Content of the CSD

Fig. 3.2. a Number of new entries added to the CSD vs year of publication b Data content of the CSD as megabytes of information vs year of publication... 

In a manner reminiscent of the self-organising maps, the methodology has been applied to produce a subset of the database that represents the best representative of each unique crystal stracture [53]. Thus, a compound with 10 CSD entries, comprising one polymorph determined seven times and a second polymorph determined three times will be reduced to two entries, which are considered to represent the two unique structure types. The details of the applied quality tests are extensive [53], but the result is a list of 231918 structures (derived from 353 666 structures in the November 2005 release) that are considered to be the best representative examples of all unique high-quality stmctures in the CSD [54], In this way, the complete contents of the CSD are reduced to a set of representative structures that contain an equivalent amount of structural information, but without any redundancy. This dataset forms an especially convenient basis for structural searches, since it is free of any duplication. 

The CSD is now approaching 100,000 entries. Hence, it covers a very broad range of both chemical and crystallographic situations. All of these must be encoded in ways which conform to the basic structure of the database, and certain conventions must, perforce, be employed. In this section, we highlight certain aspects of the database content that have not been discussed elsewhere, and which may affect the outcome of certain searches and, hence, the interpretation of numerical results derived from such searches. In particular, we hope to alert readers to areas in which a careful reading of the CSD User Manuals [15] may be a necessity rather than a chore. 

Each individual structure in the CSD is referred to as an entry and each entry is identified by a reference code (refcode) comprising six alphabetic characters and a further two numeric characters which trace its publication history. The information content of a CSD entry (Figure 1), is best summarized in terms of its dimensionality ... 

A question then arises as to whether the CSD recovery is being limited by the preasphaltene content produced from direct products of coal liquefaction or whether by low liquefaction severity a more thermally sensitive product is produced resulting in retrogressive reactions of liquefaction products to "post-asphaltenes." There is some indication that "virgin" preasphaltenes, primary products of coal dissolution, are more easily recovered via CSD as shown in Table VII however, "postasphaltenes" made from thermal regressive reactions are not. 

Of the SRC evaluated, the CSD SRC has the highest potential for success in a utility boiler application from the standpoint of achieving low carbon content fly ash under low N0X (staged) combustion operating conditions. 

It has been widely demonstrated that the preparation of oriented, and in some cases epitaxial films by CSD is possible, despite the relatively large thickness of the films that is deposited in a single step. Lange ° reviewed the various mechanisms that lead to oriented growth, and a variety of factors including reactions at the electrode interface, - organic content within the film," " and the use of seed layers" to promote homoepitaxy have been discussed. The ability to control film properties (remanent polarization, dielectric constant, etc.) through manipulation of film orientation has also been shown. 

The definitions of information content given above often give rise to edge-effects , which are irritating to some users. In particular, the CSD has obvious areas of overlap with both the ICSD and with the PDB. This problem is now being solved via the inclusion of certain structures in both of the databases concerned. Thus, the CSD now contains a number of structure types (e.g. pure metal carbonyls, boranes, etc.) which also occur in the ICSD. Likewise, a number of oligonucleotides may be found in both the PDB and in the CSD. The precise definition of overlap areas is currently under active discussion. 

In an effort to develop a fully automated procedure that can be used to screen the entire CSD, van de Streek and Motherwell considered that some CSD entries are likely to contain errors, which are more likely to be in the atomic coordinates than in the unit-cell parameters. To eliminate the contribution of the unit-cell contents, the structure factor of each reflection in the simulated PXRD profile was set to an arbitrary constant value, thereby producing a normalised reduced-cell PXRD profile that was used for comparison using the WCC procedure. Applying this methodology to the CSD prior to November 2004 produced a list of 2862 unique refcode families which contain reliable pairs of polymorphic structures, amounting to ca. 1 % of all unique chemical compounds in the CSD. Numerous strict criteria were applied in the process, so that the list represents a minimum number of reliable observations rather than an exhaustive list. 

However, it should be noted that most of the EEG studies on physical fatigue is based on the referential EEG data. It has been pointed out by [6] that the referential method may have certain drawbacks since it is dependent on the location of the reference. Thus, it has been suggested using both the Common Average Reference (CAR) method as well as the Current Source Density (CSD) method [7] hand in hand to provide a more complete view of the EEG content. 

The vast majority of results used in this survey arise from X-ray diffraction studies neutron studies comprise less than 1% of CSD content. However, wherever possible we cite relevant neutron and spectroscopic results and discuss (and attempt to quantify) the numerical differences that exist between N-, S- and X-geometries. 

The distributed CSD System comprises the database, together with software for search, retrieval, visualization, and analysis of CSD content. 

---