Modulating Supramolecular Reactivity Using Covalent “Switches” on a

Oct 25, 2012 - selective, and within this group, the carboxylic acid---pyridine heterosynthon .... weighing equimolar amounts (1:1 pyrazole/acid) in a...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/crystal

Modulating Supramolecular Reactivity Using Covalent “Switches” on a Pyrazole Platform Published as part of the Crystal Growth & Design virtual special issue In Honor of Prof. G. R. Desiraju Christer B. Aakeröy,* Evan P. Hurley, and John Desper Department of Chemistry, Kansas State University, 213 CBC Building, Manhattan, Kansas, 66502, United States S Supporting Information *

ABSTRACT: Systematic co-crystallizations of halogen- methyl- and nitrosubstituted pyrazoles with a library of 20 aromatic carboxylic acids have been carried out using melt and solution-based experiments. The solids resulting from all reactions were screened using infrared spectroscopy in order to determine if a reaction (co-crystal or salt) had taken place. The halogenated pyrazoles, including their dimethyl analogues, gave a supramolecular yield of 55−70%. Replacing a halogen atom (R-X, X = Cl, Br, I) with a nitro (R-NO2) group drops the success rate significantly to 10% due to the reduced charge on the basic nitrogen atom of the pyrazole. Eleven crystal structures were obtained: six were co-crystals and five were salts (including one hydrate). In all six co-crystals, dimeric pyrazole···acid assemblies were constructed via a combination of O−H---N(pyz) and N−H---O hydrogen bonds corresponding to a 100% supramolecular yield. A variety of weaker halogen-bonds CN---I, I---I and X---O− connect dimers into infinite onedimensional chains. In contrast, the salts displayed a variety of stoichiometries and a much wider range of noncovalent interactions, although a charge-assisted N+-H---O− hydrogen bond was present in all five structures. In general, the salts lack structural and stoichiometric predictability and stability as compared to the co-crystals. Furthermore, the overall electrostatic charge on the key binding sites on the pyrazole backbone can be modulated by using specific covalent switches, which in turn can increase (or decrease) the success rate for a reaction.

A

interaction. The ditopic binding site on the pyridine molecule combines a very strong acceptor moiety N(py), and a very weak donor, C−H. In order to determine how important the balance between these two contacts is, as far as offering a competitive and successful synthon goes, we decided to systematically examine the ability of a variety of pyrazole molecules containing halogen atoms, methyl, and nitro substituents to form co-crystals with an approaching carboxylic acid. The desired synthon is again expected to involve a ditopic site on the base, but this time, the heterocyclic nitrogen is a rather poor hydrogen-bond acceptor, whereas the adjacent N− H donor is strong (relative to N and C−H, respectively, in pyridine), Scheme 2.

n important goal of crystal engineering is to design functional materials,1 which requires access to a library of synthons capable of facilitating the construction of heteromeric solid-state architectures with predictable composition, metrics, and topologies.2 In order to assemble a toolbox of reliable and robust synthons3 that can realize the desired supramolecular target, it is necessary to carry out systematic structural studies of molecules equipped with structurally ‘active’ functional groups and to determine their binding preferences in competitive environments. Hydrogen-bond based synthons are commonly used in crystal engineering, as they are strong (in the context of noncovalent interactions), directional, and potentially very selective, and within this group, the carboxylic acid---pyridine heterosynthon, Scheme 1, has been particularly prominent.4,5 The communication between molecules in this pair is dictated by the primary O−H---N(py) hydrogen bond complemented (sometimes) by a secondary C−H---O

Scheme 2. The Targeted Carboxylic Acid---Pyrazole Heterosynthon

Scheme 1. The Carboxylic Acid---Pyridine Heterosynthon

Received: September 23, 2012 Revised: October 15, 2012

© XXXX American Chemical Society

A

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Scheme 3. The Nine Pyrazoles Used in This Studya

Potentially, carboxylic acid---pyrazole interactions, which could parallel carboxylic acid---pyridine interactions (same graph set notation, R22(7)),6 can offer a reliable and predictable motif that may expand our current crystal engineering toolbox for targeted supramolecular synthesis. However, since the angle between the N−H bond and the lone pair on the adjacent nitrogen atom is somewhat large to provide a perfect fit with a carboxylic acid, the viability of this interaction has yet to be established. In fact, a few pyrazole/carboxylic acid crystal structures have been published previously,7,8 which indicate that head-to-head acid···pyrazole dimers are less likely than a trimer or catemer-type interaction. In addition to the fundamental importance of intermolecular bonds, carboxylic acid---azole interactions, such as those occurring with pyrazoles, play an important role in active ingredients in many agrichemicals9 and in membrane protein antagonists.10 Furthermore, halogenated pyrazoles are part of the backbone of some important agrichemicals11 and have also shown promising antitumor activity.12 Thus, the ability to form cocrystals of pyrazole-containing compounds would provide access to a larger number of solid forms which could optimize physicochemical properties of a particular active ingredient. One way of ‘dialing-in’ the strength of any heterosynthon, including azole---acid combinations, is to use covalent ‘switches’ that modify the electrostatic nature of the functional groups that participate in the intermolecular interaction under consideration. Obviously, well-established covalent synthesis allows us to decorate any molecular fragment with substituents in suitable positions. For example, methyl and methoxy groups can be added for the purpose of increasing the 'negative' charge on a specific hydrogen-bond acceptor site, thus making it more likely to accept an electrostatically driven hydrogen-bond interaction; the addition of electron withdrawing groups will have the opposite effect. In this study, we set out to establish if relatively minor changes to the pyrazole backbone have a noticeable effect on the ability of pyrazole derivatives to form co-crystals13 with a series of carboxylic acids. As weak and medium strength hydrogen bonds are primarily electrostatic in nature, we postulated that the supramolecular yield of these reactions will reflect changes in electrostatic potentials on the participating species. Molecular electrostatic potential surface (MEPS) calculations were thus performed (using both semiempirical14 and DFT methods), in order to estimate the relative charge differences on the hydrogen-bond donor/ acceptor sites in the participating pyrazole molecules. The experimental work utilized nine different pyrazoles, Scheme 3, and 20 carboxylic acids, Table 1. The compounds shown in Scheme 3 have been functionalized in order to make slight alterations to the electrostatic charge on the hydrogen-bond acceptor, N(pyz), and the main hydrogen-bond donor, N−H. In Scheme 4, our approach for mapping out the synthetic landscape15 for reactions between pyrazoles and carboxylic acids is summarized. To minimize any potential problems with solubility differences between the two reactants in each case, the substances were melted together, and the solid formed upon cooling was characterized using IR spectroscopy. Any reaction that resulted in reaction (co-crystal or salt formation), as determined via infrared spectroscopy, was subsequently used for crystal growth experiments from solution in order to obtain crystals suitable

a

Pyrazole (halo and nitro analogues) on top. Dimethyl pyrazole (and halo analogues) on bottom.

for single-crystal X-ray diffraction studies. In one case, suitable crystals were directly obtained from the melt.



EXPERIMENTAL SECTION

All chemicals were purchased from Aldrich and used without any further purification unless otherwise noted. Chloropyrazole and bromopyrazole were synthesized according to a report by Zhao and co-workers.16 Iodopyrazole was synthesized according to a procedure by Elguero.17 The 4-halo-(3,5-dimethyl) pyrazoles were synthesized according to Ehlert and co-workers.18 Melting points were determined on a Fisher-Johns melting point apparatus and are uncorrected. Single crystal X-ray data were collected on a Bruker Apex diffractometer. AM1/PM3/DFT charge calculations were performed using Spartan software. Infrared data were collected on a Thermo Scientific Nicolet 380 FT-IR using a ZnSe crystal. Melt experiments were performed by weighing equimolar amounts (1:1 pyrazole/acid) in a test tube and gently heating the two solids together using a Pamran Co. Inc. (Waukesha, WS) HEJET model HJ700 heat gun until the two components appeared melted together. NMR data were collected on a Bruker 200 MHz instrument, unless otherwise noted.



SYNTHESIS Co-crystals: Synthesis of 4-Bromo-1H-pyrazole:3,5-dinitrobenzoic Acid (1:1). To a vial, 4.3 mg (0.029 mmol) of 4bromo-1H-pyrazole was added along with 1 mL of methanol. To a separate vial, 6.1 mg (0.029 mmol) of 3,5-dinitrobenzoic acid was added along with 1 mL of methanol. The two solutions were combined in a vial, covered in parafilm (1 pinhole), and left for slow evaporation. Colorless prisms were obtained after several days. mp 135−140 °C. Synthesis of 4-Iodo-1H-pyrazole:3,5-dinitrobenzoic Acid (1:1). To a vial, 12.6 mg (0.065 mmol) of 4-iodo-1H-pyrazole was added along with 0.5 mL of methanol. To a separate vial, 13.8 mg (0.065 mmol) of 3,5-dinitrobenzoic acid was added along with 0.5 mL of methanol. The two solutions were combined in a vial, covered in parafilm (3 pinholes), and left for slow evaporation. Colorless plates were obtained after several days. mp 123−125 °C. Synthesis of 4-Iodo-1H-pyrazole:4-cyanobenzoic Acid (1:1). To a vial, 15.2 mg (0.078 mmol) of 4-iodo-1H-pyrazole was added along with 0.5 mL of methanol. To a separate vial, 11.5 mg (0.078 mmol) of 4-cyanobenzoic acid was added along with 0.5 mL of methanol. The vial containing the acid was heated until fully dissolved, then added to the solution containing the pyrazole. The vial was covered in parafilm (three pinholes) and left for slow evaporation. Colorless prisms were obtained after several days. mp 189−190 °C. Synthesis of 4-Chloro-3,5-dimethyl-1H-pyrazole:4-hydroxy-3-methoxybenzoic Acid (1:1). To a vial, 7.3 mg (0.056 B

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Table 1. Twenty Carboxylic Acids Used in This Studya

a

Acids listed in order of increasing acidity. pKa values obtained from Scifinder scholar.

The two solutions were combined in a vial, covered in parafilm (1 pinhole), and left for slow evaporation. Colorless prisms were obtained after several days. mp 156−160 °C. Synthesis of 4-Bromo-3,5-dimethyl-1H-pyrazole:3,5-dinitrobenzoic Acid (1:1). To a vial, 5.7 mg (0.033 mmol) of 4bromo-3,5-dimethyl-1H-pyrazole was added along with 1 mL of methanol. To a separate vial, 7.1 mg (0.033 mmol) of 3,5dinitrobenzoic acid was added along with 1 mL of methanol. The two solutions were combined in a vial, covered in parafilm (1 pinhole), and left for slow evaporation. Colorless plates were obtained after several days. mp 155−158 °C. Salts. Synthesis of 4-Chloro-3,5-dimethyl-1H-pyrazolium:2,4-dinitrobenzoate:4-chloro-3,5-dimethyl-1H-pyrazole (1:1:1). To a vial, 5.6 mg (0.043 mmol) of 4-chloro-3,5dimethyl-1H-pyrazole was added along with 200 μL of methanol. To a separate vial, 9.2 mg (0.043 mmol) of 2,4dinitrobenzoic acid was added along with 200 μL of methanol. The two solutions were combined in a vial, covered in parafilm (1 pinhole), and left for slow evaporation. Colorless plates were obtained after several days. mp 100−102 °C. Synthesis of 4-Iodo-3,5-dimethyl-1H-pyrazolium:3,5-dinitrobenzoate (1:1). To a vial, 5.4 mg (0.024 mmol) of 4-iodo3,5-dimethyl-1H-pyrazole was added along with 200 μL of methanol. To a separate vial, 5.2 mg (0.024 mmol) of 3,5dinitrobenzoic acid was added along with 200 μL of methanol.

Scheme 4. Flow Chart for Supramolecular Synthesis of CoCrystals Employed in This Study

mmol) of 4-chloro-3,5-dimethyl-1H-pyrazole was added along with 200 μL of methanol. To a separate vial, 9.6 mg (0.057 mmol) of 4-hydroxy-3-methoxybenzoic acid was added along with 200 μL of methanol. The two solutions were combined in a vial, covered in parafilm (1 pinhole), and left for slow evaporation. Colorless spheres were obtained after several days. mp 205−209 °C. Synthesis of 4-Chloro-3,5-dimethyl-1H-pyrazole:3,5-dinitrobenzoic Acid (1:1). To a vial, 5.3 mg (0.041 mmol) of 4chloro-3,5-dimethyl-1H-pyrazole was added along with 200 μL of methanol. To a separate vial, 8.9 mg (0.042 mmol) of 3,5dinitrobenzoic acid was added along with 200 μL of methanol. C

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

The two solutions were combined in a vial, covered in parafilm (1 pinhole), and left for slow evaporation. Colorless prisms were obtained after several days. mp 138−142 °C. Synthesis of 1H-Pyrazolium:2,6-dichlorobenzoate:2,6-dichlorobenzoic Acid (1:1:1). To a test tube, 13.1 mg (0.19 mmol) of 1H-pyrazole was added along with 39.3 mg (0.21 mmol) of 2,6-dichlorobenzoic acid. The solids were heated using a heat gun until both components had melted. The melt was allowed to cool to room temperature, yielding colorless plates. mp 71−73 °C. Synthesis of 1H-Pyrazolium:3,5-dinitrobenzoate:3,5-dinitrobenzoic Acid (1:1:1). To a small beaker, 29.1 mg (0.43 mmol) of 1H-pyrazole was added and dissolved in 1 mL of absolute ethanol (200 proof). To a separate beaker, 91.4 mg (0.43 mmol) of 3,5-dinitrobenzoic acid was added and dissolved in 1 mL of absolute ethanol (200 proof). The two solutions were combined in a beaker, covered with parafilm (5 pinholes), and left for slow evaporation. Colorless plates were obtained after several days. mp 146−148 °C. Synthesis of 3,5-Dimethyl-1H-pyrazolium:3,5-dinitrobenzoate Hydrate (1:1). To a small beaker, 27.9 mg (0.29 mmol) of 3,5-dimethyl-1H-pyrazole was added and dissolved in 1 mL of absolute ethanol (200 proof). To a separate beaker, 62.3 mg (0.29 mmol) of 3,5-dinitrobenzoic acid was added and dissolved in 1 mL of absolute ethanol (200 proof). The two solutions were combined in a test tube and left for slow evaporation. Colorless prisms were obtained after several days. mp 120−124 °C. Covalent Synthesis. Six pyrazole molecules were synthesized using published procedures and characterized by 1H NMR spectroscopy and melting-point, Table 2.

Table 3. AM1, PM3, and DFT Calculations on All Nine Pyrazolesa

1.Cl 1.Br 1.I 2.Cl 2.Br 2.I

1

H NMR values [lit. value]

7.58 (s, 2H, CH) [7.55, s, 2H, CH]16 7.61 (s, 2H, CH) [7.53, s, 2H, CH]16 7.65 (s, 2H, CH) [7.64, s, 2H, CH]17 2.25 (s, 6H, CH3) [2.25, s, 6H, CH3]18 2.26 (s, 6H, CH3) [2.26, s, 6H, CH3]18 2.27 (s, 6H, CH3) [2.27, s, 6H, CH3]18

melting point [lit. value]

49

89−91 °C [91−92 °C]19

30

104−105 °C [108 °C]14

47

106−108 °C [117−118 °C]13 116−119 °C [122 °C]16

62

135−136 °C [not reported]

52

AM1 (N)

PM3 (H)

PM3 (N)

DFT (H)

DFT (N)

1.H 1.Cl 1.Br 1.I 1.NO2 2.H 2.Cl 2.Br 2.I

178 194 200 200 254 171 188 193 191

−251 −236 −230 −231 −260 −256 −238 −239 −237

156 167 169 161 280 141 150 152 150

−307 −289 −283 −282 −297 −308 −298 −293 −302

245 269 273

−186 −159 −160

315 226 260 253

−162 −195 −172 −171

Values are given in kJ/mol. DFT calculations performed at 6-31+G* level of theory. DFT calculations could not be performed on iodopyrazoles.

resulted in a reaction (salt or co-crystal). A total of 11 crystal structures were subsequently obtained (ESI). Crystal Structures (Co-Crystals). The crystal structure determination of 1.Br:N shows that the outcome is a co-crystal with the intended 1:1 stoichiometry with four unique molecules in the asymmetric unit cell. The acidic protons remain on the two unique carboxylic acids as shown by the significantly different CO/C−O(H) bond lengths, 1.219(4)/1.325(4) Å, and 1.219(4)/1.312(4) Å, respectively. The driving force for the reaction is the heteromeric O−H---N(pyz)/N−H---O synthon, which we define as “head-to-head dimers”, Figure 1. These dimers are organized into one-dimensional (1-D) chains via C−Br--O(nitro) halogen bonds; r(C−Br---O) is 3.148 Å and the C− Br---O angle is 154°. The crystal structure of 4-iodopyrazole:4-cyanobenzoic acid, 1.I:H, displays a very similar primary supermolecule, a head-tohead dimer constructed from a neutral O−H---N(pyz)/N−H--O synthon, Figure 2. The dimers assemble into infinite 1-D chains by CN---I halogen bonds, where the CN---I distance is 3.183 Å with a C−I---N angle of 179.38°. The crystal structure of 1.I:N also contains 1-D chains where dimer formation occurs via O−H---N(pyz) and N−H---O interactions. The dimers are connected into 1-D chains through C−H---O synthons. The chains are linked together in a stairstep fashion through I---I (type I) halogen bonds, Figure 3. The I---I distance is 3.698 Å and the C−I---I angle is 136.54°. The crystal structure of 2.Cl:B contains head-to-head heterodimers formed via O−H---N(pyz) and N−H---O synthons, Figure 4. The N−H proton is bifurcated between the CO oxygen from one acid and the O−H oxygen from a second symmetry-related neutral acid. The synthons combine to form an extended zigzag chain. The structure of 2.Cl:N also contains neutral heterodimers formed by O−H---N(pyz) and N−H---O synthons. The dimers extend into 1-D chain via C−Cl---O synthons. The C−Cl---O distance is 3.065 Å and the C−Cl---O angle is 169.98°, Figure 5. Similarly, in the crystal structure of 2.Br:N, an infinite chain is observed where O−H---N(pyz) and N−H---O synthons drive cocrystal formation. The C−Br---O− distance is 3.022 Å and the C−Br---O− angle is 174.25°, Figure 6. Salts. The first example of a salt was revealed in the structure determination of the 2.Cl:T. Furthermore, not only has the proton been transferred from the acid to the base, but

% yield

73−74 °C [75−76 °C]13

AM1 (H)

a

Table 2. 1H NMR Data, Yields, and Melting Points for Six Pyrazoles molecule

molecule

38

Charge Calculations. AM1, PM3, and DFT calculations were performed to establish the maxima and minima in the MEPS at the relevant hydrogen-bond donor/acceptor sites, Table 3. Co-Crystal Screen. A co-crystal screen using IR spectroscopy was carried out for a total of 180 (9 × 20) pyrazolecarboxylic acid combinations, Table 4. A reaction, co-crystal or salt, was deemed to take place when the resulting solid displayed broad stretches in the 2500 cm−1 and 1900 cm−1 region, indicative of O−H---N(pyz) intermolecular hydrogenbond interactions that are only possible if the solid contains both components. Single-Crystal Data. On the basis of IR spectroscopy we were able to determine that 101 of the 180 experiments D

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Table 4. Summary of Prominent IR Stretches (cm−1) from the Melt Experiments between Nine Pyrazoles and 20 Carboxylic Acidsa acids

1.H

1.Cl 2499, 1897, 1666

A

1.Br

1.I

1.NO2

2507, 1874, 1674

2.H 2488, 1855, 1679 2496, 1879, 1667

B C D,E, F G H I J

2382, 1883, 1695 2397, 1928, 1690

K L M N O P Q R S T supramolecular yield (%) a

2533, 1912, 1679 2517, 1936, 1679 2436, 1866, 1702 2492, 1932, 1704 2448, 1874, 1702 2357, 1912, 1720 2440, 1885, 1709 2296, 1932, 1647 2429, 1866, 1713 11/20 = 55%

2440, 1862, 1682 2464, 1862, 1729 2436, 1885, 1702

2535, 1865, 1693 2396, 1865, 1693 2425, 1862, 1698

2535, 1872, 1686 2469, 1885, 1699 2401, 1835, 1679 2499, 1867, 1675 2417, 1846, 1687

2487, 1874, 1690 2433, 1881, 1694 2515, 1878, 1682 2413, 1854, 1698

2429, 1854, 1690 2495, 1878, 1683 2495, 1881, 1682 2405, 1823, 1691

2526, 1874, 1674 2464, 1893, 1706 2483, 1893, 1709 2421, 1909, 1702 2425, 1878, 1706 13/20 = 65%

2522, 1885, 1678 2491, 1893, 1702 2499, 1897, 1713

2464, 1885, 1704 2415, 1897, 1704

2366, 1881, 1708 12/20 = 60%

2374, 1881, 1709 12/20 = 60%

2491, 1893, 1721

2413, 1827, 1691

2546, 1862, 1698

2/20 = 10%

2.Cl

2491, 1878, 1667 2452, 1838, 1671

2.Br

2.I

2507, 1897, 1691 2491, 1881, 1671

2448, 1870, 1675

2357, 1850, 1695 2429, 1889, 1695 2456, 1862, 1716 2429, 1866, 1683 2417, 1893, 1716 2448, 1885, 1687 2433, 1846, 1698

2503, 1897, 1699 2484, 1891, 1687 2350, 1858, 1704

2553, 1893, 1687 2632, 1889, 1720 2429, 1850, 1708

2476, 1891, 1695 2476, 1904, 1695 2492, 1851, 1683 2521, 1887, 1720 2459, 1842, 1708 2488, 1887, 1712 2491, 1885, 1708 2435, 1822, 1699 2522, 1893, 1717 14/20 = 70%

2397, 1881, 1712 2499,1917, 1679 2519, 1842, 1701

2382, 1870, 1679 2541, 1889, 1675 2448, 1878, 1688 2468, 1897, 1708 2433, 1893, 1683 2390, 1901, 1720 2511, 1921, 1679 2413, 1889, 1704

2483, 1909, 1712 2476, 1901, 1716

2554, 1897, 1699 2358, 1893, 1675 2483, 1881, 1704

2386, 1901, 1691 2452, 1909, 1712 2468, 1889, 1708

2464, 1889, 1704 11/20 = 55%

2456, 1878, 1699 14/20 = 70%

2358, 1885, 1704 12/20 = 60%

Note: numbers represent wavenumbers (cm−1) observed.

Figure 1. Infinite 1-D chain in the crystal structure of 1.Br:N driven by O−H---N(pyz) and N−H---O synthons. N−O---Br interactions extend the dimers into a 1D chain. Figure 3. An overview of the 2-D layer in the crystal structure of 1.I:N.

Figure 2. Infinite 1-D chain in the crystal structure of 1.I:H driven by O−H---N(pyz) and N−H---O synthons. CN---I interactions extend the dimers into 1D chains.

Figure 4. Infinite 1-D zigzag chain in the structure of 2.Cl:B..

an additional neutral pyrazole molecule is also included in the lattice, Figure 7. The three building blocks are assembled into discrete trimers, driven by N−H---O, N−H---N, and chargeassisted N−H+---O− synthons.

The reaction between 2.I and N also results in an ionic compound. A combination of charge-assisted N−H+---O−, N− E

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 5. 1-D chain of in the structure of 2.Cl:N.

Figure 6. 1-D chain in the crystal structure of 2.Br:N. Figure 9. Discrete tetramers formed between in the crystal structure of 1.H:R.

Figure 10. A zigzag chain in the crystal structure of pyrazolium 3,5dinitrobenzoate 3,5,-dinitrobenzoic acid.

Figure 7. A hydrogen-bonded trimer in the crystal structure of 4chloro-3,5-dimethylpyrazolium 2,4-dintrobenzoate 4-Cl-3,5-dimethylpyrazole.

H---O, and N−H---O− synthons produce tetramers, Figure 8, which are connected into a ladder-like motif via C−I---O− interactions, with a C−I---O− distance of 3.089 Å and a C−I--O− angle of 177.59°.

forms an O−H---O synthon to complete the structure while a C−H---O− interaction generates the 1-D chains. The proton transfer from N to 2.H produces the only solvated crystal structure in this series. The main feature is a trimer involving all three components, Figure 11. The N−H proton interacts directly with the oxygen from water to form a N−H---O synthon. The O−H(water)---O(carbonyl) synthon completes the trimer. Table 5 displays the relevant hydrogen-bond distances for the crystal structures obtained.

Figure 8. Tetramers (subsequently linked into ribbon) formed in the solid product resulting from a combination of 2.I and N.

The reaction between 1.H and R also produces an ionic compound where the main feature of the crystal structure is a tetramer, Figure 9, held together by charge-assisted N−H+---O− synthons. Figure 10 depicts a zigzag chain formed by charge-assisted N−H+---O− synthons in the crystal structure of the ionic compound produced when 1.H was allowed to react with N. The N−H proton is bifurcated between a CO oxygen from a second neutral acid and the carboxylate oxygen directly involved in the hetero dimer. The second neutral acid also

Figure 11. A hydrated ion-pair in the crystal structure of 3,5dimethylpyrazolium 3,5,-dinitrobenzoate hydrate. F

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Table 5. Summary of Hydrogen-Bond Geometries structure

D−H···A

d(D−H)

d(H···A)

1.Br:N

N111−H111···O221 N112−H112···O222 O211−H211···N121 O212−H212···N122 O(11)−H(11)···N(22) N(21)−H(21)···O(12) O(11)−H(11)···N(22) N(21)−H(21)···O(12) O(31)−H(31)···N(42) N(41)−H(41)···O(32) N(11)−H(11)···O(22) O(21)−H(21)···N(12) O(24)−H(24)···O(22)#1 N(11)−H(11)···O(22) O(21)−H(21)···N(12) N(11)−H(11)···O(22) O(21)−H(21)···N(12) N(11)−H(11)···O(38) N(12)−H(12)···N(22) N(21)−H(21)···O(37) N11A H11A O21 N11B H11B O21 N12A H12A O23 N12B H12B O23 N11 H11 O21 N12 H12 O21 O31 H31 O22 N11A2−H11A2···O312 N12A2−H12A2···O322 N11B2−H11B2···O312 N12B2−H12B2···O222 O211−H211···O311 O212−H212···O312 N11 H11 O21 N12 H12 O1W O1W H1A O22 O1W H1B O22

0.87(5) 0.75(5) 0.88(5) 0.89(5) 0.96(2) 0.79(2) 0.87(2) 0.78(2) 0.83(2) 0.76(2) 0.89(3) 0.90(3) 0.84(4) 0.758(18) 0.814(18) 0.98(5) 1.28(5) 1.15(2) 0.92(2) 0.78(3) 0.88 0.88 0.880.88

2.11(5) 2.18(5) 1.80(5) 1.79(5) 1.70(2) 2.16(2) 1.75(2) 2.31(2) 1.84(2) 2.28(2) 2.18(3) 1.76(3) 2.01(4) 2.619(17) 1.810(19) 2.49(5) 1.30(5) 1.41(2) 1.91(3) 1.98(3) 1.85 2.20 2.191.91

2.834(4) 2.828(4) 2.674(4) 2.657(4) 2.6521(17) 2.8225(17) 2.6225(17) 2.9250(17) 2.6671(18) 2.9050(18) 2.875(2) 2.654(2) 2.636(2) 3.1647(13) 2.5946(13) 3.180(5) 2.578(4) 2.541(2) 2.811(2) 2.740(2)

0.87(4) 0.92(4) 0.83(5) 0.88 0.88 0.89(4) 0.83(4) 0.86(4) 0.79(4) 0.995(17) 0.967(16) 0.890(19) 0.888(18)

1.94(4) 1.78(4) 1.75(5) 2.30 2.22 1.73(4) 1.99(4) 1.67(4) 1.72(4) 1.595(17) 1.713(16) 1.851(19) 1.880(18)

2.797(4) 2.650(4) 2.575(4) 2.843(17) 3.059(17) 2.620(3) 2.741(3) 2.522(3) 2.517(3) 2.5747(14) 2.6490(14) 2.7390(13) 2.7681(13)

1.I:H 1.I:N

2.Cl:B

2.Cl:N 2.Br:N 2.Cl:T

2.I:N

1.H:R

1.H:N

2.H:N



DISCUSSION

d(D···A)

2.713(6) 3.005(9) 3.024(6) 2.772(7)

∠(DHA) 140(4) 145(5) 175(5) 167(5) 171(2) 141(2) 177(2) 136(2) 176(2) 140(2) 136(3) 174(3) 131(3) 130.6(16) 161.6(18) 128(3) 173(4) 167(2) 166(2) 167(3) 168.4 151.1 158.6 164.3 166(4) 158(4) 171(5) 119.7 160.6 178(4) 149(3) 172(3) 177(4) 167.2(15) 162.0(15) 174.9(17) 178.9(18)

solid is obtained (one anion, one cation, one neutral acid) even though CO band is again shifted by 12 cm−1. Our results from the co-crystal screening indicate that adding a halogen atom to the pyrazole backbone does not change the overall efficiency of the supramolecular yield as compared to the unsubstituted pyrazole and dimethylpyrazole. For the latter two the success rate is 55−70%, compared with a 60%−70% success rate for the halogenated pyrazoles. The similarities in supramolecular yield can be ascribed to the fact that neither methyl groups nor halogen atoms significantly alter the electrostatic potential on the primary binding sites of the pyrazole moiety. However, our calculations did indicate that nitropyrazole is considerably less basic (and the N−H hydrogen atom is substantially more acidic) than the corresponding moieties in the other pyrazoles. We therefore assumed that any hydrogen-bond interaction between nitropyrazole and a carboxylic acid would be relatively weak and less likely to drive a successful co-crystallization. In fact, the supramolecular yield for nitropyrazoles dropped to 10%, which is consistent with the hypothesis that the magnitude of the electrostatic component of the hydrogen-bond is critical to supramolecular effectiveness of such interactions.

Theoretical Calculations. According to the calculations, the presence of two methyl groups or a halogen atom on the pyrazole backbone has a relatively minor effect on the electrostatic charges. The addition of the nitro group, however, has a pronounced effect on the charges, notably on the acidic N−H proton on the pyrazole molecule. Although the actual numbers differ between different methodologies (AM1, PM3, and DFT), the relative ranking of binding sites remain unchanged. The question is, are these differences significant enough to change the supramolecular yield? Co-Crystal Screen. Although it is easy to distinguish between reaction and no-reaction using IR spectroscopy, an unambiguous determination of whether a salt or co-crystal has formed is not always possible, especially since some reactions produce solids that contain both ions and neutral molecules. For example, the single-crystal data for 2.Cl:N show that the composition in this solid is a 1:1 co-crystal and in the corresponding IR spectrum the CO band of the acid has shifted by 12 cm−1 compared with that of the free acid. However, when the same pyrazole is combined with T, a 1:1:1 G

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Predicting salt or co-crystal formation from analysis of the CO stretches (normalized against the neutral acid) turned out to be difficult due the small changes in peak position (4−12 cm−1). However, the more acidic acids (based upon pKa values) did show a higher propensity to form salts as found from singlecrystal diffraction data. Co-crystals were formed by acids with pKa values in the 2.77−4.45 range, whereas salts were produced by acids with pKa values in the 1.43−2.77 range. Single Crystal Data. Analysis of the Cambridge Structural Database (CSD) reveals the majority of structures containing pyrazole are used as ligands in metal coordination complexes.20 In Table 6, the different motifs involving pyrazole molecules are listed. A detailed analysis has been provided by Elguero and co-workers.21

Table 7. Bond Angles for Relevant Structures Containing Halogen Bonding

dimers

trimers

tetramers

2 (CUMQAD, CUMQEH, CUQGOK, DASXEA, DASXEA10, EBILUX, GOQXIT, HOQHUQ, WIKZUL)

3 (CUQHAX)

a

C−X---O− (X = Cl, Br) (deg)

1.Br:N 1.I:N 2.Cl:N 2.Br:N

154 164 170 174

occasions a neutral acid was included, on one occasion a neutral pyrazole molecule was included, and finally one structure was hydrated. This behavior is fully consistent with a previous report13 which suggests that co-crystals are more likely to generate desirable stoichiometries whereas organic salts are far more likely to appear with unexpected stoichiometries. As a result of the presence of neutral acids or bases, or a solvent of crystallization, it is not possible to make many general conclusions about how the competing intermolecular interactions in this group of five compounds should be ranked in terms of strength or proclivity.

Table 6. Summary of Crystal Structure Type Obtained for Pyrazoles from Data in the CSDa 8 (CUMQOR, FITQEE, FITQEE01, OMEKAS, TIDLIB, TIDLIB01)

compound



CONCLUSIONS This study has established several points of general importance to practical crystal engineering. First, the head-to-head pyrazole---carboxylic acid synthon has shown itself to be a valid and robust heterosynthon that can find considerable use in the construction of co-crystals and extended architectures in organic solids. Second, the connection between electrostatic charge on a hydrogen-bond donor/acceptor site and the supramolecular yield of subsequent co-crystallizations is emphasized yet again. However, we have also refined this conclusion by examining which substituents are capable of effecting a noticeable change; the -NO2 moiety has a far greater effect than -Me, -Cl, -Br, or -I. Third, practical crystal engineering is likely to be more successful (in terms of predicting stoichiometry and connectivity) if the solid form is a neutral co-crystal as opposed to an organic salt. The structural landscape15 of the latter is far more difficult to map out primarily because chemical composition of the resulting solid is highly unpredictable.

Ref codes shown in parentheses.

Most of the structures in Table 6 arise from 1H-pyrazole or the 3,5-dimethyl-1H-pyrazole analogue: the majority of our structures contain halogen-atom substituted pyrazoles. Co-crystals: Six of the 11 crystal structures that were obtained in this study were cocrystals, i.e., all components were charge neutral. In each case, the desired head-to-head acid--dimer was driving the co-crystal synthesis through a two point O−H---N/CO---H−N synthon, [R22(7)]. The dimers were subsequently linked into chains through a variety of secondary contacts depending on the nature of the substituents on the pyrazole and/or acid backbone. We identified I---O2N, Br--O2N, and Cl---O2N as well as I---N≡C and I---I (Type I) interactions, and finally an O−H---OOHC hydrogen bond all of which served to create chains of dimers. In the case of 1.I:H the distance between nitrogen and iodine is 3.183 Å, which is considerably shorter than the sum of their van der Waals radii (3.53 Å)22 and in good agreement with similar CN---I cyanohalogen interactions observed by Desiraju and Harlow.23 This also indicates a favorable n → σ* donation from the nitrogen lone-pair to the positive lobe on the iodine atom.24 It is noteworthy that placing halogen-bond donors in the 4- position on the pyrazole may allow for carboxylic acids, containing suitable halogen-bond acceptors in the 4- position, to form predictable supramolecular chains driven by the interplay between hydrogen and halogen bonds.25 As expected, the only halogen atom capable of forming Type I interactions in these structures is the iodine atom due to its greater polarizability. However, all three halogen atoms are capable of forming weakly interacting halogen bonds with an O2N-acceptor site, Table 7. Salts. In five of the 11 crystal structures that were obtained, proton transfer was found to have taken place. In this series of compounds, a charge assisted N−H+---−OOC hydrogen bond was present in all five structures, again demonstrating the importance of charge for structure directing hydrogen bonds. A striking general difference between co-crystals and salts is the fact that in the former group, all six compounds displayed the expected 1:1 stoichiometry, whereas four of the five salts offered an unexpected chemical stoichiometry. On two



ASSOCIATED CONTENT

S Supporting Information *

Tabulated crystallographic data. This information is available free of charge via the Internet at http://pubs.acs.org/.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We thank NSF (CHE-0957607) and NSF GK-12 for financial support. REFERENCES

(1) Aakeröy, C. B.; Champness, N. R.; Janiak, C. CrystEngComm 2010, 12, 22−43. (2) (a) Aakerö y, C. B.; Hussain, I.; Forbes, S.; Desper, J. CrystEngComm 2007, 9, 46−54. (b) Shattock, T. R.; Arora, K. K.; Vishweshwar, P.; Zaworotko, M. J. Cryst. Growth Des. 2008, 8, 4533− H

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

4545. (c) Aakeröy, C. B.; Beatty, A. M.; Helfrich, B. A. J. Am. Chem. Soc. 2002, 124, 14425−14432. (3) Thalladi, V. R.; Goud, B. S.; Hoy, V. J.; Allen, F. H.; Howard, J. A. K.; Desiraju, G. R. Chem. Commun. 1996, 3, 401−402. (4) Mohamed, S.; Tocher, D. A.; Vickers, M.; Karamertzanis, P. G.; Price, S. L. Cryst. Growth Des. 2009, 9, 2881−2889. (5) (a) Bhogala, B. R.; Basavoju, S.; Nangia, A. CrystEngComm 2005, 7, 551−562. (b) Aakeröy, C. B.; Schultheiss, N. C.; Rajbanshi, A.; Desper, J.; Moore, C. Cryst. Growth Des. 2009, 9, 432−441. (6) (a) Etter, M. C.; MacDonald, J.; Bernstein, J. Acta Crystallogr., Sect. B 1990, B46, 256−262. (b) Etter, M. C. Acc. Chem. Res. 1990, 23, 120−126. (7) (a) Foces-Foces, C.; Infantes, L.; Aguilar-Parrilla, F.; Golubev, N. S.; Limbach, H. H.; Elguero, J. J. Chem. Soc., Perkin Trans. 2 1996, 3, 349−353. (b) Lopez, C.; Claramunt, R. M.; Garcia, M. A.; Pinilla, E.; Torres, M. R.; Alkorta, I.; Elguero, J. Cryst. Growth Des. 2007, 7, 1176−1184. (8) Basu, T.; Sparkes, H. A.; Mondal, R. Cryst. Growth Des. 2009, 9, 5164−5175. (9) Vors, J. P.; Gerbaud, V.; Gabas, N.; Canselier, J. P.; Jagerovic, N.; Jimeno, M. L.; Elguero, J. Tetrahedron 2003, 59, 555−560. (10) Sasmal, P. K.; Reddy, D. S.; Talwar, R.; Venkatesham, B.; Balasubrahmanyam, D.; Kannan, M.; Srinivas, P.; Kumar, K. S.; Devi, B. N.; Jadhav, V. P.; Khan, S. K.; Mohan, P.; Chaudhury, H.; Bhuniya, D.; Iqbal, J.; Chakrabarti, R. Bioorg. Med. Chem. Lett. 2011, 21, 562− 568. (11) Lamberth, C. Heterocycles 2007, 71, 1467−1502. (12) Chou, L.-C.; Huang, L.-J.; Hsu, M.-H.; Fang, M.-C.; Yang, J.-S.; Zhuang, S.-H.; Lin, H.-Y.; Lee, F.-Y.; Teng, C.-M.; Kuo, S. C. Eur. J. Med. Chem. 2010, 45, 1395−1402. (13) Aakeröy, C. B.; Fasulo, M. E.; Desper. J. Mol. Pharm. 2007, 4, 317−322. (14) Musumeci, D.; Hunter, C. A.; Prohens, R.; Scuderi, S.; McCabe, J. F. Chem. Sci. 2011, 2, 883−890. (15) Mukherjee, A.; Grobelny, P.; Thakur, T. S.; Desiraju, G. R. Cryst. Growth Des. 2011, 11, 2637−2653. (16) Zhao, Z.-G.; Wang, Z.-X. Syn. Commun. 2007, 37, 137−147. (17) Vasilevsky, S. F.; Klyatskaya, S. V.; Tretyakov, E. V.; Elguero, J. Heterocycles 2003, 60, 879−886. (18) Ehlert, M. K.; Storr, A.; Thompson, R. C. Can. J. Chem. 1992, 70, 1121−1128. (19) Foces-Foces, C.; Saiz-Llamas, A. L.; Elguero, J. Z. Kristallogr. 1999, 214, 237−241. (20) Perez, J.; Riera, L. Eur. J. Inorg. Chem. 2009, 33, 4913−4925. (21) Alkorta, I.; Elguero, J.; Foces-Foces, C.; Infantes, L. ARKIVOC 2006, 2, 15−30. (22) Bondi, A. J. Phys. Chem. 1964, 68, 441−451. (23) Desiraju, G. R.; Harlow, R. L. J. Am. Chem. Soc. 1989, 111, 6757−6764. (24) Metrangelo, P.; Resnati, G. Chem.Eur. J. 2001, 7, 2511−2519. (25) (a) Aakeröy, C. B.; Chopade, P. D.; Ganser, C.; Desper, J. Chem. Commun. 2011, 47, 4688−4690. (b) Aakeröy, C. B.; Desper, J.; Helfrich, B. A.; Metrangolo, P.; Pilati, T.; Resnati, G.; Stevenazzi, A. Chem. Commun. 2007, 41, 4236−4238.

I

dx.doi.org/10.1021/cg301391s | Cryst. Growth Des. XXXX, XXX, XXX−XXX