Note: This article was scanned using OCR from the Fall 1986 CCCE Newsletter. Please contact us if you identify any OCR errors.
Computer use has been expanding rapidly in all types of industrial chemical laboratories, so it should surprise few readers that this trend also extends to the flavor and fragrance industries. Since changes in these industries are not always widely followed, the specific directions of computer development may not be familiar to most of us. This book consists of a series of papers presented at a symposium sponsored by the Division of Agricultural and Food Chemistry at the 186th Meeting of the American Chemical Society, Washington, DC, August 28-September 2, 1983.
The first five articles deal with using computers for structure-activity correlations, information handling, and statistical. analysis. Many of the processes in these articles, such as computer-assisted molecular design, information storage, and information retrieval, will be familiar to those who regularly follow computer developments in chemistry, but the specific applications are sufficiently different to be of interest. In addition, sensory scientists must collect and process data from evaluation panels, individuals who express their opinions about the flavor and/or odor of possible new products. These panels play a key part in product development, and the computer can contribute significantly to this work.
Most chemists are aware that the pharmaceutical industry has played a pioneering role in research on computer-assisted molecular synthesis but may not be familiar with similar work done in the flavor and fragrance industry. These papers discuss several methods that are being studied to find structure- activity correlations. Taken together, they present a useful perspective on some of the progress in this area. Even though the relation between sensory properties of flavor compounds and their molecular properties is not yet understood, progress is being made towards predicting the odor and/or taste of new products.
Computer processing of information from evaluation panels is especially important since it can both decrease the cost and increase the reliability of these procedures. The criteria here can be complicated. As an example, the public may associate effectiveness of medicines with unpleasant taste, and consumers can be just as likely to reject a product that tastes too good as one that tastes too bad. To quantitatively measure these trends, statistical analysis plays a role that is as important as analytical instrumentation and measures of organic reactivity.
Of course, the computer is not essential for this work. Acceptable products were produced long before the development of the microprocessor, but the computer offers savings of both time and expense when used for product modeling and optimization. It can streamline the process by recording evaluations, organizing data, and determining the most effective new product formulations.
Chapter Six discusses how to determine whether the purchase of a Laboratory Information Management System (LIMS) is economically justified. Papers on this topic often po1nt out the technical advantages but fail to consider economic factors. This treatment is not detailed, but it does ident1fy the major financial and workload considerations which should be surveyed. It may provide an interesting perspective for those who are thinking about purchasing a LIMS system as well as those who don't normally deal with factors like interest payments, analyst workload, and speed of analysis.
Those readers who feel that instrumentation is overemphasized at the expense of human capabilities should be encouraged by a statement in the next chapter (page 75) that, "Despite the inherent sensitivity and wide applicability of conventional GC and GC/MS techniques, they will never replace the well trained nose as a means of identifying odiferous components." Unfortunately the ego boost is of limited duration, since the author goes on to explain how a new instrument, the Atmospheric Pressure Chemical Ionization Tandem Triple Quadrupole Mass Spectrometer (APCI/MS/MS), is able to analyze air-borne fragrances directly from the atmosphere without intermediate trapping or concentration steps, just like the human nose. The applications used to test this instrument are quite interesting, including a determination of whether men emit different chemicals from their skin than women, and also an investigation of the advertising claims that a certain perfume creates a different aroma on every woman who wears it.
The next three chapters describe other computer-based analytical instruments that are used in the flavor and fragrance industry. An article on near infrared reflectance analysis (NIRA) states that this procedure is becoming very popular for food analysis because it is not only rapid and non-destructive but usually requires little, if any, sample preparation. These advantages should also make the method useful in other areas of chemistry. Gas chromatography has been used for many years to analyze the individual compounds responsible for product flavor, but the effectiveness of this instrument is increased by coupling it with multivariate methods of statistical analysis. The efficacy of this combination is demonstrated here by two studies, an aroma analysis of cigarette smoke from different tobaccos and the correlation of analytical and sensory data for quality control.
The last article on instrumentation describes the use of a commercially available laboratory automation system for chromatographic analysis, data processing, and data analysis of flavor extracts from food. The data treatment again includes statistical procedures known as multivariate analysis. The final chapter concerns the use of a laboratory robot system for routine analytical tasks. This study reports that sample preparation by the robot system offers significant time savings over manual methods. Accuracy and precision are almost identical by robot and manual methods.
The book is printed from camera-ready copy, but all of the chapters are quite clear and readable. Each article provides a short bibliography with references through 1983, and the book includes an adequate index.
Although the computer appliCations outlined in this book are derived from the fields of flavor and fragrance research, many of the techniques should be applicable to other areas of chemistry. These articles could provide an interesting perspective both for those who are currently involved in computer applications in the industrial laboratory, as well as for those who only wish to keep informed of recent developments.
*Professor of Chemistry
SUNY-Oneonta
Oneonta, NY 13820