Crystal Engineering: Identification of a Unique Supramolecular


Crystal Engineering: Identification of a Unique Supramolecular...

0 downloads 147 Views 239KB Size

CRYSTAL GROWTH & DESIGN

Crystal Engineering: Identification of a Unique Supramolecular Synthon Based on CdO‚‚‚X Interaction in Halogen-Substituted Aromatic Carboxaldehydes

2003 VOL. 3, NO. 4 581-585

J. Narasimha Moorthy,*,† P. Venkatakrishnan,† Prasenjit Mal,† Seema Dixit,‡ and P. Venugopalan*,‡ Department of Chemistry, Indian Institute of Technology, Kanpur 208 016, India, and Department of Chemistry, Panjab University, Chandigarh 160 014, India Received January 5, 2003;

Revised Manuscript Received May 3, 2003

ABSTRACT: The well-known CdO‚‚‚X interactions control the crystal packing of o-haloaromatic carboxaldehydes as revealed by the X-ray crystal structures of a set of aldehydes 2-8. The intramolecular C-H‚‚‚X interaction causes the CdO‚‚‚X interactions to manifest in a readily recognizable pattern of association, i.e., synthon 1. The robustness of synthon 1 is amply suggested from the isostructurality exhibited by the dichloro- and dibromo-dialdehydes 3,6 and 4,7. Further, it is found that the aldehyde 2, which can exploit the CdO‚‚‚Br interactions in two distinct ways crystallizes in two crystal modifications. Introduction In crystal engineering,1 one of the goals is to recognize and develop supramolecular synthons2 that are robust to be exchanged from one network structure to another. Quite often, the molecular packing in organic solids can be readily predicted/understood based on well-recognized association patterns2-4 of certain functional groups that they contain.2,5,6 Thus, a priori knowledge of how certain functional groups or combinations thereof interact to result in specific supramolecular synthons that decisively control the crystal packing is most desirable.7,8 Because the recurring robust patterns exhibited by the functional groups not only aid one to predict but also design solid-state structures with predefined functions, the quest for new and novel supramolecular synthons continues unabated. In this context, the structure-directing influence of strong hydrogen bonds such as O-H‚‚‚O, O-H‚‚‚N, and N-H‚‚‚O has lent itself to identification of a variety of synthons.2,3,9 Conspicuously, the synthons not containing the so-called strong hydrogen bonds are only a few.5,10,11 During our recent investigations on solid-state photoreactivity of o-haloaromatic aldehydes,12 we have had an occasion to determine the X-ray crystal structures of two of the dihalo-substituted dialdehydes, i.e., 3 and 7 (Chart 1). A closer inspection of their crystal packing revealed a specific molecular packing mediated by CdO‚‚‚Br interactions. We recognized that the centrosymmetric pairs of the o-chloro/bromoaromatic aldehydes selfassociate to form dimers via two CdO‚‚‚Br interactions (synthon 1) in a fashion that resembles the selfassembly of carboxylic acids via the centrosymmetric dimer motif. This prompted us to undertake the present study to establish the recurrence and generality of synthon 1. Herein, we report our results on the identification, prevalence, and robustness of synthon 1 based on X-ray structural analyses of aromatic carboxalde* To whom correspondence should be addressed. Fax: 91-5122597436; Tel: 91-512-2597438. E-mail: [email protected]. † Indian Institute of Technology. ‡ Panjab University.

Chart 1

hydes 2-8. Further, we show how the manifestation of CdO‚‚‚Br interaction in two distinct ways does indeed lead to two crystal modifications in the case of dialdehyde 2. Results and Discussion The dialdehyde 2 was found to crystallize in two modifications (P1 h and P21/c); whereas the crystals grown from chloroform-hexane appeared more like thick blocks and corresponded to a triclinic crystal system, platelike monoclinic crystals were obtained from a combination of solvents, namely, methanol, chloroform, and hexanes. In Figure 1 are shown the crystal packing diagrams of both modifications. It can be readily recognized that the aldehydes in the triclinic modification (2T) form linear tapes via the synthon 1. The linear tapes are interconnected through weaker C-H‚‚‚O (involving the formyl oxygen and the aromatic hydrogen, dC‚‚‚O ) 3.45 Å and θC-H‚‚‚O ) 163.4°) and Br‚‚‚Br (d ) 3.62 Å) interactions.1,13,14 A similar scenario, i.e., formation of linear tapes, is observed with the monoclinic modification (2M), but with the difference that the linear tape is formed via CdO‚‚‚Br interactions between the halogen atom and the oxygen atom of the formyl group at 1,3 positions. Thus, the self-assembly in this modification occurs via a ring motif consisting of 12 atoms, as

10.1021/cg034001p CCC: $25.00 © 2003 American Chemical Society Published on Web 05/21/2003

582

Crystal Growth & Design, Vol. 3, No. 4, 2003

Moorthy et al.

Figure 3. The crystal-packing diagram of dialdehyde 5. The intermolecular CdO‚‚‚Br interactions are shown with a dotted line. Figure 1. The crystal packing diagrams of triclinic (top) and monoclinic (bottom) modifications of dialdehyde 2.

Figure 4. The crystal packing of the dialdehyde 7. Figure 2. The crystal packing of the dialdehyde 3.

opposed to 10 atoms in the triclinic modification. As in the case of the latter, the linear tapes are interconnected, but in the present instance through C-H‚‚‚O hydrogen bonds involving the formyl hydrogen and oxygen (dC‚‚‚O ) 3.43 Å and θC-H‚‚‚O ) 133.4°). It is noteworthy that in both modifications, the carbonyl oxygens orient themselves away from the o-halogen atoms as a result of intramolecular C-H‚‚‚X interactions.15 A linear tape pattern (based on synthon 1) as that observed for 2T is observed for aldehydes 316 and 4, which were found to be isostructural (see, Supporting Information). The crystal packing of 3 is shown in Figure 2. Here again, the tapes are interconnected in both cases through C-H‚‚‚O hydrogen bonds involving the methyl hydrogen and the formyl oxygen (in 3, dC‚‚‚O ) 3.64 Å and θC-H‚‚‚O ) 147.4°). We examined the crystal packing of dialdehyde 5 with an objective of establishing the persistence of synthon 1 to variation in the substituents. While the synthon is robust to substitution of hydrogens in 2 by methyls as in 4, an entirely different crystal packing is observed for substitution with methoxy groups as in 5 (Figure 3). The reason for such a representative shift in the crystal packing is because the formyl hydrogens in 5 intramolecularly hydrogen bond to methoxy oxygens so that the formyl oxygens are oriented toward the bromo groups, while it is exactly the opposite in 2-4; evidently, the preference of the formyl hydrogen to interact with

the methoxy oxygen over that of the bromine is established. This conformational change thus prevents synthon 1 from operating in 5. Nonetheless, it is important to note that the crystal packing is mediated exclusively by CdO‚‚‚Br interactions. It is thus obvious that for synthon 1 to be structure-determining, the orientation of formyl oxygen away from the o-halogen is imperative. Indeed, one does observe synthon 1 in the crystal packing of the dialdehyde 6 and 7,16 for which such a criterion is met for only one of the formyl groups. Again, the dialdehydes 6 and 7 are found to be isostructural. The crystal packing of 7 is typically shown in Figure 4. The zero-dimensional dimers resulting from synthon 1 associate further via CdO‚‚‚Br interactions (dCdO‚‚‚Br ) 3.40 Å and θC-Br‚‚‚O ) 160.3°) involving the second formyl carbonyl oxygen and halogen atoms; although the distance for this interaction is slightly larger than the van der Waals sum, the observed angular parameter is strongly indicative of the attractive and anisotropic interaction (vide infra). The trialdehyde 8 was designed and synthesized to mimic the features of trimesic acid;17 as each o-bromobenzene carboxaldehyde moiety should typically behave as a site for linear self-assembly through synthon 1, we anticipated the crystal packing to assume a rosette organization. Ostensibly, such a crystal packing should lead to creation of cavity for guest inclusion. In the absence of an appropriate guest molecule, the observed crystal packing in Figure 5 clearly attests to the robustness of synthon 1. Clearly, a zigzag tape pattern

Identification of a Unique Supramolecular Synthon

Crystal Growth & Design, Vol. 3, No. 4, 2003 583

Table 1. Geometrical Parameters for CdO‚‚‚X Interaction in Aldehydes 2-8 compound

2T

2M

3

4

5

6

7

8(1)

8(2)

8(3)

dCdO‚‚‚X/Å θC-X‚‚‚O/°

3.44 151.9

3.30 143.8

3.27 159.6

3.27 161.1

3.10 168.9

3.23 157.4

3.23 156.6

3.10 172.7

3.22 163.4

3.04 169.7

Figure 5. The crystal-packing of the trialdehyde 8. Notice that the synthon 1 leads to a zigzag tape.

is formed along the b-axis via the operation of synthon 1. These tapes are interconnected again through CdO‚‚‚Br interactions along c-axis, as shown in Figure 5. The existence of attractive interactions between halogens and n-donor atoms was recognized almost four decades ago by Hassel and co-workers;18 these interactions were attributed to the “charge-transfer” type in analogy to the transfer of charge from Lewis base to a Lewis acid.19,20 The prevalence of these interactions between C-X (Cl, Br, and I) as an acceptor and a heteroatom (O, N, S, Se, etc.) as a donor, and their manifestation in the crystal structures of a variety of molecules to result in specific packing patterns were analyzed within the next few years.19,21,22 An early database investigation of the geometry of C-I‚‚‚O interactions was performed by Murray-Rust and Motherwell.23 A thorough analysis of the angular preferences for interactions around the halogen centers in the crystal structures of halogen-containing compounds reported until then was subsequently conducted by Ramasubbu et al.24 Their analyses revealed that the electrophiles tend to approach the halogens of C-X (X ) Cl, Br, I) bonds at an angle of ca. 100.0°, while the nucelophiles do so at ca. 165.0°. Thus, the interaction of X with other atoms (depending on the nature and direction of approach) were clearly shown to be anisotropic and the attractive interactions were interpreted as arising from “electrophile-nucleophile pairing”. On the basis of a closely related study involving Cl‚‚‚Cl interactions, Price et al.25 argued that the highly directional short intermolecular contacts are observed as a result of close packing of atoms with nonspherical atomic charge distribution26 (polar flattening) and not because of any specific attractive interactions. An exptrapolation of this to short intermolecular contacts observed between halogens (Cl, Br, and I) and heteroatoms calls into question the “charge-transfer” or “electrophile-nucleophile pairing” character traditionally associated with such short contacts. The contention of Price and co-workers has, however, been disputed recently by Lommerse et al.27 The latter have shown that highly directional and attractive intermolecular contacts between halogens (Cl, Br, and I) and electronegative atoms can indeed exist and that the attractive

nature of interactions is due to electrostatic effects with polarization, charge-transfer, and dispersion contributions all playing an important role. A perusal of the geometrical parameters in Table 1 readily suggests that the CdO‚‚‚X interactions control the crystal packing in all of the aldehydes 2-8; the electronegative carbonyl oxygen lies along the C-X bond axis with an angle around ca. 150-170° in all cases with interpenetration, except in 2T,28 of van der Waals volumes (see values of d in Table 1, the van der Waals radii of O and Br atoms are 1.52 and 1.85 Å,29 respectively). That the CdO‚‚‚X interaction is indeed structuredetermining in o-haloaromatic aldehydes is further suggested from the isostructural relationship exhibited by the pairs 3 and 4 and 6 and 7;30,31 the crystal packing is robust to the exchange of chlorine atoms in 3 and 6 by bromine atoms as in 4 and 7. That the interactions are polarization-induced, in addition to being anisotropic, is clearly evident from the comparison of geometrical parameters for CdO‚‚‚X interaction in the chloro and bromo analogues, i.e., 3 and 4, and 5 and 6. One observes that the CdO‚‚‚X distance is accidentally the same for the pairs 3 and 4, and for 6 and 7, although the van der Waals radius of Cl atom is lesser than that of the Br atom by ca. 0.1 Å. This suggests that the interaction is weaker in the case of Cl-substituted aldehydes, if one relates the distance to the extent of interaction. The observed result is thus in line with the tendency of the halogens to form short C-X‚‚‚heteroatom contacts according to the order of their polarizabilities (I > Br > Cl). Although recognized long ago, the interactions between halogens (Cl, Br, and I) and n-donor atoms have now come to be called “halogen bonds”,32 and their application in supramolecular synthesis has been well proven. The synthon 1 reported herein is unique in that the simultaneous operation of two CdO‚‚‚X (Cl/Br) interactions by virtue of the ortho relationship between halogen and the CdO groups leads to a ring motif, which is heretofore unknown. It is remarkable that the dialdehyde 2 crystallizes in two modifications based on two distinct modes in which it may exploit the CdO‚‚‚Br interactions. The selfassembly in both cases via CdO‚‚‚Br interactions leads eventually to the formation of tapes. This dimorphic behavior of the dialdehyde 2 points one to consider, in the context of crystal engineering, not only the patterns of associations of the functional groups, but also the number of different ways in which the functional groups may interact. Polymorphism is of tremendous contemporary interest,33,34 and it has been noted that the occurrence of polymorphism is more probable for compounds that contain a multiple number of functional groups of the same type.35 While this is indeed the case, the results observed with the aldehyde 2 also suggest that the number of polymorphs that may be readily obtainable may be, to some extent, assessed from the alternative, but equally efficient (in terms of crystal lattice stabilization) packing modes.

584

Crystal Growth & Design, Vol. 3, No. 4, 2003

In summary, the results described herein constitute identification of a unique supramolecular motif 1 (based on CdO‚‚‚X interactions) for crystal packing of ohalogen-substituted aromatic carboxaldehydes. The synthon 1 formally resembles the centrosymmetric carboxylic acid dimer motif and is robust to permit predictability with regard to the molecular self-assembly as revealed from a limited but sufficient number of examples. Further, it is found that the aldehyde 2, which can exploit the CdO‚‚‚Br interactions in two distinct ways, crystallizes in two crystal modifications. Experimental Section Synthesis of the Aldehydes 2-8. The aldehydes 2,36 5,37 and 838,39 were synthesized following the reported procedures and characterized completely (IR, 1H, 13C NMR spectroscopy). The synthesis of other aldehydes, viz., 3, 4, 6, and 7, has been previously reported by us.12 X-ray Crystal Structure Analyses of the Aldehydes 2, 4, 5, 6, and 8. The protocol adopted for the X-ray crystal structure determination of aldehydes 2, 4, 5, 6, and 8 is described below. The X-ray structure determination details for 3 and 7 have been previously reported.12 The individual parameters and refinement conditions pertaining to each compound are recorded in the table entitled “Crystal Data and Structure Refinement” in Supporting Information. The aldehyde 2 was found to crystallize in two different modifications; while slow evaporation of a solution of 2 in methanol-chloroform-hexanes afforded colorless monoclinic plates (mp. 182 °C), the solution in chloroform-hexanes mixture led to isolation of colorless triclinic thick blocks (mp. 190 °C). The crystals of all other aldehydes, i.e., 4, 5, 6, and 8, amenable for X-ray crystallographic studies were grown by slow evaporation of their solutions in chloroform-hexanes mixtures. A good single crystal in each case was mounted along its largest dimension and used for data collection. The intensity data were collected on a Siemens P4 single-crystal diffractometer equipped with Molybdenum sealed tube (λ ) 0.71073 Å) and highly oriented graphite monochromator. The lattice parameters and standard deviations were obtained by a leastsquares fit to 40 reflections (10.0° < 2θ < 30.0°). The data were collected by 2θ-θ scan mode with a variable scan speed ranging from 2.0° to a maximum of 60.0°/min. Three reflections were used to monitor the stability and orientation of the crystal and were remeasured after every 97 reflections. Their intensities did not change significantly during the entire data collection period. All other relevant information about the data collection is presented in the tables for each compound. The structure was solved in each case by Direct Methods using the SHELX-9740 package and also refined using the same program. All the non-hydrogen atoms were refined anisotropically. The hydrogen atoms were included in the ideal positions with fixed isotropic U values and they were riding with their respective non-hydrogen atoms. A weighting scheme of the form w ) 1/[σ2(Fo2) + (aP)2 + bP] was used. The difference Fourier map, after the refinement, was essentially featureless in all the cases. Final atomic coordinates, bond lengths, bond angles, anisotropic displacement parameters, hydrogen atom coordinates, and torsion angles have been recorded in the tables for all of the aldehydes.

Acknowledgment. We thank the Department of Science and Technology (DST), India, for finanical support. P.V. and P.M. are grateful to CSIR and IIT Kanpur, India, for junior and senior research fellowships, respectively. We thank Dr. P. Dastidar (Bhavnagar, India) for his help in performing the Cambridge Structural Database search. We thank one of the reviewers for critical comments and suggestions. Supporting Information Available: The crystal data (details of crystal structure determination, refinement, atomic

Moorthy et al. coordinates, temperature factors, tables of bond lengths, and bond angles) for aldehydes 2T, 2M, 4, 5, 6, and 8. This material is available free of charge via the Internet at http://pubs. acs.org.

References (1) Desiraju, G. R. Crystal Engineering: The Design of Organic Solids; Elsevier: Amsterdam, The Netherlands, 1989. (2) Desiraju, G. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 2311. (3) Etter, M. C. Acc. Chem. Res. 1990, 23, 120. (4) MacDonald, J. C.; Whitesides, G. M. Chem. Rev. 1994, 94, 2383. (5) Desiraju, G. R. Chem. Commun. 1997, 1475 and references therein. (6) Lawrence, D. S.; Jiang, T.; Levett, M. Chem. Rev. 1995, 95, 2229. (7) The functional group approach for crystal structure design and prediction is not entirely viable. Recent investigations have shown that the simultaneous operation of weaker interactions and strong and directional O-H‚‚‚O hydrogen bonds leads to departure from the expectations based on molecular association via strong interactions alone, see Desiraju, G. R.; Steiner, T. The Weak Hydrogen Bond in Structural Chemistry and Biology; Oxford University Press: Oxford, U. K., 1999. (8) For our recent contribution on the influence of C-H‚‚‚X interactions on self-assembly of benzene carboxylic acids, see Moorthy, J. N.; Natarajan, R.; Mal, P.; Venugopalan, P. J. Am. Chem. Soc. 2002, 124, 6530. (9) Allen, F. H.; Motherwell, W. D. S.; Raithby, P. R.; Shields, G. P.; Taylor, R. New J. Chem. 1999, 23, 25. (10) Navon, O.; Bernstein, J.; Khodorkovsky, V. Angew. Chem., Int. Ed. Engl. 1997, 36, 601. (11) Bosch, E.; Barnes, C. L. Cryst. Growth Des. 2002, 2, 299. (12) Moorthy, J. N.; Venkatakrishnan, P.; Mal, P.; Venugopalan, P. J. Org. Chem. 2003, 68, 327. (13) Sarma, J. A. R. P.; Desiraju, G. R. Acc. Chem. Res. 1986, 19, 222. (14) Desiraju, G. R.; Parthasarathy, R. J. Am. Chem. Soc. 1989, 111, 8725. (15) We have preferred to term “C-H‚‚‚X” as an interaction to a “hydrogen bond”. The question of whether the C-H‚‚‚X interaction involving formyl hydrogens can be called a hydrogen bond will be addressed elsewhere. (16) The details of the crystal structure determination for 3 and 7 have been already reported, see ref 12. (17) Herbstein, F. H. In Comprehensive Supramolecular Chemistry; MacNicol, D. D., Toda, F., Bishop, R., Eds.; Pergamon Press: Oxford, 1996; Vol. 6, p 61. (18) Hassel, O.; Hvoslef, J. Acta Chem. Scand. 1954, 8, 873. (19) Hassel, O.; Romming, C. Quart. Rev. 1962, 1, 16. (20) Gaultier, J.; Hauw, C.; Schvoerer, M. Acta Crystallogr. 1971, B 27, 2199. (21) Bent, H. A. Chem. Rev. 1968, 68, 587. (22) Bernstein, J.; Cohen, M. D.; Leiserowitz, L. In The Chemistry of the Quinonoid Compounds. Part 1; Patai, S., Ed.; John Wiley & Sons: New York, 1974; p 37. (23) Murray-Rust, P.; Motherwell, W. D. S. J. Am. Chem. Soc. 1979, 101, 4374. (24) Ramasubbu, N.; Parthasarathy, R.; Murray-Rust, P. J. Am. Chem. Soc. 1986, 108, 4308. (25) Price, S. L.; Stone, A. J.; Lucas, J.; Rowland, R. S.; Thornley, A. E. J. Am. Chem. Soc. 1994, 116, 4910. (26) Nyburg, S. C.; Wong-Ng, W. Proc. R. Soc. London A 1979, 367, 29. (27) Lommerse, J. P. M.; Stone, A. J.; Taylor, R.; Allen, F. H. J. Am. Chem. Soc. 1996, 118, 3108. (28) The CdO‚‚‚Br distance of 3.44 Å in the case of triclinic modification, i.e., 2T, is larger than the sum of van der Waals radii. This is presumably due to crystal packing forces. The directionality as reflected from the C-X‚‚‚OdC angle leaves no doubt as to the anisotropic interaction. Indeed, the assessment of the strength of an interaction based on distance criterion alone can be misleading. It has been shown that the position of minimum in the intermolecular potential is ca. 0.4 Å larger than the van der Waals distance, see Dance, I. New J. Chem. 2003, 27, 22. (29) Bondi, A. J. Phys. Chem. 1964, 68, 441.

Identification of a Unique Supramolecular Synthon (30) A similar isostructurality in the crystal structures of 2,5dichloro- and 2,5-dibromo-1,4-benzoquinones has been attributed to the effectiveness of CdO‚‚‚X interactions, see ref 27. For discussion on the crystal packing of a variety of halogen-substituted benzoquinones, see ref 22. (31) The CSD 2003 search for structures containing o-halobenzaldehyde moiety (halogen ) F, Cl, Br, and I) was performed to establish the occurrence of synthon 1. The search, however, yielded only eight hits. Of these, the dimeric motif 1 is observed only in two compounds, viz., 1-bromonaphthalene-2-carboxaldehyde (code: FADWIQ; dCdO‚‚‚Br ) 3.88 Å and θC-Br‚‚‚O ) 149.0 °) and 2,4-dibromo-3-methoxy6-nitrobenzaldeyde (code: XEGNIG; dCdO‚‚‚Br ) 3.26 Å and θC-Br‚‚‚O ) 168.7°). The absence of the dimeric motif 1 in the remaining compounds is found to be due to the presence of other functional groups such as Br, OH, COOH, etc., which are capable of stabilizing the crystal lattice through interactions of their own. The alternative and equally or better-stabilizing interactions appear to obviate the adoption of dimeric motif 1 in all these cases.

Crystal Growth & Design, Vol. 3, No. 4, 2003 585 (32) Metrangolo, P.; Resnati, G. Chem. Eur. J. 2001, 7, 2511. (33) Desiraju, G. R. Science 1997, 278, 404 and references therein. (34) Bernstein, J.; Davey, R. J.; Henck, J.-O. Angew. Chem., Int. Ed. Engl. 1999, 38, 3440. (35) Sarma, J. A. R. P.; Desiraju, G. R. In Crystal Engineering. The Design and Applications of Functional Solids; Seddon, K. R.; Zaworotko, M, Eds.; Kluwer: Dordrecht, 1999; p 325. (36) Thulin, B.; Wennerstrom, O.; Somfai, I.; Chmielarz, B. Acta Chem. Scand. 1977, B 31, 135. (37) Syper, L.; Mlochowski, J. Synthesis 1984, 747. (38) Anthony, J. E.; Khan, S. I.; Rubin, Y. Tetrahedron Lett. 1997, 38, 3499. (39) Bruns, D.; Miura, H.; Stanger, A.; Vollhardt, K. P. C. Org. Lett. 2003, 5, 549. (40) SHELX-97 Sheldrick, G. M. Program for the Solution and Refinement of Crystal Structures; University of Go¨ttingen, Go¨ttingen, Germany, 1997.

CG034001P