ARTICLE pubs.acs.org/crystal
Avoiding “Synthon Crossover” in Crystal Engineering with Halogen Bonds and Hydrogen Bonds Published as part of the Crystal Growth & Design virtual special issue on Halogen Bonding in Crystal Engineering: Fundamentals and Applications Christer B. Aaker€oy,* Prashant D. Chopade, and John Desper Department of Chemistry, Kansas State University, Manhattan, Kansas 66506, United States
bS Supporting Information ABSTRACT: A combination of halogen bonds and hydrogen bonds has been used for effective assembly of three co-crystals containing desired one-dimensional architectures where the interactions within each assembly can be modulated using tunable electrostatics. The central tecton in these structures, 2-aminopyrazine, can interact with suitable hydrogen-bond donors and halogen-bond donors simultaneously without any “synthon crossover”. When different 2-aminopyrazine-based molecules are co-crystallized with 1,4-diiodo-tetrafluorobenzene (DITFB), a N 3 3 3 I halogen bond is driving the co-crystal synthesis in each case, whereas the N H 3 3 3 N/N 3 3 3 H N homosynthon is responsible for creating infinite chains of alternating pyrazine and DITFB molecules in the three crystal structures. The importance of electrostatic and geometric complementarity for refining strategies for supramolecular synthesis is emphasized.
fundamental principle of supramolecular chemistry1 de scribes a molecule as being built from atoms connected by covalent bonds, whereas a supermolecule is constructed from molecules using intermolecular interactions such as hydrogen bonds,2 halogen bonds,3 and π π interactions as the connectors.1,4 One of the biggest challenges in supramolecular synthesis arises due to the reversible nature of these interactions which effectively restricts noncovalent synthesis to one-pot processes.5 An added complication with using relatively weak chemical bonds as primary synthetic tools is, of course, that the outcome of the supramolecular reaction can be quite unpredictable, which explains the ongoing quest for more robust synthons that can be combined into reliable and reproducible noncovalent synthesis. Hydrogen bonds remain the most commonly utilized tools in the assembly of co-crystals due to their relatively high strength and directionality,2,6 and many successful studies have been reported where overall assembly of binary and ternary cocrystals is guided by a synthetic scheme based on hydrogen bonds of varying strengths in a predictable manner. However, for designing co-crystals of higher complexity, a strategy that relies solely on hydrogen bonds could soon fail because of unavoidable competition between the intended hydrogen-bond donors/acceptors which could lead to “synthon crossover”. Supramolecular strategies, that can accommodate two or more different noncovalent interactions in such a way that they are unlikely to interfere with each other, would, in principle, alleviate this problem. A suitable accompaniment to a hydrogen-bond based strategy may be provided by halogen bonds, which are typically formed between activated iodine or bromine atoms (the halogen-bond donor) and an appropriate halogen-bond acceptor (electron-pair donor)
A
r 2011 American Chemical Society
such as an N-heterocycle.7 Although halogen bonding is similar to hydrogen bonding in some regards,3a a halogen bond will mostly involve single-point synthons,3 whereas hydrogen-bonded synthons are often multipoint interactions, Scheme 1. As part of our continuing efforts to explore ways to combine hydrogen- and halogen-bond interactions in order to minimize synthon crossover,8 we decided to employ 2-aminopyrazine as a tecton that can interact simultaneously via hydrogen bonds and halogen bonds at two distinct binding sites: a two-point amino: N-pyrazine end and a single point N-pyrazine end.8c Recently, we demonstrated that 2-aminopyrazine can form binary co-crystals with ditopic hydrogen-bond (HB)/halogen-bond (XB) donor molecules using a robust and predictable combination of HB/ XB based synthons (Scheme 2)8c where structural control was attained without synthon crossover due to the specific geometric complementarity of the XB and HB sites, respectively. In previous work, Nangia and co-workers also demonstrated that carboxylic acid 3 3 3 py (HB) and iodo 3 3 3 nitro (XB) synthons can exist side-by-side in co-crystals of nitrobenzoic acids and iodopyridine.9 Of course, in the absence of a strong and complementary twopoint hydrogen-bond donor/acceptor moiety such as a carboxylic acid, the amino-pyrazine site is perfectly capable of forming a self-complementary homo synthon, Scheme 3b. A CSD search on 2-aminopyrazine and its derivatives uncovered several examples of HB homosynthon formation between 2-aminopyrazine molecules.10 In addition, this homodimer is frequently Received: July 15, 2011 Revised: October 27, 2011 Published: November 16, 2011 5333
dx.doi.org/10.1021/cg2009013 | Cryst. Growth Des. 2011, 11, 5333–5336
Crystal Growth & Design Scheme 1. Examples of Commonly Occurring Supramolecular Synthons, (a) Acid Acid, (b) Amide Amide, (c) Acid Amide, (d) Acid Aminopyridine, (e) Iodo-N(py)
ARTICLE
Scheme 4. MEP Surface Calculations of 2-Aminopyrazine, B1; 2-Amino-5-bromopyrazine, B2; and 2-Amino-3,5-dibromopyrazine, B3a
a
Scheme 2. Infinite 1-D Chain Formation via Combination of HB and XB Based Synthons (X = I, Br).8c
Scheme 3. Observed Synthons in the CSD Database Search for (a) Co-Crystals of Substituted Pyridines and 1,4-Diiodotetrafluorobenzene, (B) Substituted 2-Aminopyrazine by Itself, and (c) Its Extended Network
accompanied by an anti NH(amino) 3 3 3 N4(pyrazine) secondary interaction (Scheme 3c). Consequently, a potential halogen bond, Scheme 3a, would have to compete successfully with the N H 3 3 3 N hydrogen bond if co-crystal formation is to take place. Literature reports indicate that this may be possible, based on the estimated strength, ∼24 kJ/mol, of an activated I 3 3 3 N(pyridine) interaction.8a,11 The goal of this study is to map out part of the structural landscape surrounding supramolecular reactions between 2-aminopyrazine based compounds and a powerful halogen-bond donor with a view to answering the following questions: (1) Can we synthesize co-crystals of 2-aminopyrazine and its derivatives using halogen bonds? (2) Will an N 3 3 3 I halogen bond prevail in the face of a potential N H 3 3 3 N hydrogen bond? (3) Can the strength and geometry of the halogen bond be modulated as a function of charge on the halogen-bond acceptor? We opted for 1,4-diiodotetrafluorobenzene (DITFB) as the halogen-bond donor due to its proven ability to form halogen bonding with N-heterocyclic acceptors (Scheme 3a).12
Numbers reflect the MEPs on N4 nitrogen atom in each molecule.13
We also wanted to examine the role of the electrostatic charge at the intended halogen-bond acceptor site (N4 on 2-aminopyrazine). We therefore attached one and two electron-withdrawing substituents, respectively, to the aminopyrazine backbone (Scheme 4) in order to alter the charge on the halogen-bond acceptor. Bromination (using N-bromosuccinamide) of 2aminopyrazine (B1) at 0 °C yielded 2-amino-5-bromopyrazine (B2) and 2-amino-3,5-dibromopyrazine (B3). Since most halogen bonds are primarily electrostatic in nature, a decrease in negative charge at the XB acceptor site should result either in an increase in the N4 3 3 3 I intermolecular distance or in the complete absence of the halogen bond upon moving from B1 to B3. To determine how the presence of the electron-withdrawing groups influence the charge on the heterocyclic nitrogen atoms, we used molecular electrostatic potential (MEP) surface calculations,13,14 which showed that the charge on the N4 nitrogen atom of 2-aminopyrazine indeed decreased upon consecutive brominations (Scheme 4). Co-crystallizations between B1 B3 and DITFB yielded crystals suitable for structure determinations using single-crystal X-ray diffraction, Table 1. The supramolecular reaction between B1 (containing the most basic halogen-bond acceptor in this study) and DITFB yielded a co-crystal with a 2:1 stoichiometry. The crystal structure determination of B1 3 DITFB shows an anticipated twopoint homosynthon (Scheme 3b) comprising two symmetry related NH 3 3 3 N hydrogen bonds, N(21) 3 3 3 N(22) 3.006(4) Å. The driving force for the co-crystal formation is a halogen bond between an iodine atom and single-point N-heterocyclic acceptor site, N(24) 3 3 3 I(1) 2.822(3) Å. The combination of these two different interactions results in an infinite 1-D chain, Figure 1. The remaining amino N H hydrogen-bond donor forms a close contact with the middle part of an iodine atom on a neighboring molecule with a geometry that is akin to a Type II halogen bond; the C I 3 3 3 H(-N) angle is approximately 92°. The noncovalent reaction between B2 and DITFB also produced a co-crystal with a 2:1 stoichiometry as expected. The driving force for the construction of this binary compound is a one-point I(1) 3 3 3 N(24) 3.025(2) Å halogen bond. In addition, a homomeric self-complementary synthon, N(21) 3 3 3 N(22) Å, 3.086(3) Å serves to create an infinite 1-D chain of alternating molecules of B2 and DITFB, Figure 2, similar to what was observed in the crystal structure of B1 3 DITFB. The remaining N H hydrogenbond donor does not participate in any notable short contacts in this structure. The final reaction between 2-amino-3,5-dibromopyrazine, B3, the weakest base of the three acceptors and DITFB also resulted in a co-crystal with a 2:1 stoichiometry. The co-crystal assembly is affected by the desired halogen bond, I(1) 3 3 3 N(24) 3.0459(18) Å, Figure 3, and the primary motif in this crystal structure, the infinite chain of alternating B3 and DITFB molecules, is completed with the aid of a self-complementary 5334
dx.doi.org/10.1021/cg2009013 |Cryst. Growth Des. 2011, 11, 5333–5336
Crystal Growth & Design
ARTICLE
Table 1. Crystallographic Data for B1 3 DITFB, B2 3 DITFB, and B3 3 DITFB B1 3 DITFB
B2 3 DITFB
B3 3 DITFB
formula moiety
(C4H5N3)2(C6I2F4)
(C4H4N3Br)2(C6F4I2)
(C4H3N3Br2)2(C6F4I2)
empirical formula
C14H10F4I2N6
C14H8Br2F4I2N6
C14H6Br4F4I2N6
molecular weight
592.08
749.88
907.69
color, habit
colorless plate
orange prism
colorless prism
crystal dimensions
0.24 0.22 0.12 mm
0.26 0.20 0.14 mm
0.26 0.22 0.14 mm
temp, K
120
120
120
crystal system
monoclinic
monoclinic
triclinic
space group, Z a, Å
P2(1)/c, 2 7.9013(6)
P2(1)/n, 2 5.6316(6)
P1, 1 5.6428(4)
b, Å
5.6116(4)
8.1219(9)
9.7705(7)
c, Å
20.1258(16)
21.226(2)
10.4014(7)
α, °
90.00
90.00
100.821(3)
β, °
101.260(4)
93.035(4)
103.743(3) 93.752(3)
γ, °
90.00
90.00
V, Å3
875.18(11)
969.48(18)
543.49(7)
X-ray wavelength μ, mm 1
0.71073 3.645
0.71073 7.416
0.71073 10.294
absorption corr
multiscan
multiscan
multiscan
trans min/max
0.4749/0.6688
0.2487/0.4233
0.1749/0.3267
collected
10448
11360
21278
independent
3084
3386
4133
observed
2633
3034
3850
threshold expression R1 (observed)
>2σ(I) 0.0336
>2σ(I) 0.0296
>2σ(I) 0.0259
reflections
wR2 (all)
0.0808
0.0668
0.0646
S
0.739
1.056
1.064
Figure 1. Infinite 1-D chains produced through a combination of hydrogen- and halogen bonds in the crystal structure of B1 3 DITFB.
Figure 2. 1-D motif in the crystal structure of B2 3 DITFB constructed from HB homosynthons and near-linear N 3 3 3 I halogen bonds (the latter is responsible for the primary assembly of the co-crystal).
homosynthon, N(13) 3 3 3 N(14), 3.051(3) Å. In this crystal structure, the nearest neighbor to the remaining N H hydrogenbond donor is one of the fluorine atoms; the approach is sideways, Type II interaction, as the C F 3 3 3 N angle is close to 108°. As expected for attractive intermolecular interactions, the N 3 3 3 I distances in all three structures are shorter than the sum of the van der Waals radii of the participating atoms. Furthermore, the N 3 3 3 I distance increases as the charge on the acceptor site N4 decreases: B1 3 DITFB 2.822(3) Å; B2 3 DITFB 3.025(2) Å, and B3 3 DITFB 3.046(2) Å. Despite the lower charge at the binding site N4, the I 3 3 3 N interaction prevailed against the potentially competing N 3 3 3 H N(amine) hydrogen bond. A space filling representation also indicates that the observed lengthening of the I 3 3 3 N bond is unlikely to be due to steric congestion and can therefore be ascribed to electrostatics (Figure 4). These findings are also in agreement with previous
Figure 3. Part of an infinite chain in the structure of B3 3 DITFB produced through hydrogen bonds and halogen bonds.
suggestions that halogen bonding involves some degree of electron donation from the halogen-bond acceptor (N4 in this case) to the antibonding bonding orbital (σ*) of C I.15 The C I bond is longer, 2.095(3) Å in B1 3 TFDIB (which contains the shorter I 3 3 3 N distance), than in the other two structures (B2 3 TFDIB and B3 3 TFDIB), 2.085(3) and 2.088(2) Å, respectively. 5335
dx.doi.org/10.1021/cg2009013 |Cryst. Growth Des. 2011, 11, 5333–5336
Crystal Growth & Design
Figure 4. Space-filling representation of a portion of the crystal structure of B3 3 DITFB.
Systematic co-crystallization reactions of DITFB with B1 B3 enabled the assembly of predictable 1-D chains in each of three new crystal structures, where the nature of the specific halogen bond directly responsible for the co-crystal assembly in each case, I 3 3 3 N(py), could be altered by modulating the electrostatic charge on the acceptor site through simple covalent modifications. The self-complementary homosynthon N H 3 3 3 N of 2-aminopyrazine was robust, and appeared intact in each of the three structures. Despite possibilities for “synthon crossover” each primary interaction showed high fidelity, as the amino anti proton was unable to compete with I 3 3 3 N(pyrazine) interaction. The anti proton (of amino group) did not establish any structural patterns, and it is therefore unlikely that it plays any significant structure-directing role. The effective use of different intermolecular interactions in the synthesis of supramolecular architectures can be facilitated by ensuring that geometric complementarity minimizes possible “synthon crossover” during the assembly process. A two-point contact is more easily incorporated within a hydrogen-based synthon, whereas the vast majority of halogen-bond interactions are based on single point halogen atom 3 3 3 lone pair synthons. Since hydrogen bonds and halogen bonds are frequently of comparable strength, driven primarily by electrostatic attractions, it is likely going to be helpful to utilize any geometric bias that these interactions display, when devising strategies for crystal engineering of multicomponent systems with an increased degree of structural complexity.
’ ASSOCIATED CONTENT
bS
Supporting Information. X-ray crystallographic files in CIF format for structures B1 3 DITFB, B2 3 DITFB, and B3 3 DITFB. Synthesis, characterization of all the compounds, details of the crystallographic work. This information is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT We are grateful for financial support from NSF (CHE0957607) and from the Johnson Center for Basic Cancer Research.
ARTICLE
(e) Moulton, B.; Zaworotko, M. J. Chem. Rev. 2001, 101, 1629. (f) Wenger, M.; Bernstein, J. Angew. Chem., Int. Ed. 2006, 45, 7966. (g) Childs, S. L.; Hardcastle, K. I. CrystEngComm 2007, 9, 64. (h) Bosch, E. CrystEngComm 2007, 9, 191. (i) Pedireddi, V. R.; Chatterjee, S.; Ranganathan, A.; Rao, C. N. R. J. Am. Chem. Soc. 1997, 119, 10867. (3) (a) Metrangolo, P.; Neukirch, H.; Pilati, T.; Resnati, G. Acc. Chem. Res. 2005, 38, 386. (b) Metrangolo, P.; Pilati, T.; Resnati, G.; Stevenazzi, A. Chem. Commun. 2004, 1492. (c) De Santis, A.; Forni, A.; Liantonio, R.; Metrangolo, P.; Pilati, T.; Resnati, G. Chem.—Eur. J. 2003, 9, 3974. (d) Cincic, D.; Friscic, T.; Jones, W. J. Am. Chem. Soc. 2008, 130, 7524. (e) Cincic, D.; Friscic, T.; Jones, W. Chem.—Eur. J. 2008, 14, 747. (f) Shirman, T.; Freeman, D.; Posner, Y. D.; Feldman, I.; Facchetti, A.; van der Boom, M. E. J. Am. Chem. Soc. 2008, 130, 8162. (g) Nguyen, H. L.; Horton, P. N.; Hursthouse, M. B.; Legon, A. C.; Bruce, D. W. J. Am. Chem. Soc. 2004, 126, 1617. (h) Triguero, S.; Llusar, R.; Polo, V.; Fourmigue, M. Cryst. Growth Des. 2008, 8, 2241. (i) Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew. Chem., Int. Ed. 2003, 42, 1210. (j) Espallargas, G. M.; Brammer, L.; Sherwood, P. Angew. Chem., Int. Ed. 2006, 45, 435. (4) (a) Barooah, N.; Sarma, R. J.; Baruah, J. B. CrystEngComm 2006, 8, 608. (b) Motohiro, N. CrystEngComm 2004, 6, 130. (5) Aaker€oy, C. B.; Beatty, A. M.; Helfrich, B. A. Angew. Chem., Int Ed. 2001, 40, 3240. (6) (a) Etter, M. C. J. Phys. Chem. 1991, 95, 4601. (b) Etter, M. C Acc. Chem. Res. 1990, 23, 120. (c) Steiner, T Angew. Chem.,Int. Ed. 2002, 41, 48. (d) Thalladi, V. R.; Goud, B. S.; Hoy, V. J.; Allen, F. H.; Howard, J. A. K.; Desiraju, G. R Chem. Commun. 1996, 401. (7) Halogen Bonding: Fundamentals and Applications. Metrangolo, P.; Resnati, G., Eds.; Structure and Bonding (Berlin); Springer: Berlin, 2008; Vol. 126. (8) For our previous co-crystal synthesis based on combination of hydrogen bond and halogen bond see:(a) Aaker€ oy, C. B.; Desper, J.; Helfrich, B. A.; Metrangolo, P.; Pilati, T.; Resnati, G.; Stevenazzi, A. Chem. Commun. 2007, 4236. (b) Aaker€oy, C. B.; Schultheiss, N.; Rajbanshi, A.; Desper, J.; Moore, C. Cryst. Growth Des. 2009, 9, 432. (c) Aaker€oy, C. B.; Chopade, P. D.; Ganser, C.; Desper, J. Chem. Commun. 2011, 47, 4688. (9) Saha, B. K.; Nangia, A.; Jask olski, M. CrystEngComm 2005, 7, 355. (10) CSD search carried out on ConQuest Version 1.12 (Updated till February 2010). CSD codes for the 2-aminopyrazine based compounds showing homosynthon are AMCPYZ, AMPYRZ, AMXPYZ. Allen, F. A. Acta Crystallogr. Sect. B: Struct. Sci. 2002, 58, 380. (11) (a) Walsh, R. B.; Padgett, C. W.; Metrangolo, P.; Resnati, G.; Hanks, T. W.; Pennington, W. T. Cryst. Growth Des. 2001, 1, 165. (b) Corradi, E.; Meille, S. V.; Messina, M. T.; Metrangolo, P.; Resnati, G. Angew. Chem., Int. Ed. 2000, 39, 1782. (12) A CSD analysis shows that there are currently about 37 co-crystals containing halogen bonds between DITFB and pyridine derivatives. (13) (a) Hunter, C. A. Angew. Chem., Int. Ed. 2004, 43, 5310. (b) Musumeci, D.; Hunter, C. A.; Prohens, R.; Scuderi, S.; McCabe, J. F. Chem. Sci. 2011, 2, 883. (14) Charge calculations for B1 B3 were performed using Spartan ’04 (Wave function, Inc., Irvine, CA). All three molecules were optimized using PM3, with the maxima and minima in the electrostatic potential surface (0.002 e au 1 isosurface) determined using a positive point charge in the vacuum as a probe. (15) (a) Ananthavel, S. P.; Manoharan, M. Chem. Phys. 2001, 269, 49. (b) Lu, Y.-X.; Zou, J.-W.; Wang, Y.-H.; Jiang, Y.-J.; Yu, Q.-S. J. Phys. Chem. A 2007, 111, 10781.
’ REFERENCES (1) Lehn, J.-M. Supramolecular Chemistry, Concepts and Perspectives; VCH: Weinheim, 1995. (2) (a) Aaker€oy, C. B.; Salmon, D. J. CrystEngComm 2005, 7, 439. (b) MacGillivray, L. CrystEngComm 2004, 6, 77. (c) Lehn, J.-M Science 2002, 295, 2400. (d) Desiraju, G. R. Acc. Chem. Res. 2002, 35, 565. 5336
dx.doi.org/10.1021/cg2009013 |Cryst. Growth Des. 2011, 11, 5333–5336
