Making crystals from crystals: three solvent-free routes to the hydrogen bonded co-crystal between 1,1′-di-pyridyl-ferrocene and anthranilic acid

Abstract

The organometallic complex [Fe(η⁵-C₅H₄–C₅H₄N)₂] has been reacted with the anthranilic acid (C₆H₄)NH₂COOH to generate the hydrogen bonded supramolecular macrocycle {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂. The product has been fully characterized by means of X-ray powder diffraction, ¹³C and ¹⁵N solid-state NMR and single crystal X-ray diffraction. It has been shown that the same product can be obtained, quantitatively, by three different processes, namely kneading, i.e. solid-state mixing in the presence of a catalytic amount of MeOH, wet compression, i.e. pressure without mixing in the presence of MeOH, and vapour digestion, i.e. a mixture of the solid reactants is left in an atmosphere of MeOH vapours. The product can also be obtained via thermally induced reaction of a mixture of the two solid reactants, while no reaction is observed by dry mixing and dry compression. These experiments provide new insights into the factors controlling the process suggesting that the reaction requires the intermediacy of a catalytic amount of solvent or melting of one reactant to take place.

---

Introduction

Solvent-free reactions, such as the mechanical mixing of solid reactants, often provide fast and quantitative routes to the preparation of novel organic and inorganic compounds.^1–3 In recent years, under the impetus of crystal engineering,⁴ solvent-free processes have begun to be investigated for the preparation of crystalline materials.^5,6 The number of papers reporting preparation of molecular co-crystals,⁷ coordination networks,⁸ salts,⁹ as well as the investigation of polymorphs¹⁰ is increasing rapidly. Beside inter-crystal reactions, those occurring in the solid state, such as cycloaddition s,¹¹ and many other organic reactions¹² have been the subject of successful investigations. Undoubtedly, the number of reactions and processes that can be carried out in the solid state is enormous and the statement that “all (organic) reactions can be conducted in solvent free conditions” is largely justified.¹³

We have reported a large number of cases where preparation of hydrogen-bonded co-crystals has been achieved by simple grinding of the solid reactants.¹⁴ However, the nature of the processes leading from reagents to products is not fully understood because the experimental conditions are not completely under control. For instance, it is difficult to estimate the extent of thermal effects associated with sheering and pressuring during manual or mechanical co-grinding. Furthermore, ambient humidity or intentional kneading, viz. the use of a micro quantity of solvent, can play a crucial role for the occurrence of the reaction.

In order to gain insight into these aspects we have chosen as a benchmark case the preparation of the 1 ∶ 1 hydrogen bonded co-crystal {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂. The compound has been obtained from the solid-state reaction of crystalline [Fe(η⁵-C₅H₄–C₅H₄N)₂] and anthranilic acid. It is worth recalling that the starting reactant [Fe(η⁵-C₅H₄–C₅H₄N)₂] can also be prepared in solvent-free conditions.¹⁵

Anthranilic acid was chosen also because it is known to exist in three polymorphic modifications,¹⁶ with the so called form I and form II thermally interconverting at ca. 90 °C. Importantly, the mechanically induced conversion of form I into II by grinding had been investigated by Etter,¹⁷ while Shan and co-workers have found several new kneading ways for the interconversion among the three known forms.^16d

We have made use of a multi-technique approach for the investigation of the solid state reactions products, which have been characterized via a complementary use of both single-crystal and powder X-ray diffraction, solid-state NMR spectroscopy and DSC analysis.

Results and discussion

The initial solid-state preparation was carried out by manual grinding in an agate mortar of equimolar quantities of the two solid materials. In order for the reaction to take place, kneading was required, i.e. a drop of MeOH was added during workup. For anthranilic acid the most stable form I was used. After grinding, the polycrystalline material was used as such for powder diffraction experiments. In a separate experiment, equimolar quantities of acid and base were dissolved in methanol and the solvent was allowed to evaporate at a temperature of 5 °C. By comparison with the diffractograms measured on the raw reactants, it was possible to ascertain that in both cases the starting material had been fully converted into the product. Methanol (99.8%) was used to grow single crystals of the co-crystal for the X-ray diffraction experiment. The structure determined from the single-crystal experiment (see below) was in turn used to calculate the reference powder diffraction pattern. In the crystal structure the hydrogen bond interactions (Fig. 1) show that anthranilic acid molecules act as bridges between the organometallic molecules, which are in a cisoid (eclipsed) conformation.

Fig. 1 The anthranilic acid molecules bridge together two organometallic sandwich molecules [H_CH atoms not shown for clarity].

The complex can be described as a supramolecular macrocycle whereby two (C₆H₄)NH₂COOH molecules and two [Fe(η⁵-C₅H₄–C₅H₄N)₂] complexes are joined by alternate O–H⋯N and N–H⋯N hydrogen bonding interactions. This is the most common arrangement for co-crystals involving two [Fe(η⁵-C₅H₄–C₅H₄N)₂] molecules and polyprotic acid units, although the alternative geometry based on an infinite acid–base hydrogen bonded network has been observed in at least one case.¹⁸ On the basis of the single crystal X-ray structure, it appears that the hydrogen atoms are ordered along the hydrogen bonds with the formation of three hydrogen bonding interactions, e.g. an intermolecular O–H⋯N [N(1)⋯O(2) 2.609(7) Å], an intramolecular O–H⋯N [N(3)⋯O(1) 2.718(9) Å] and an N⋯H–N interaction [N(2)⋯N(3) 3.079(10) Å]. This description of the hydrogen bonds, together with the structural information on the C–O and C=O distances, indicate that the compound can be described as a neutral hydrogen bonded adduct. Fig. 2 shows a comparison between the X-ray powder diffractogram of the ground polycrystalline product (top) and that calculated on the basis of the structure determined by single crystal X-ray diffraction (bottom). Even though the crystallinity of the sample of {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂ obtained via grinding is not high, all significant peaks can be easily recognized.

Fig. 2 {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂. Comparison between the X-ray powder diffractogram of the ground polycrystalline product (top) and that calculated on the basis of the structure determined by single crystal X-ray diffraction (bottom).

Solid state NMR characterization of the {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂ co-crystal

All the NMR data are listed in Table 1 and the number of carbon atoms follows the numeration proposed in Scheme 1. The ¹³C CPMAS spectrum of {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂ (Fig. 3) is characterized by the presence of a signal at 172.3 ppm attributed to the anthranilic COOH group (C16). The chemical shift value is typical of a carboxylic group, in agreement with the formation of an O–H⋯N interaction [N(1)⋯O(2) 2.609(7) Å] with no proton transfer from the acid to the nitrogen base. A similar chemical shift for this group (171.5 ppm) has been previously observed by Harris and Jackson for the neutral molecule present in form I of the free acid, confirming the presence of a COOH instead of a COO^– group.¹⁹

Fig. 3 ¹³C CPMAS NMR spectrum of the co-crystal {[Fe(η⁵-C₅H₄ C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂ recorded at 5.5 kHz.

Scheme 1

Table 1 ¹³C and ¹⁵N NMR data for {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂. For the assignment we followed the carbon numeration shown in Scheme 1

δ ^13C/ppm	Note	δ ^15N/ppm	Note
172.3	C16	276.9	N2–H⋯N3
153.9	C10	243.1	N1⋯H–O2
151.0	C8	45.3	N3H2
148.8	C8′
147.0	C6, C9, C9′
144.2	C6′
132.5	C12, C14
122.1	C7, C8
119.7	C13, C7′
116.8	C15, C8′
113.8	C11
82.4	C1
81.3	C1′
71.7	C2, C5, C2′, C5′
67.2	C3, C4, C3′, C4′

---

The pyridine and the anthranilic aromatic carbon resonances fall in the range between 154 and 113 ppm. The complete assignment is reported in Table 1. It is worth noting that it was possible to detect by ¹³C NQS experiment two quaternary carbons (C6 and C6′) for the pyridine moiety at 147.0 and 144.2 ppm. This is due to the presence of two different hydrogen bonds formed between pyridine and anthranilic acid, i.e. the O–H⋯N [N(1)⋯O(2) 2.609(7) Å], and the N⋯H–N interaction [N(2)⋯N(3) 3.079(10) Å]. The difference in hydrogen bonding induces differences in the pyridine rings of the same [Fe(η⁵-C₅H₄–C₅H₄N)₂] moiety. This is confirmed also by the ¹⁵N CPMAS spectrum (Fig. 4), which shows two signals for the nitrogen pyridine atoms at 276.9 and at 243.1 ppm assigned to the nuclei involved in the N(2)⋯N(3) and N(1)⋯O(2) interactions, respectively. With respect to the chemical shift value of the nitrogen atom in pyridine (293.8 ppm), the observed shifts of 16.9 and 50.7 ppm are in agreement with the presence of a weak N(H)⋯N interaction and of a stronger N(H)⋯O hydrogen bond interaction, respectively.

Fig. 4 ¹⁵N CPMAS NMR spectrum of the co-crystal {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂ recorded at 5.0 kHz.

This difference affects also the Cp rings which, in fact, show two signals for the C1 and C1′ carbons (82.4 and 81.3 ppm) and two resonances for the other Cp atoms at 71.7 and 67.2 ppm attributed to C2, C2′, C5, C5′ and to C3, C3′, C4, C4′, respectively.

By comparing our results with the chemical shifts reported by Harris et al. for the free known forms, we can argue that in the {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂ adduct the acidic moiety is in a neutral form similar to that of polymorph I. The only remarkable difference is observed for the C10 carbon atom, whose peak shifts from 148 ppm (free acid) to 153.3 ppm (adduct). This is probably due to the different hydrogen bonding environment around the amine moiety in the adduct with respect to the free acid. In the former two hydrogen bonds are present (N(2)–N(3)_inter (3.079 Å) and N(3)–O(1)_intra (2.718 Å)), while in the latter three hydrogen bonds are involved (O–N_intra (2.688 Å), N–N_inter (2.872 Å) and O–N_inter (2.894 Å)). The different HB network greatly influences the ¹⁵N chemical shift of the amine nitrogen N(3), with a ¹⁵N resonance at 45.3 ppm for the NH₂ group (Fig. 4) shifted to higher frequencies by about 23.7 ppm with respect to that of the pure anthranilic acid (form I, neutral molecule).

Alternative solvent-free preparations of the co-crystal {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂

In order to gain insight into the factors controlling the process leading to formation of the supramolecular macrocycle {[Fe(η⁵-C₅H₄–C₅H₄N)₂]·[(C₆H₄)NH₂COOH]}₂, we attempted its preparation in different ways, as detailed in the following:

(i) Dry compression

Powdered samples of anthranilic acid form I and of [Fe(η⁵-C₅H₄–C₅H₄N)₂] (both carefully dried for 24 h in a desiccator containing P₂O₅) were gently mixed together at room temperature. An X-ray powder diffractogram showed that no reaction had taken place. The sample was then pressed in an IR pellet maker. The resulting pellet was then ground and subjected to an X-ray measurement. The resulting superimposition of the two reactant powder patterns showed that no reaction had taken place.

(ii) Wet compression, viz. solvent drop and compression

The same experiment was repeated with the addition of a drop of MeOH to the mixture before compression. The XRPD pattern showed formation of the compound {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂, thus indicating that the tiny amount of solvent was necessary for the reaction to take place.

(iii) “Digestion” in the presence of solvent vapour

The same reaction was carried out by placing an equimolar mixture of anthranilic acid and ferrocenyl complex in a closed beaker containing MeOH vapour and leaving the reactants in the MeOH atmosphere. After 5 d, the change of colour, from orange–red to purple indicated formation of the compound {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂, which was then ascertained by powder diffraction. The important role of solvent vapours in increasing the efficiency of a solid state reaction was also reported by Toda et al.^1h

(iv) Thermally induced reaction

Finally, the reaction between anthranilic acid and [Fe(η⁵-C₅H₄–C₅H₄N)₂] was carried out by heating the mixture of reactants directly under X-ray diffraction conditions. The process was also analysed by differential scanning calorimetry , which shows that the polymorphism of anthranilic acid does not play any significant role, because transition from form I to form II takes places in the heating process at 90 °C.

The thermally induced reaction experiment requires a more detailed description of the results. Fig. 5a shows the variation with temperature of the powder diffraction patterns collected on a static mixture of the reactants from 25 to 140 °C. It can be seen that, on increasing the temperature to ca. 130 °C, peaks of the product start to appear, and they are evident at 140 °C (which corresponds to the melting point of anthranilic acid), viz. the reaction occurs between a crystalline solid and a melt. Fig. 5b shows a comparison between the room temperature pattern after reaction and the calculated pattern.

Fig. 5 Thermally induced reaction of a mixture of solid [Fe(η⁵-C₅H₄–C₅H₄N)₂] and [(C₆H₄)NH₂COOH]. X-Ray diffraction patterns were collected on a static mixture of the reactants from 25 to 140 °C. (a) The variation with temperature of the XRPD pattern. (b) Comparison between (top) the experimental pattern measured at room temperature after the reaction and (bottom) the calculated one. Partial sublimation of the anthranilic acid prevents completeness of the reaction, as indicated by the presence of unreacted [Fe(η⁵-C₅H₄–C₅H₄N)₂] (starred peaks).

The previous reaction was also conducted in a glass vial immersed in an oil bath, in order to show that the reaction can be visually detected. Solid [Fe(η⁵-C₅H₄–C₅H₄N)₂] and [(C₆H₄)NH₂COOH] are orange and white, respectively, at room temperature, while the colour of the product is purple. Fig. 6 shows the glass vial at room temperature and after the bath temperature has reached 143 °C.

Fig. 6 The orange–white mixture of solid reagents (left) and the purple solid product (right) can be visually appreciated if the thermally induced reaction is carried out in a glass vial.

In summary, kneading (i.e. manual grinding of a wet mixture), wet compression and vapour digestion achieve the same result as the static heating of the reactants and lead to formation of the product {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂, while dry compression and dry grinding do not. Altogether, these experiments indicate that the process leading from solid anthranilic acid and solid [Fe(η⁵-C₅H₄–C₅H₄N)₂] to the co-crystal can be more appropriately described as a solvent-free, rather than a solid-state reaction because, in order to occur, it requires either the intervention of a solvent, albeit in catalytic amount, or that one of the reactants is in the liquid state.

Conclusions

In this paper we have reported that solid [Fe(η⁵-C₅H₄–C₅H₄N)₂] reacts with the anthranilic acid (C₆H₄)NH₂COOH to generate the hydrogen bonded supramolecular macrocycle {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂. The product has been characterized by means of X-ray powder diffraction, ¹³C and ¹⁵N solid-state NMR and of single crystal X-ray diffraction. Both spectroscopic and diffraction experiments concur to indicate that no protonation of the organometallic molecule takes place, so that the overall structure can be described as a supramolecular macrocycle where two neutral anthranilic acid molecules link together two ferrocenyl units via O–H⋯N hydrogen bonds. This compound belongs to a family of supramolecular adducts of [Fe(η⁵-C₅H₄–C₅H₄N)₂] with organic acids, all obtained by mechanical mixing of the reactants.¹⁸

The focus of this study is, however, on the preparation conditions. Different mixing conditions of the solid reactants [Fe(η⁵-C₅H₄–C₅H₄N)₂] and (C₆H₄)NH₂COOH have been tested with the aim of understanding the factors controlling product formation.

The results of the five experiments described above indicate that product formation is controlled by molecular diffusion, which can be attained by grinding in the presence of “catalytic” amounts of solvent (kneading or wet compression), by exposing the dry mixture of reactants to vapours of the solvent (solvent digestion) or by thermal treatment. In the latter case, we have shown that one of the components, namely anthranilic acid, melts before reaction.

We would thus argue in favour of experimental conditions (grinding, kneading etc.) that can assist diffusion of the molecules of anthranilic acid in the lattice of the ferrocenyl complex. Clearly, once the reaction has occurred and the new supramolecular bonds between the organic and the organometallic molecules are established, the new crystal phase needs to nucleate and grow as the remaining crystalline phase is destroyed. The driving force appears to be the formation of the hydrogen bonds between the anthranilic acid and the ferrocenyl complex, a highly favourable process because there are no strong hydrogen bonds in the crystals of the organometallic molecules while those in the organic crystals have been demolished by thermal treatment and/or by the intervention of methanol in a process of “local solvation ”. It can be correctly argued that these processes cannot be regarded as bona fide solid state reactions.

Experimental

Anthranilic acid was purchased from Sigma-Aldrich and used without further purification. Reagent grade solvents were used.

Synthesis of Fe[η⁵-C₅H₄-1-(4-C₅H₄N)]₂

The ferrocenyl derivative was prepared as previously reported. A solution of 4-bromopyridine hydrochloride (0.177 g, 0.91 mmol) in dioxane (6.5 mL) was stirred under nitrogen atmosphere with a solution of Na₂CO₃ (1 M, 4.3 mL) to obtain the 4-bromopyridine. The solution was heated at 50 °C for 30 min. PdCl₂[1,1′-bis(diphenylphosphino)ferrocene] (0.006 g, 0.007 mmol) was then added followed by ferrocene diboronic acid (0.100 g, 0.37 mmol) and NaOH (3 M, 0.24 mL) in DME (3 mL). The solution was refluxed for 2 d, quenched with water and extracted with ethyl acetate (3 × 10 mL). The organic layer was washed with NH₄Cl and water, dehydrated with Na₂SO₄ and concentrated. The crude product was purified by column chromatography CH₂Cl₂/MeOH (95 ∶ 5) (yield 53%).

Solid-state preparation of {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂

A stoichiometric 1 ∶ 1 mixture of [Fe(η⁵-C₅H₄–C₅H₄N)₂] and (C₆H₄)NH₂COOH was manually ground in an agate mortar for 5 min and subjected to X-ray powder diffraction measurement (see below). A change in colour from orange–red to purple was also diagnostic of product formation. It is noteworthy that the same solid-state reaction carried out with an excess of either reactant invariably led to formation of a mixture of the same product and of the reactant in excess. Single crystals of {[Fe(η⁵-C₅H₄–C₅H₄N)₂][(C₆H₄)NH₂COOH]}₂ suitable for X-ray diffraction were obtained by slow evaporation of a solution obtained dissolving [Fe(η⁵-Cp–C₅H₄N)₂]₂ and (C₆H₄)NH₂COOH in 5 mL of methanol 99.8%.

Solution synthesis of {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂

[Fe(η⁵-C₅H₄–C₅H₄N)₂] and (C₆H₄)NH₂COOH were dissolved in stoichiometric amount in methanol 99.8% and stirred for 5 min. The solution colour changed from orange–red to purple. A polycrystalline product was then recovered after evaporation of the solvent to dryness. The nature of the product was confirmed by X-ray powder diffraction.

Digestion of {[Fe(η⁵-C₅H₄–C₅H₄N)₂] [(C₆H₄)NH₂COOH]}₂ in the presence of methanol vapour

The compounds obtained from grinding reactions were exposed at room temperature to vapours of methanol (99.8%). After 7 days the diffraction pattern showed that crystallinity of all samples had improved.

Crystal structure determination

Crystal data were collected at room temperature on a Nonius CAD4 diffractometer. Crystal data and details of measurements are summarised in Table 2. Mo Kα radiation, λ = 0.71073 Å, monochromator graphite. SHELX97^20a was used for structure solution and refinement based on F². All non-hydrogen atoms were refined anisotropically. Hydrogen atoms bound to N and O atoms were directly located from a Fourier difference map and refined, while the H_CH atoms were added in calculated positions. Data were corrected for absorption by azimuthal scanning of high-χ reflections. SCHAKAL99^20b was used for the graphical representation of the results. The program PLATON^20c was used to calculate the hydrogen bonding interactions. These data can be obtained free of charge viahttp://www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB21EZ, UK; fax: (+44)1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk). CCDC reference number 621628. For crystallographic data in CIF or other electronic format see DOI: 10.1039/b613569b

Table 2 Crystal data

Compound	{[Fe(η^5-C5H4–C5H4N)2][(C6H4)NH2COOH]}2
Formula	C27H23FeN3O2
M	477.33
System	Triclinic
Space group	P1̄
a/Å	9.565(9)
b/Å	9.897(4)
c/Å	13.201(4)
α/°	99.77(3)
β/°	94.17(4)
γ/°	115.74(7)
U/Å^3	1094(1)
Z	2
Density/g cm^–3	1.449
µ (Mo Kα)/mm^–1	0.720
Reflections collected	4010
Indep refls, R(int)	3829
Observed [Fo > 4σ (Fo)]	2740
R1 [Fo > 4σ (Fo)]	0.0843
wR2 (all data)	0.2491

---

Powder diffraction measurements

Powder diffraction for all samples was measured on a Philips PW-1100 automated diffractometer, Cu Kα, monochromator graphite, using quartz sample holders; for the pure reagents 30 mg of substance were employed. The program PowderCell 2.2^20d was used for calculation of X-ray powder patterns on the basis of the single crystal structure determinations. Prolonged grinding of the compounds did not alter the diffraction patterns significantly.

Solid state NMR measurements

All ¹³C and ¹⁵N spectra were recorded on a JEOL GSE 270 equipped with a Doty probe operating at 67.8 MHz for ¹³C and at 27.25 MHz for ¹⁵N. A standard cross-polarization pulse sequence has been used with a contact time of 3.5 ms for ¹³C and of 5 ms for ¹⁵N, a 90° pulse of 4.5 ms, recycle delay of 10–15 s and a number of 600–4000 transients. Powdered samples were spun at 4–5 kHz at room temperature in a cylindrical 5 mm od zirconia rotors with sample volume of 120 mL. The ¹³C results are reported with respect to TMS assuming the hexamethylbenzene methyl peak is at 17.4 ppm. ¹⁵N chemical shifts were referenced via the resonance of solid (NH₄)₂SO₄ (–355.8 ppm with respect to CH₃NO₂). For all samples the magic angle was carefully adjusted from the ⁷⁹Br spectrum of KBr by minimising the linewidth of the spinning sideband satellite transitions.

Acknowledgements

We acknowledge financial support from the University of Bologna and Torino and from MIUR (FIRB and PRIN2004).

References

1. (a) R. P. Rastogi, P. S. Bassi and S. L. Chadha, J. Phys. Chem., 1963, 67, 2569; (b) R. P. Rastogi and N. B. Singh, J. Phys. Chem., 1966, 70, 3315; (c) R. P. Rastogi and N. B. Singh, J. Phys. Chem., 1968, 72, 4446; (d) A. O. Patil, D. Y. Curtin and I. C. Paul, J. Am. Chem. Soc., 1984, 106, 348; (e) J. F. Fernandez-Bertran, Pure Appl. Chem., 1999, 71, 581; (f) J. Fernandez-Bertran, J. C. Alvarez and E. Reguera, Solid State Ionics, 1998, 106, 129; (g) N. Shan, F. Toda and W. Jones, Chem. Commun., 2002, 2372; (h) S. Nakamatsu, S. Toyota, W. Jones and F. Toda, Chem. Commun., 2005, 3808.
2. (a) K. Tanaka and F. Toda, Chem. Rev., 2000, 100, 1025; (b) D. Braga and F. Grepioni, Angew. Chem., Int. Ed., 2004, 43, 2; (c) D. Braga and F. Grepioni, in Topics in Current Chemistry, ed. F. Toda, Springer, Berlin, 2005, 254, p. 71; (d) A. V. Trask and W. Jones, in Topics in Current Chemistry, ed. F. Toda, Springer, Berlin, 2005, 254, p. 41; (e) G. Kaupp, in Topics in Current Chemistry, ed. F. Toda, 2005, Springer, Berlin, vol. 254, p. 95; (f) G. W. V. Cave, C. L. Raston and J. L. Scott, Chem. Commun., 2001, 2159; (g) G. Rothenberg, A. P. Downie, C. L. Raston and J. L. Scott, J. Am. Chem. Soc., 2001, 123, 8701.
3. (a) D. W. Brice and D. O'Hare, Inorg. Mater., Wiley & Sons, New York, 1992; (b) M. Gielen, R. Willen and B. Wrackmeyer, Solid State Organometallic Chemistry, Methods and Applications, Wiley & Sons, New York, 1999; (c) V. V. Boldyrev, Solid State Ionics, 1995, 63–65, 537.
4. (a) G. R. Desiraju, Crystal Engineering: The Design of Organic Solids, Elsevier, Amsterdam, 1989; (b) Crystal Engineering: from Molecules and Crystals to Materials, ed. D. Braga, F. Grepioni and A. G. Orpen, Kluwer Academic Publishers, Dordrecht, 1999; (c) M. D. Hollingsworth, Science, 2002, 295, 2410; (d) D. Braga, F. Grepioni and G. R. Desiraju, Chem. Rev., 1998, 98, 1375; (e) D. Braga, G. R. Desiraju, J. Miller, A. G. Orpen and S. Price, CrystEngComm, 2002, 4, 500; (f) G. R. Desiraju, in The Crystal as a Supramolecular Entity. Perspectives in Supramolecular Chemistry, Vol. 2, Wiley, Chichester, 1996, p. 107; (g) D. Braga, G. R. Desiraju, J. Miller, A. G. Orpen and S. Price, CrystEngComm, 2002, 4, 500; (h) Crystal Design, Structure and Function. Perspectives in Supramolecular Chemistry, ed. G. R. Desiraju, Whiley & Sons, Chichester, UK, 2003, vol. 7; (i) D. Braga, L. Brammer and N. Champness, CrystEngComm, 2005, 7, 1; (j) Frontiers in Crystal Engineering, ed. E. R. T. Tiekink and J. J. Vittal, Wiley & Sons, Chichester, 2006.
5. See for instance: (a) M. Ogawa, T. Hashizume, K. Kuroda and C. Kato, Inorg. Chem., 1991, 30, 584; (b) D. Braga, L. Maini and F. Grepioni, Chem. Commun., 1999, 937; (c) J. M. Thomas, R. Raja, G. Sankar, B. F. G. Johnson and D. W. Lewis, Chem.–Eur. J., 2001, 7, 2973; (d) P. J. Nichols, C. L. Raston and J. W. Steed, Chem. Commun., 2001, 1062; (e) R. Kuroda, Y. Imal and T. Sato, Chirality, 2001, 13, 588.
6. (a) M. R. Caira, L. R. Nassimbeni and A. F. Wildervanck, J. Chem. Soc., Perkin Trans., 1995, 2213; (b) L. R. Nassimbeni and H. Su, New J. Chem., 2002, 26, 989; (c) M. R. Caira, L. R. Nassimbeni, F. Toda and D. Vujovic, J. Am. Chem. Soc., 2000, 122, 9367; (d) S. Apel, M. Lennartz, L. R. Nassimbeni and E. Weber, Chem.–Eur. J., 2002, 83, 67849; (e) L. R. Nassimbeni, Acc. Chem. Res., 2003, 36, 631; (f) L. R. Nassimbeni, CrystEngComm, 2002, 5, 200.
7. (a) M. C. Etter, J. Phys. Chem., 1991, 95, 4601; (b) M. C. Etter, S. M. Reutzel and C. G. Choo, J. Am. Chem. Soc., 1993, 115, 4411; (c) W. H. Ojala and M. C. Etter, J. Am. Chem. Soc., 1992, 114, 10228; (d) V. R. Pedireddi, W. Jones, A. P. Chorlton and R. Docherty, Chem. Commun., 1996, 987; (e) A. V. Trask, W. D. S. Motherwell and W. Jones, Chem. Commun., 2004, 890.
8. (a) D. Braga, M. Curzi, A. Johansson, M. Polito, K. Rubini and F. Grepioni, Angew. Chem., Int. Ed., 2006, 45, 142; (b) A. Pichon, A. Lazuen-Garay and S. L. James, CrystEngComm, 2006, 8, 211; (c) D. Braga, S. L. Giaffreda, F. Grepioni, A. Pettersen, L. Maini, M. Curzi and M. Polito, Dalton Trans., 2006, 1249.
9. (a) D. Braga, M. Curzi, M. Lusi and F. Grepioni, CrystEngComm, 2005, 7, 276; (b) A. V. Trask, D. A. Haynes, W. D. S. Motherwell and W. Jones, Chem. Commun., 2006, 51.
10. (a) J. Bernstein, Polymorphism in Molecular Crystals, Oxford University Press, Oxford, 2002; (b) D. Braga and F. Grepioni, in Crystal Design, ed. G. R. Desiraju, John Wiley & Sons, Weinheim, Editor edn, 2003; (c) D. Braga and J. Bernstein, in Making Crystals by Design. Methods, Techniques, Applications, ed. D. Braga and F. Grepioni, John Wiley & Sons, Weinheim, Editor edn, 2006; (d) J. Bernstein, R. J. Davey and J.-O. Henck, Angew. Chem., Int. Ed., 1999, 38, 3440.
11. (a) L. R. MacGillivray, J. L. Reid and J. A. Ripmeester, J. Am. Chem. Soc., 2000, 122, 7817; (b) G. S. Papaefstathiou, A. J. Kipp and L. R. MacGillivray, Chem. Commun., 2001, 2462; (c) D. B. Varshney, G. S. Papaefstathiou and L. R. MacGillivray, Chem. Commun., 2002, 1964; (d) T. Friščić and L. R. MacGillivray, Chem. Commun., 2003, 1306; (e) X. Gao, T. Friščić and L. MacGillivray, Angew. Chem., Int. Ed., 2004, 43, 232; (f) L. R. MacGillivray, CrystEngComm, 2002, 4, 37.
12. (a) Organic Solid-State Reactions, ed. F. Toda, Kluwer Academic Publishers, 2002; (b) K. Tanaka and F. Toda, Solvent-Free Organic Synthesis, Wiley-VCH, London, 2003; (c) F. Toda, CrystEngComm, 2002, 4, 215–222; (d) G. Kaupp, Angew. Chem., Int. Ed. Engl., 1992, 31, 592; (e) G. Kaupp, Angew. Chem., Int. Ed. Engl., 1992, 31, 595; (f) K. Tanaka, F. Toda, E. Mochizuki, N. Yasui, Y. Kai, I. Miyahara and K. Hirotsu, Angew. Chem., Int. Ed., 1999, 38, 3523; (g) V. P. Balema, J. W. Wiench, M. Pruski and V. K. Pecharsky, Chem. Commun., 2002, 1606; (h) V. P. Balema, J. W. Wiench, M. Pruski and V. K. Pecharsky, Chem. Commun., 2002, 724; (i) V. P. Balema, J. W. Wiench, M. Pruski and V. K. Pecharsky, J. Am. Chem. Soc., 2002, 124, 6244.
13. G. Kaupp, CrystEngComm, 2003, 5, 117.
14. (a) D. Braga, L. Maini, S. L. Giaffreda, F. Grepioni, M. R. Chierotti and R. Gobetto, Chem.–Eur. J., 2004, 10, 3261; (b) D. Braga, L. Maini, M. Polito, L. Mirolo and F. Grepioni, Chem.–Eur. J., 2003, 9, 4362; (c) D. Braga, M. Curzi, F. Grepioni and M. Polito, Chem. Commun., 2005, 2915.
15. D. Braga, M. Polito, M. Bracaccini, D. D'Addario, E. Tagliavini and L. Sturba, Organometallics, 2003, 22, 2142.
16. (a) H. Takazawa, S. Ohba and Y. Saito, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1986, 42, 1880; (b) T. H. Lu, P. Chattopadhyay, F. L. Liao and J. M. Lo, Anal. Sci., 2001, 17, 905; (c) C. J. Brown and M. Ehrenberg, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1985, C41, 441; (d) A. T. Trask, N. Shan, W. D. S. Motherwell, W. Jones, S. Feng, R. B. H. Tan and K. J. Carpenter, Chem. Commun., 2005, 880.
17. H. O. William and M. C. Etter, J. Am. Chem. Soc., 1995, 114, 10288.
18. D. Braga, S. L. Giaffreda and F. Grepioni, Chem. Commun., 2006 10.1039/b608691h.
19. R. K. Harris and P. Jackson, J. Phys. Chem. Solids, 1987, 48, 813.
20. (a) G. M. Sheldrick, SHELXL97, Program for Crystal Structure Determination; University of Göttingen: Göttingen, Germany, 1997; (b) E. Keller, SCHAKAL99, Graphical Representation of Molecular Models; University of Freiburg, Germany, 1999; (c) A. L. Spek, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990, 46, C31; (d) PowderCell programmed by W. Kraus and G. Nolze (BAM Berlin) © subgroups derived by Ulrich Müller (Gh Kassel).

---