Prehistory of FIA

There is universal agreement that the paper entitled Flow Injection Analysis. Part 1. A New Concept of Continuous Flow Analysis [1], which appeared in Analytica Chimica Acta in the spring of 1975, is the very first publication in which the name of the method was introduced, and for this reason alone this reference rightfully appears as the first one in the List of References. Any reader interested in the history of FIA will find, however, by further scrutiny of this paper that, besides coining the new, name, it also contains the concepts and experiments outlining the principle and the scope of this new method, and that its immediate sequels, published with our co-workers and associates from Brazil, Denmark, England, India, and Sweden [1-20], mapped the novel area and plotted the course of future research. 

There have been, however, expressed other views and claims as to the discovery of FIA and they, besides offering interesting reading [216, 607, 1056, 1352], confirm that the history of human endeavour, and scientific discoveries in particular, tells many stories, often confused by conflicting claims. Valcarcel and Luque de Castro have in Chapter 2 of their FIA monograph [665] reviewed material pertinent to this topic published prior to 1975. The reason for selecting this year is that the principles and experiments described in the first FIA publication were summarized in an earlier Danish Patent Application filed in September 1974 [1]. 

Considering that FIA is based on sample injection, controlled dispersion of the injected sample zone, and reproducible timing and that it is physically embodied in the system shown in Fig. 2.4, it becomes evident that none of the papers published until 1975 conforms with that description. Closest to this principle was undoubtedly the work of Nagy et al. [7.1], who suggested a novel approach to hydrodynamic voltammetry, based on the injection of a small sample volume into a supporting elec- 

Reproducible results were obtained only if rapid and total homogenization of the solution to be analyzed and the supporting electtrolyte was ensured. Preliminary experiments and visual observations of the streamlines by a suitable technique showed that the test solution injected into the supporting electrolyte did not mix with the latter to the desired extent, even at high flow rates. The reproducibility of the measurement was impeded by a concentration gradient perpendicular to the direction of flow [our emphasis]. 

Frantz and Hare [7.10] described in an internal report in 1973 a system where a sample was inserted into a nonsegmented stream by manually breaking a capillary of a defined volume. A colorimeter was used to measure reaction product. Hare later joined Beecher and Stewart who replaced the awkward sampling device by a chromatographic injection valve 

---