1. Introduction
1.1 Beginnings
This database was conceived of in about 2002 after two of us (RL and XZ) attended the 21st conference on Dyes in History and Archaeology in Avignon, France. The Boston University Chemistry Department had recently acquired a HPLC system with tandem diode array and mass detectors (HPLC-DAD-MSD). We decided this would be useful for characterizing textile dyes since little mass spectrometric data on dyes was available at that time.
We chose to focus on yellow dyes because there are a lot of them and because they had not been well characterized as dyes. In particular, the existing methods for extracting dyes from textile fibers utilized strongly acidic conditions (e.g., 3 M HCl) that cleaved glycosidic bonds. Thus the sugars attached to aglycones in flavonoids, the largest category of yellow dyes (there are over 8000 different known flavonoids), were lost. This led to the development of “soft” methods that permitted extraction without loss of the sugars. In our own work, our initial studies utilized methanol and formic acid (a weak acid), but now we use water and pyridine (a very weak base), with oxalic acid to complex with the mordant (see section 3.2.7).
This database is not exhaustive, but contains only materials that we have worked with ourselves.
1.2 Caveats
This database should be used as a guide, rather than as a statement of the absolute truth, as the data here are subject to a number of possible errors and variables:
1. Specimen variation. The metabolites (e.g., dye components) produced by plants can vary depending on many environmental factors [time of harvest, soil, weather, and other conditions], as well as post-harvest processing. Thus a plant may produce variable amounts of individual colorants. However, specific species of plants have the same genes and thus generally produce the same colorants, though the ratio of one to another may vary significantly.
2. Errors in plant identification. Many of these specimens were obtained or provided by persons, including some of the compilers, who are not trained plant taxonomists.
3. Errors in interpretation of data. The compilers have attempted to provide accurate interpretations but may not have always succeeded.
4. Typographical errors.