382
E. R. T. Tiekink
19.1 Introduction
The intriguing term kryptoracemate has been ascribed [1] to Ivan Bernal, a renowned
chemical crystallographer with a particular interest in chirality in crystals, especially
in metalorganic systems. It is understood that Ivan Bernal employed the term during a
meeting of the American Crystallographic Association in the mid-1990s [2]. Clearly,
the term refers to the crystallisation of racemic compounds. Less obvious is the origin
of the prefix “krypto”, which comes from the Greek and loosely translates as hidden.
Thus, kryptoracemic behaviour refers to crystallisation of racemic molecules about
a hidden, i.e. a non-crystallographic, centre of inversion. For organic molecules, the
most recent comprehensive review of the topic by Fábián and Brock [3] highlights
how rare this phenomenon is, accounting for only 0.1% of all-organic structures
included in the Cambridge Structural Database (CSD) [4]; a complimentary survey
of kryptoracemic behaviour in metalorganic systems has been published by Bernal
and Watkins [5]. The aim of the present chapter is to alert the reader to this crystallographic peculiarity by surveying the structural characteristics of representative
examples of all-organic molecules known with a certain degree of confidence to
exhibit kryptoracemic behaviour.
There are at least three possible outcomes from the crystallisation from a solution of a racemic compound. In descending order of importance, these crystallisation
outcomes are: (i) an ordered racemic crystal, (ii) a physical mixture comprising equal
numbers of enantiopure crystals and (iii) a kryptoracemate; herein, solid solutions
and enantiomorphous twins are ignored. Racemic compounds have equal numbers
of the enantiomers in the crystal, i.e. an equal distribution of mirror images of the
molecules. This is normally accomplished by having the molecules disposed about
a crystallographic centre of inversion. There are exceptions to this general principle,
whereby a racemate can crystallise in a non-centrosymmetric space-group, but a
space-group having glide symmetry so that the criterion of having equal numbers
of mirror images pertains. An early survey of this later phenomenon by Dalhus
and Görbitz [6] revealed that 90% of crystals in this category crystallised in five
non-centrosymmetric space-groups, namely Pc, Cc, Pca2 1 , Pna2 1 and Fdd2. The
second phenomenon leading to a physical mixture of crystals, or a conglomerate
of homochiral crystals, is usually termed spontaneous resolution and was famously
recognised by Pasteur in his work on ammonium sodium tartrate crystals [7]. The
third crystallisation outcome for racemic compounds results in kryptoracemic crystals which are characterised as crystallising in space-groups lacking a centre of inversion, mirror planes and rotary-inversion centres, i.e. not having symmetry operators
of the second kind, as discussed further below.
The reality is that most organic materials containing a pair of resolvable isomers
(also those with meso-symmetry and achiral compounds) crystallise in one of the
centrosymmetric space-groups leading to hetero-chiral crystals; this has been termed
an enantiophilic trait [8]. It is estimated that 99% of molecules (neutral and charged)
that can crystallise in a centrosymmetric space-group will do so [9]. This is probably
E. R. T. Tiekink
19.1 Introduction
The intriguing term kryptoracemate has been ascribed [1] to Ivan Bernal, a renowned
chemical crystallographer with a particular interest in chirality in crystals, especially
in metalorganic systems. It is understood that Ivan Bernal employed the term during a
meeting of the American Crystallographic Association in the mid-1990s [2]. Clearly,
the term refers to the crystallisation of racemic compounds. Less obvious is the origin
of the prefix “krypto”, which comes from the Greek and loosely translates as hidden.
Thus, kryptoracemic behaviour refers to crystallisation of racemic molecules about
a hidden, i.e. a non-crystallographic, centre of inversion. For organic molecules, the
most recent comprehensive review of the topic by Fábián and Brock [3] highlights
how rare this phenomenon is, accounting for only 0.1% of all-organic structures
included in the Cambridge Structural Database (CSD) [4]; a complimentary survey
of kryptoracemic behaviour in metalorganic systems has been published by Bernal
and Watkins [5]. The aim of the present chapter is to alert the reader to this crystallographic peculiarity by surveying the structural characteristics of representative
examples of all-organic molecules known with a certain degree of confidence to
exhibit kryptoracemic behaviour.
There are at least three possible outcomes from the crystallisation from a solution of a racemic compound. In descending order of importance, these crystallisation
outcomes are: (i) an ordered racemic crystal, (ii) a physical mixture comprising equal
numbers of enantiopure crystals and (iii) a kryptoracemate; herein, solid solutions
and enantiomorphous twins are ignored. Racemic compounds have equal numbers
of the enantiomers in the crystal, i.e. an equal distribution of mirror images of the
molecules. This is normally accomplished by having the molecules disposed about
a crystallographic centre of inversion. There are exceptions to this general principle,
whereby a racemate can crystallise in a non-centrosymmetric space-group, but a
space-group having glide symmetry so that the criterion of having equal numbers
of mirror images pertains. An early survey of this later phenomenon by Dalhus
and Görbitz [6] revealed that 90% of crystals in this category crystallised in five
non-centrosymmetric space-groups, namely Pc, Cc, Pca2 1 , Pna2 1 and Fdd2. The
second phenomenon leading to a physical mixture of crystals, or a conglomerate
of homochiral crystals, is usually termed spontaneous resolution and was famously
recognised by Pasteur in his work on ammonium sodium tartrate crystals [7]. The
third crystallisation outcome for racemic compounds results in kryptoracemic crystals which are characterised as crystallising in space-groups lacking a centre of inversion, mirror planes and rotary-inversion centres, i.e. not having symmetry operators
of the second kind, as discussed further below.
The reality is that most organic materials containing a pair of resolvable isomers
(also those with meso-symmetry and achiral compounds) crystallise in one of the
centrosymmetric space-groups leading to hetero-chiral crystals; this has been termed
an enantiophilic trait [8]. It is estimated that 99% of molecules (neutral and charged)
that can crystallise in a centrosymmetric space-group will do so [9]. This is probably