556 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The bag method can also be used for sensory evaluation of the odors of tampons, napkins, and cotton pads worn under axillae, etc. Such sample col- lectors are placed into the bag, the bag is inflated with an odorless air, and left to equilibrate for 10 to 30 min. Use of this technique with samples ob- tained by polymer pads ( Fig. 2B) has not yet been explored. For sensory evaluation of bag samples, bags are placed in a cylinder such as the one in Fig. 2D, and the sample is pushed out by pumping air into the cylinder at a controlled rate. SENSORY EVALUTIONS Odor measurement is a measttrement of sensation and is conducted by obtaining the responses of panelists. Odors exhibit several sensory dimen- sions (9-11) as follows: 1. odor intensity, of which odor threshold is a sub- sidiary measure, indicating the dilution of sample needed to make the odor sensation so weak that either its detectability or its recognizability is at sta- tistical threshold value 2. •odor character or quality and 3. hedo•fic value, that is the pleasantness-unpleasantness characteristic. In the cosmetic correction of undesirable odors, detection threshold mea- surements have limited utility. If a fragrance is present, a reduction of mal- odorant emission may not be noticeable, since the threshold will relate to. the odor of fragrance. Recognition threshold-that at which the malodorant character becomes recognizable as the degree of sample dilution is decreased, is a more suitable characteristic, but it is better incorporated as a part of the odor character evaluation. The odor detection threshold is, however, a useful characteristic when dealing either with the malodor or with the frag- rance alone. Odor Intensity and Threshold Although the Weber-Fechner law is frequently utilized to express odor in- tensity as a function of the odorant concentration (9), the recent trend is to use another mathematical function, known as the psychophysical power law* (12-14) I ---- kC n or logI = logk + n log C *It is interesting to note that at the New York Academy of Sciences 1973 conference on odors (15) not a single paper dealt with Weber's law, but many utilized the power law.
EVALUATION OF HUMAN BODY ODOR 557 In the above, I is the psychophysical sensory magnitude of the odor sensa- tion C is the concentration of the odorant in air, which reaches the nose of the panelist and K and n are coefficients. For odors, n is 0.2 to 0.8, depending on the odorant. The consequence of this relation is that, e.g., for n -- 0.5, the odorant con- centration must be reduced by a factor of 4 to make the odor weaker by a factor of 2. For most odorants, a plot of log I versus log C is indeed a straight line. Until recently, odor intensities have been recorded using some form of cate- gory scale, e.g., from 0 (no odor) to 5 (extremely strong odor) (16). Such a scale is already in proportional relation to log I and log C. Odor intensity, which is expressed in some numbers proportional to the intensity of sensa- tion, increases by a factor of 3 to 4 per category step for the 0 to 5 scale. The category scale creates difficulties in standardizing the meaning of the scale values. Therefore, more recently (17) an effort has been underway at the American Society of Testing Materials (ASTME) to adapt an odor refer- encing method in which odors are compared to odors on a scale consisting of a series of concentrations of 1-butanol. The odor of the sample is then documented to have an intensity which matches some certain concentrations of 1-butanol. Panelists readily do this, disregarding the differences in the odor character of the sample and 1-butanol. Figure 3 depicts one physical form of such scale there are 8 scale stimuli, each next higher in 1-butanol concentration than the preceeding by a factor of 2 (binary scale). Still more recently (18) a proposal has been made to standardize the 1- butanol-referenced scale even further. Odor intensity of 250 ppm v/v in air is defined as I = 10 this odor is well noticeable in its intensity, but not too strong. Since for 1-butanol n -- 0.66 (18), the following relation results I = 0.261 (ppm 1-butanol) o.66 The resulting numbers, for different samples, are in an approximate ratio of the sensory odor intensities of these samples. As an example (which will be used later) isovaleric acid (component of per- spiration odors) at a concentration of 2.8 x 10 -7 g/1. to air, was evaluated versus 1-butanol scale by 9 panelists, and its odor found to be as intense as that of 89 ppm v/v of 1-butanol in air. This corresponds, and it can be docu- mented, as the sensory intensity of 5. The above scale has a fixed middle point and is open on both ends. Odors of I below unity approach the odor threshold intensity. The effect of different values of exponent n for different odorants results in a different rate of odor intensity increase with odorant concentration. There- fore, measuring odor intensities in terms of multiples of odor threshold does not give correct comparisons of the odor intensities of the sample before dilu- tion. This is best clarified by considering the odor intensity of perspiration in the axillar area of a shirt.
Previous Page Next Page