Fish freshness detection by a computer screen photoassisted based gas sensor array

In the last years a large number of different measurement methodologies were applied to measure the freshness of fishes. Among them the connection between freshness and headspace composition has been considered by gas chromatographic analysis and from the last two decades by a number of sensors and biosensors aimed at measuring some characteristic indicators (usually amines). More recently also the so-called artificial olfaction systems gathering together many non-specific sensors have shown a certain capability to transduce the global composition of the fish headspace capturing the differences between fresh and spoiled products. One of the main objectives related to the introduction of sensor systems with respect to the analytical methods is the claimed possibility to distribute the freshness control since sensors are expected to be "portable" and "simple". In spite of these objectives, until now sensor systems did not result in any tool that may be broadly distributed. In this paper, we present a chemical sensor array where the optical features of layers of chemicals, sensitive to volatile compounds typical of spoilage processes in fish, are interrogated by a very simple platform based on a computer screen and a web cam. An array of metalloporphyrins is here used to classify fillets of thawed fishes according to their storage days and to monitor the spoilage in filleted anchovies for a time of 8 h. Results indicate a complete identification of the storage days of thawed fillets and a determination of the storage time of anchovies held at room temperature with a root mean square error of validation of about 30 min. The optical system produces a sort of spectral fingerprint containing information about both the absorbance and the emission of the sensitive layer. The system here illustrated, based on computer peripherals, can be easily scaled to any device endowed with a programmable screen and a camera such as cellular phones offering for the first time the possibility to fulfil the sensor expectation of diffused and efficient analytical capabilities. (c) 2006 Elsevier B.V. All rights reserved.