Sniffing procedure

As the pig house odour research continued, it became clear that the one observer procedure is time consuming. Only two odour measurements per day were possible. Futhermore, panel members observed a stress which was caused by the noise of the olfactometers. Therefore it was decided to order a set of three double sniffing port low noise olfacometers at the University of Utrecht. The Psychological Laboratory of this university had some experience in the construction of olfactometers. The flowrate was set at 10 1/min arbitrary. 

After a considerable construction time the olfactometers became available for use. With these olfactometers and a recently created sniffing room at the IMAG premises, the poultry house odour research was repeated with a panel of eight observers. The necessary diluting air was obtained from a rotary vane compressor. Filtration of the compressed air was applied. The experimental procedures made it possible to use all panel members at the time. As a result, two experimenters were able to complete four odour measurements per day. 

The aroma of foods is caused by volatile compounds which are perceived by the human nose. Many studies (reviews in [1, 2]) have indicated that only a small fraction of the hundreds of volatiles occurring in a food sample contribute to its aroma. To detect these compounds, a method proposed by Fuller et al. [3] is used. In this procedure, which is designated gas chromatography-olfactometry (GC-O), the effluent from a gas chromatography column is sniffed by an expert who marks in the chromatogram each position at which an odour impression is perceived. 

After concentration of the extract by microdistillation [25] or by special procedures [26] to facilitate the identification of the odorants, an aliquot is separated by high-resolution GC and the effluent is split into a flame ionisation detector (FID) and a sniffing port [27]. The positions of the odorants in the gas chromatogram are assessed by sniffing the carrier gas as it flows from the port. This procedure is denoted GC-O. 

There are many samples that can be injected into a gas chromatograph in order to determine what aroma active volatile is present or to quantify a particular one (Maarse and van der Heij, 1994). Unfortunately, many of the protocols that were followed in the past did not rely on meaningful standards or retention indexing, or failed to use GC sniffing or the more formal GC/O procedures to establish that the peaks detected had odor activity however, if the object of the analysis is to simply monitor one or two odorants known to contribute positively or... 

This procedure allows the differentiation of odor active compounds from odorless substances within a complex mixture of volatiles. For decades this procedure has been successfully applied for aroma analyses of foods (Grosch, 1993). The mixture of volatile compounds either collected in a purified organic solvent extract or in a defined headspace volume is separated into its different components by means of GC and the effluent gas flow at the end of the GC capillary column is split between a FID and an experienced test person s nose. By sniffing the column effluent, the human nose is able to perceive the odor active compounds contained in a complex mixture and the test person can mark the corresponding spot in the FID chromatogram recorded in parallel and attribute a certain odor quality. A sample GC—O chromatogram of a solvent extracted material is shown in Figure 8.7. 

To detect the odour-active volatiles. Fuller et al. [6] described a system for the sniffing of GC effluents which was improved and applied to food samples by Dravnieks and O Donnell [7]. The new technique, named GC olfactometry (GCO), was the starting point for the development of a systematic approach to the identification of the compounds which cause food aromas. As summarised in Table 6.23 the analytical procedure consists of screening for key odorants by special GCO techniques, quantification and calculation of OAVs as well as aroma-recombination studies. During the last decade these steps have been critically reviewed by Acree [8], Blank [9], Grosch [10, 11 ], Mistry et al. [12] and Schieberle /fi/. 

Charm analysis and AEDA both constitute screening procedures, as the results are not corrected for the losses of odorants during the isolation procedure. Furthermore, in GCO, the odorants are completely volatilised and then evaluated by sniffing, whereas the volatility of the aroma compounds in foods depends on their solubility in the aqueous and/or oily phase as well as their binding to non-volatile food constituents. To elucidate which of the compounds revealed in the dilution experiments contribute with high OAV to the aroma, quantification of the odorants with higher FD factors and calculation of their OAVs are the next steps in the analytical procedure. 

To conclude this section, it should be recalled that data treatment is an important part of the analytical approach. The results of a SNIF-NMR experiment constitute a matrix of data where the variables are the isotope ratios or signal intensities of the different isotopomers and the individuals are the NE observations for a given sample. In this sense, SNIF-NMR is a second-order procedure (ref. 3) and multi-variate analysis is the appropriate method for evaluating the results if a linear behaviour of the variables may be assumed. 

ACT-R is composed of perceptual-motor and memory modules implemented as production rules. Both declarative and procedural memory are modeled. A pattern-matching mechanism is used to select production rules applicable to the situation. Buffers are used to interact with and represent the state of modules. ACT-R is implemented as a computer language. Users create a program containing relevant task data to model the task and task performance. An extension of ACT called SNIF-ACT that incorporates a Bayesian navigation mechanism was described earlier in this chapter (Fu and Pirolli 2007). 

Sniffing time of odors of urine, feces and mid-ventral gland secretion of conspecifics of own and opposite sex was examined in a two-stimulus test (Beauchamp, 1973 Johnston, 1981) the order of testing with different odors varied across subjects. There were at least 48 hours between two tests on the same subject. Technically, the procedure consisted of presentation of a pair of stimuli applied on cotton swabs that were placed in glass tubes (internal diameter 0.5 cm.) at a depth of 0.5 cm. from the open end. A pair of tubes, 4 cm apart, were inserted approximately 1 cm. into the cage through the wire lid. A new pair of glass tubes was used for each presentation. 

An on-line method for wine classification was also suggested by Fauhl and Wittkowski. H NMR spectroscopy has also been used to aid determination of site-specific D/H ratios in glycerol from different wine sources, since glycerol is found in wine as a byproduct of glycolysis. The applications of the SNIP method have also been extended to spirits and brandies. Since many of these studies are based on ethanol determination, a critical analysis of the accuracy of the SNIF-NMR method compared with other methods such as a time domain least-squares procedure has been carried out. ... 

The next development has been to do site specific stable isotope analysis (SNIF-NMR). While total or D can now be done by combined GC-MS analysis, which makes the method both sensitive and rapid, site specific analysis requires the isolation and purification of components and nmr time [28]. This adds complexity to the procedure, and it must be recognized that the analysis potentially will alter the isotope distribution/composition of the component being monitored [29]. However, Remaud et al. [28] have demonstrated how this analysis can be applied to the detection of adultered vanilla. 

Until recently, there has been only one report of an attempt to quantitate an odorant using GC-O. Probits of NIF values of l-octen-3-one in a model solution and in coffee were compared to a calibration curve (38). Results showed that GC- SNIF can compete with most sensitive and selective techniques, such as tandem-MS, to quantify extremely intense odorants. In the present example, the GC-O sensitivity was 75 to 500 times higher than MS for the quantitation procedure. 

---