.0011 (3)	0.0003 (3)
N2	0.0192 (4)	0.0193 (4)	0.0139 (4)	0.0014 (3)	−0.0001 (3)	0.0015 (3)
N3	0.0194 (4)	0.0204 (4)	0.0136 (4)	0.0019 (3)	−0.0012 (3)	0.0010 (3)
N4	0.0173 (4)	0.0178 (4)	0.0185 (4)	0.0024 (3)	−0.0002 (3)	0.0016 (3)

Geometric parameters (Å, º)

C1—N1	1.3665 (14)	C7—H7B	0.99
C1—C2	1.3811 (17)	C8—N3	1.4684 (14)
C1—H1A	0.95	C8—H8A	0.99
C2—C3	1.4108 (17)	C8—H8B	0.99
C2—H2	0.95	C9—N3	1.2798 (14)
C3—C4	1.3886 (14)	C9—C10	1.4457 (15)
C3—H3	0.95	C9—H9	0.95
C4—N1	1.3776 (14)	C10—N4	1.3796 (14)
C4—C5	1.4469 (15)	C10—C11	1.3875 (15)
C5—N2	1.2806 (13)	C11—C12	1.4160 (16)
C5—H5	0.95	C11—H11	0.95
C6—N2	1.4680 (14)	C12—C13	1.3776 (16)
C6—C7	1.5287 (15)	C12—H12	0.95
C6—H6A	0.99	C13—N4	1.3675 (14)
C6—H6B	0.99	C13—H13	0.95
C7—C8	1.5264 (15)	N1—H1	0.903 (15)
C7—H7A	0.99	N4—H4	0.888 (15)
N1—C1—C2	108.59 (10)	N3—C8—H8A	109.3
N1—C1—H1A	125.7	C7—C8—H8A	109.3
C2—C1—H1A	125.7	N3—C8—H8B	109.3
C1—C2—C3	107.16 (10)	C7—C8—H8B	109.3
C1—C2—H2	126.4	H8A—C8—H8B	107.9
C3—C2—H2	126.4	N3—C9—C10	122.72 (10)
C4—C3—C2	107.47 (10)	N3—C9—H9	118.6
C4—C3—H3	126.3	C10—C9—H9	118.6
C2—C3—H3	126.3	N4—C10—C11	107.73 (9)
N1—C4—C3	107.70 (9)	N4—C10—C9	123.60 (9)
N1—C4—C5	123.24 (9)	C11—C10—C9	128.63 (10)
C3—C4—C5	128.79 (10)	C10—C11—C12	107.46 (10)
N2—C5—C4	122.34 (9)	C10—C11—H11	126.3
N2—C5—H5	118.8	C12—C11—H11	126.3
C4—C5—H5	118.8	C13—C12—C11	107.01 (10)
N2—C6—C7	110.97 (9)	C13—C12—H12	126.5
N2—C6—H6A	109.4	C11—C12—H12	126.5
C7—C6—H6A	109.4	N4—C13—C12	108.85 (10)
N2—C6—H6B	109.4	N4—C13—H13	125.6
C7—C6—H6B	109.4	C12—C13—H13	125.6
H6A—C6—H6B	108	C1—N1—C4	109.08 (9)
C8—C7—C6	111.31 (9)	C1—N1—H1	125.1 (9)
C8—C7—H7A	109.4	C4—N1—H1	124.7 (9)
C6—C7—H7A	109.4	C5—N2—C6	117.04 (9)
C8—C7—H7B	109.4	C9—N3—C8	116.85 (9)
C6—C7—H7B	109.4	C13—N4—C10	108.93 (9)
H7A—C7—H7B	108	C13—N4—H4	125.6 (9)
N3—C8—C7	111.77 (9)	C10—N4—H4	124.5 (9)
N1—C1—C2—C3	1.13 (13)	C10—C11—C12—C13	−1.30 (12)
C1—C2—C3—C4	−1.11 (12)	C11—C12—C13—N4	1.41 (12)
C2—C3—C4—N1	0.67 (12)	C2—C1—N1—C4	−0.74 (12)
C2—C3—C4—C5	−173.42 (10)	C3—C4—N1—C1	0.03 (12)
N1—C4—C5—N2	−9.46 (16)	C5—C4—N1—C1	174.52 (10)
C3—C4—C5—N2	163.79 (11)	C4—C5—N2—C6	−177.76 (9)
N2—C6—C7—C8	166.96 (9)	C7—C6—N2—C5	103.56 (11)
C6—C7—C8—N3	169.72 (9)	C10—C9—N3—C8	179.49 (9)
N3—C9—C10—N4	−11.13 (16)	C7—C8—N3—C9	102.49 (11)
N3—C9—C10—C11	166.20 (11)	C12—C13—N4—C10	−0.98 (12)
N4—C10—C11—C12	0.72 (12)	C11—C10—N4—C13	0.15 (12)
C9—C10—C11—C12	−176.95 (10)	C9—C10—N4—C13	177.96 (9)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N1—H1···N3^i	0.902 (14)	2.059 (14)	2.935 (2)	163.5 (13)
N4—H4···N2^ii	0.889 (14)	2.073 (14)	2.949 (2)	168.4 (12)

Symmetry codes: (i) x, y, z−1; (ii) x, y, z+1.