Threshold odor intensity

The other results in Table III are those of data sets not having noncolinear physical-chemical properties. Log P was highly correlated with these data sets as well. Ethylesters threshold data in air weis linearly related to log P (eq. 4) while 3-alkyl-2-methoxy pyrazines had threshold odor intensity which was parabolically related to log P (eq. 7). The pyrazine data indicates that 3-alkyl-2-methoxy pyrazines having a log P value of 2.43 would have the most intense odor of the series. 

Odors are characterized by quaUty and intensity. Descriptive quaUties such as sour, sweet, pungent, fishy, and spicy are commonly used. Intensity is deterrnined by how much the concentration of the odoriferous substance exceeds its detection threshold (the concentration at which most people can detect an odor). Odor intensity is approximately proportional to the logarithm of the concentration. However, several factors affect the abiUty of an individual to detect an odor the sensitivity of a subject s olfactory system, the presence of other masking odors, and olfactory fatigue (ie, reduced olfactory sensitivity during continued exposure to the odorous substance). In addition, the average person s sensitivity to odor decreases with age. 

Four characteristics of odor are subject to measurement by sensory techniques intensity, detectability, character (quality), and hedonic tone (pleasantness-unpleasantness) (16). Odor intensity is the magnitude of the perceived sensation and is classified by a descriptive scale, e.g., faint-moderate-strong, or a 1-10 numerical scale. The detectability of an odor or threshold limit is not an absolute level but depends on how the odorant is present, e.g., alone or in a mixture. Odor character or qualit) is the characteristic which permits its description or classification by comparison to other odors, i.e., sweet or sour, or like that of a skunk. The last characteristic is the hedonic type, which refers to the acceptability of an odorant. For the infrequent visitor, the smell of a large commercial bread bakery may be of high intensity but pleasant. For the nearby resident, the smell may be less acceptable. 

The first element, dynamic dilution, provides a reproducible sample for each panelist. The system must minimize the loss of the odorant to the walls of the delivery apparatus, provide clean dilution air of odor-free quality, maintain a constant dilution ratio for the duration of a given test, and have no memory effect when going from high to low concentrations or switching between odorants of different character. The type of mask or port and the delivery flow rate have been found to influence the response of panelists in determining odor threshold and intensity. 

Thresholds may also be determined by extrapolation of dose-response plots. In this approach, the perceived odor intensity is measured at several... 

Compounds used in the database were obtained from literature reports of chemical structure and odor quality (7-11). In Table 1, a list of the compounds comprising the database is given. The macrocyclic and nitroaromatic musks are of strong, medium, weak, or unspecified odor intensity the nonmusks are odorless or have an odor other than musk. Information about odor quality and intensity is contained in the activity label associated with each compound. It should be emphasized that a musk compound labeled as weak, medium, or strong refers only to the change in its odor threshold, not to any change in its odor quality. Structural classes present in the dataset are shown in Fig. 1. Natural musks, whose sources include both rare animal and plant species, are... 

If qualitative data are required, and the enantiomers differ significantly in their odorqual-ity, then acceptable results may be obtained. If odor intensity measurements (OAVs or OSVs) or threshold values are required, then the conditions described above must be obtained if the data is to be of value. Bernreuther et al. (1997) and Koppenhoefer et al. (1994) have published the enantiospecific sensory data for a variety of chiral odorants. 

Considering that at least two studies should have shown effects and that the thresholds should be relevant for the low level indoor exposure range (e.g., below 10mg/m3) then Table 14.5 shows a list of symptoms and thresholds (mg/m3) in relation to M22. Assuming that M22 is a best case, then at exposures below about 2 mg/m3 (TVOC) perceived air quality, odor intensity, irritation of eyes or nose, additional ventilation needed, and cough, are expected to be the most sensitive indicators ofVOC exposures. However, it must be kept in mind that these symptoms are unspecific and may have many other causes. Therefore, the presence or absence of these symptoms cannot infer that VOC concentrations indoors are the responsible agent, only indicate the possibility. 

GC-MS system. The similarity of odors and odor intensities (at relatively equivalent levels) and odor thresholds (in AEDA or Charm experiments) must be used as an additional confirmatory criterion. 

The relationship between the molecular structure of an aroma compound and its threshold is still unclear. Volatility of a compound may not relate to its threshold. For example, the threshold of ethanol (boiling point is 78°C) is much higher than octanol (boiling point is 195°C) or other homologous alcohol. Ethanol has high volatility but low odor intensity. It is often used as a solvent in compounded flavors. 

No simple relationship exists between VOC concentrations of mixtures and odor intensity. However, as TVOC concentration increases, the likelihood of a multicomponent mixture being odorous increases also. Odor thresholds are generally lower than irritative thresholds. The consensus of the group is that although TVOC is expected to be appropriate for irritation effects it may also turn out to be useful for acute and long-term odors (as perceived by the sense of olfaction). 

The TON is the number of times a given volume of the gaseous sample is to be diluted in clean air to bring it to the threshold odor level determined by 50% of a panel of observers. Its intensity is expressed in odor units. 

In still another study Dravnieks (6) correlated 1 1 structural features with odor threshold and suprathreshold data. More recently Dravnieks correlated odor intensity equivalent to 87 ppm (Vol/vol) of 1-butanol with 20 structural features represMted by Wiswesser line notation. The molecular weight term, (log mw), was reported to be the most statistically significant term. 

The use of computer techniques in the correlation of biological activity with substrate physical-chemical properties has received much attention in the area of medicinal chemistry. The use of these techniques, denoted Quantitative Structure Activity Relationships (QSAR), were developed mostly by Hansch and his coworkers eind have been reviewed by Tute te), Purcell et. al. (9) and Dunn (10). These techniques were utilized by Greenberg (1 l) ln the correlation of odor threshold and suprathreshold data with Log P, the log (n-octanol/water partition coefficient). In the same study it was reported that steric and polar effects as measured by the Taft Steric and Polar Constants poorly correlated with odor intensity data. 

Suprathreshold odor intensity data from Dravnieks (7) equating odor intensity equivalent to 87 ppm (vol./vol.) of n-butanol was used since it eliminated errors between laboratories which occur for threshold measurements and the n-butanol reference scale has been approved by the ASTM (24) as a standard method of measuring odor intensity. 

Threshold data for ediphatic edcohols also correlated well with log P and poorly with 2 eind o . Results in nations 11 and 14 indicate a parabolic dependence of alcohol odor intensity upon P. Log Po weis found to be 3.17. Thus aliphatic alcohols having a log P value of 3.17 should have maximum odor intensity beised upon threshold data. Poor correlations were also found for the Charton eind Sterimol parameters. For the threshold data the Sterimol parameters L eind B. were each highly colinear with log P and thus were not included in Teible 1. ... 

Aldehyde and ketone suprathreshold odor intensity correlated well with log P and HB as shown in Table II. No significant relationship between steric or electronic parameters with aldehyde-ketone suprathreshold data was found with the exception of the Sterimol parameter L which was highly correlated to log P (R=0.95). Aldehyde threshold data weis found to be linearly related to log P as shown in equations 10 eind 13. The same data was poorly correlated with E andvas shown in Table II (eq. 12, 15 and 16). Note that two different aldehyde threshold data sets from two different sources produced very similar equations having slopes, intercepts, correlation coefficients and standard deviations which are not statistically different at the 95% level of confidence (eq. 10 and 13). 

The use of log P and HB parameters as a tool for predicting odor intensity seems promising. Although many excellent correlations were obtained as presented in Tables I-V further studies are needed to investigate several unresolved areas. The question on whether log P is linearly or parabolically related to odor intensity for a specific medium needs to be resolved. Six equations in Tables 1-V linearly related log P to odor intensity, while five parabolic relationships were observed which had an optimum hydrophobicity (log P) associated with maximum odor intensity. Log Po values observed were 3.17 and 2.90 for alcohols (threshold-aic). Alkanes had a log Po value of 5.33 (threshold-air). In aqueous media alcohols had a log Po value of 3.98 while 3-alkyl-2-methoxy pyrazines had a value of 2A3. The animal data indicates that rats had log Po values of 5.90 for acetates and 7.91 for alcohols. 

The equations in Table I indicate that for alcohols odor intensity is paraboliccilly dependent upon log P for threshold values determined in air (Eq. no. 11,19) and in water (Eq. no. 20) and linearly dependent upon log P for suprathreshold values in air (Eq. no. 1). The alcohol odor intensity also could be parabolically dependent upon log P for the suprathreshold values in... 

Another area of further study is the reproducibility and accuracy of derived predictive equations. Two different data sets of aliphatic aldehyde threshold values in water were subjected to QSAR techniques to determine whether log P can be used to accurately reproduce predictive equations for odor intensity data of a compound in a given medium determined by two different laboratories. Results in Table II indicate that equations 10 and 13 have slopes, intercepts, correlation coefficients and standard deviations which are not statistically different at the 95% level of confidence. Both data sets also produced equations giving poor correlations of E and log (1/c) which were not statistically significant. 

The use of the QSAR technique known as the Hansch Approach in the investigation of odor intensity and odorant physico-chemical properties has indicated that hydrophobic properties of homologous series of compounds, not steric or polar properties, are highly correlated to the level of odor intensity. This was shown to be the case for literature odor threshold and suprathreshold data determined at different laboratories using various media. The poor correlation between odor intensity and the steric properties of molecules (Taft Steric Constant) which had been reported earlier by this author (11) have been further verified by the use of Charton and Verloop Sterimol steric parameters. 

Odor Threshold. Most solvents have a characteristic odor. Human perceptibility and sensitivity to solvent vapors depends on habituation, which varies markedly from one person to another. Odors that are regarded as pleasant in small concentrations may be considered intolerable at high doses and under constant exposure. Other vapors that are initially considered objectionable may subsequently be regarded as tolerable [14.112]. It is therefore impossible to give objective rules for determining when an odor becomes objectionable. Odor intensity is subdivided into four levels ... 

Odor intensity ievei (odor threshold number) n. A test to determine the intensity of an odorant, or the number of dilutions required for an odorant (gas, vapor, or hquid) in order to become odorless or barely detectable as evaluated by a panel of humans sniffing samples. Also the character of the odorant sample may be evaluated (e.g., sweet, sour, ethereal, and putrid) (ASTM D1292-86, Standard Test Methods for Odor in Water and ASTM Publication DS 61, Atlas of Odor Character Profiles). 

Florhydral possesses a fresh floral and melon odor. It is reminiscent of the flavoring qualities found in convallaria majalis (commonly known as lily-of-the valley or muguet) and linden blossom. It has a threshold odor of 0.07 ngl and may be used to improve the floral character of an aroma composition by adding a watery or dewy quality. With its floral, fresh, trendy, and natural odor, it conveys together with other aroma compounds the perfume Good life woman (Davidoff, 1999), the typical fresh, marine, and ozonic touch. In citrus products it enhances the intensity with significant effects on the freshness. (-H)-Florhydral was found to be a much more powerful odorant than its (-)-enantiomer, which would encourage the further search for an improved asymmetric version of the hydroformylation [163]. 

---