Problems with nomenclature are necessary evils in the development of a new subject. As the subject of crystal engineering progresses, it is possible and indeed advantageous to assess terminology, keep what is good, and discard what is not required. Names are coined easily in the early days of a subject. This arises partly from enthusiasm and partly because it is difficult to describe new concepts with old names. Or so we seem to feel. As crystal engineering has evolved into a more mature discipline, one also realizes that some of the terms that we used during the last 15 years or so in describing crystals and their design and properties might not have been really necessary and that standard terminology in the existing chemical and crystallographic literature might well have sufficed. More particularly, I refer to the term co-crystal (also written as cocrystal). I do not know exactly when this term came into the literature. Perhaps it was in the paper of Etter and Panunto on the complexing ability of 1,3-bis(m-nitrophenyl)urea. Whatever the case, the term gained easy acceptance. The operational word here is ‘easy’. Anyone could understand that a co-crystal was what one got in a co-crystallization experiment, and gradually the term came to be used for just about any twoand higher-component crystal. At this point, it would do one good to remember that we had (and still have) a perfectly good word to describe multi-component crystals, which have specific noncovalent interactions between the distinct molecules, and this is molecular complex. For someone whose serious early chemical education included the study of Foster’s book on donor– acceptor complexes and Herbstein’s illuminating work in this area (for those too young to know or care, my Ph.D. topic was in the area of quinone–hydroquinone donor acceptor complexes), the term molecular complex was perfectly acceptable and I have used it invariably in my papers. Recently, however, the term co-crystal slipped into a paper (that too in the title) where I was not the corresponding author and this has prompted the present letter. I was always uncomfortable with the term co-crystal but not in ways I could express satisfactorily. For a start, the word crystal is too important, too meaningful and too evocative. For those in the subject of crystal engineering, it is a sacrosanct word—something like bond to a chemist. Given the centrality of the concept of a crystal, what is a co-crystal? I never had any problems with co-crystallization or co-crystallize and indeed I use these terms regularly. It was the simple, easy word cocrystal that caused me problems. Now, a crystal is described both by a lattice, which gives us the symmetry information, and also by the atoms/ions/molecules that are contained in it, the chemical information. So how can something like 1 : 1 1,3bis(m-nitrophenyl)urea–triphenylphosphine oxide masquerade as a co-crystal, when it has nothing in common with the crystal of either 1,3-bis(m-nitrophenyl)urea or that of triphenylphosphine oxide? What is coto what? In the end, would brass become a co-crystal of copper and zinc? Or are my questions too naı̈ve? It was Herbstein again who, through his recent paper on composite crystals of 5-oxatricyclo[5.1.0.0]octan-4-one, resolved my nebulous doubts and brought my objections to the present use of the term co-crystal into sharper focus. He defines a composite crystal as something that is formed by ‘ordered agglutination of crystals of the same or different type’. He is speaking literally about two different crystals stuck together at the molecular level. This to my mind is a real cocrystal—two crystals that are joined together. In his paper written 40 years ago on hexabromobenzene, he provides us with another beautiful example of a composite crystal. Because this crystal is pseudo-hexagonal with c/a y 1/d3, planes like (101) and (1 0 21) or (301) and (001) coincide, but not exactly, and what one gets is a mosaic of crystals joined along these nearly common planes. In a recent authoritative work, Chapuis and his group discuss urotropin azelate. These authors actually call this an ‘unwilling co-crystal’ in their paper title because each of the components, urotropin and azelaic acid, retain many of their original crystal traits in the conjunction of these substances. Now this is what I would call a good co-crystal because, once more, vestigial influences of former beings persist. However, so prevalent has the present connotation of the term co-crystal become that something that is a really good example of a co-crystal is referred to in almost exactly the opposite terms! The Chapuis group have another paper on the s-structure of b-tantalum, and here the composite portions of a true co-crystal have different space groups. This is not all. Alivisatos and co-workers have recently shown how a nanocrystal tetrapod of CdTe grows. When it nucleates, it starts off as a cubic phase and when it branches out it switches to a hexagonal phase. Two crystals living together—I’d call this a co-crystal for sure. I will round off this selection of examples with Boese’s fascinating studies of what he calls oligodiffractometry. Here we have crystals of polymorphs that are necessarily examined together in the diffraction experiment and their patterns deconvoluted later—a different type of cocrystal maybe but still a co-crystal. In none of these cases, however, has my image of a crystal been dimmed and yet we have two or more crystals somehow connected with each other intimately. So what do we do with the 1 : 1 1,3-bis(m-nitrophenyl)urea– triphenylphosphine oxide co-crystal? For a start, I’d revert to calling it the 1 : 1 1,3-bis(m-nitrophenyl)urea–triphenylphosphine oxide molecular complex. This is a good term and a clear one at that. It tells us what is happening, and it is general enough. It may be distinguished from the terms solid solution or inclusion compound and the reader who is further interested in all these terminologies is well advised to read Kitaigorodskii’s book. To conclude, when new terms enter the literature in a big way, and there is some controversy regarding their use, I would suggest that there are two possible courses of action. If the term is merely ambiguous, it may be easiest to just retain it because it has become too common. A good example of this situation is the use of the term pseudopolymorph. But if something is scientifically suspect it must go, howsoever popular it might be. What is easy is sometimes not what is best. Letter

Crystal and co-crystal

The explicit hydrogen version of the transferable potentials for phase equilibria (TraPPE-EH) force field is extended to benzene, pyridine, pyrimidine, pyrazine, pyridazine, thiophene, furan, pyrrole, thiazole, oxazole, isoxazole, imidazole, and pyrazole. While the Lennard-Jones parameters for carbon, hydrogen (two types), nitrogen (two types), oxygen, and sulfur are transferable for all 13 compounds, the partial charges are specific for each compound. The benzene dimer energies for sandwich, T-shape, and parallel-displaced configurations obtained for the TraPPE-EH force field compare favorably with high-level electronic structure calculations. Gibbs ensemble Monte Carlo simulations were carried out to compute the single-component vapor−liquid equilibria for benzene, pyridine, three diazenes, and eight five-membered heterocycles. The agreement with experimental data is excellent with the liquid densities and vapor pressures reproduced within 1 and 5%, respectively. The critical temperatures and normal boi...

Transferable potentials for phase equilibria. 9. Explicit hydrogen description of benzene and five-membered and six-membered heterocyclic aromatic compounds

Three novel cadmium-organic frameworks built-up from 1,3,5-benzenetricarboxylate anions (HXBTC(x-3)) and 4,4'-trimethylenedipyridine (TMD) have been hydrothermally synthesized, and characterized using single-crystal X-ray diffraction, thermoanalytical measurements, elemental analysis, and IR and Raman spectroscopies: [Cd(HBTC)(TMD)(2)].8.5H(2)O (I), [Cd(HBTC)(TMD)(H(2)O)].4.5H(2)O (II), and [Cd(2)(BTC)(TMD)(2)(NO(3))].3H(2)O (III), with structures I and II being isolated as a mixture of crystals. Structure I contains an undulating infinite two-dimensional [Cd(HBTC)(TMD)(2)] framework, with a (4,4) topology and rectangular pores, ca. 3.4 x 11.0 A in cross-section, distributed in a herringbone manner. The crystal structure of I is obtained by parallel packing of this 2D framework in an [ABAB.] fashion. Compound II has a porous 3D diamondoid framework with channels running in several directions of the unit cell, which allows 2-fold interpenetration to occur. The most prominent channels are distributed in a brick-wall fashion along the c axis and have a cross-section of ca. 3.2 x 13.2 A. Structure III can be seen as the three-dimensional assembly of binuclear secondary building units (SBU), which leads to a compact, neutral, and coordinatively bonded eight-connected framework, [Cd(2)(BTC)(TMD)(2)(NO(3))], exhibiting an unusual 3(6)4(22) topology. The increased flexibility of the TMD ligands (brought about by the three methylene groups between the two 4-pyridyl rings) can lead, for the same reactive system, to a large variety of crystal architectures.

Two- and Three-Dimensional Cadmium−Organic Frameworks with Trimesic Acid and 4,4‘-Trimethylenedipyridine

Crystallization of 1,3,5-cyclohexanetricarboxylic acid (CTA) from EtOH affords the 1:1 hydrate, CTA.H 2 O, with 4-fold inclined interpenetrated (6,3) hydrogen-bonded networks. Crystallization of CTA with 4,4'-bipyridine (bipy) furnishes the complex, CTA.bipy.H 2 O (2:3:1), that has 3-fold interweaving (6,3) networks with parallel interpenetration. The striking similarity of these hydrogen bond networks to those found in the crystal structure of trimesic acid and its complex with bipy suggests that such interpenetrated networks may be engineered using retrosynthetic strategies.

Four-Fold Inclined Interpenetrated and Three-Fold Parallel Interpenetrated Hydrogen Bond Networks in 1,3,5-Cyclohexanetricarboxylic Acid Hydrate and Its Molecular Complex with 4,4‘-Bipyridine

Boronic acids in the design and synthesis of supramolecular assemblies

Topological equivalences between organic and coordination polymer crystal structures: an organic ladder formed with three-connected molecular and supramolecular synthons.

A new supramolecular crystal engineering approach toward quadratic nonlinear optics is inferred from an acentric cocrystalline template made of two different bi-dimensional octupolar molecules:  tr...

Gautam R. Desiraju

Papers

Topological equivalences between organic and coordination polymer crystal structures: an organic ladder formed with three-connected molecular and supramolecular synthons.

Coupling octupoles in crystals: The case of the 1,3,5-trinitrobenzene-triphenylene 1:1 molecular co-crystal