VOLATILE PROFILE AND SENSORY PROPERTY OF GARDENIA JASMINOIDES 375 Table III Sensory Evaluation of Each Gardenia Extracts in 43 Volunteers Sample Likert score Satisfaction Intensity Concrete Absolute ethanol 2.53 ± 0.15 2.79 ± 0.20 Petroleum ether 2.33 ± 0.16 4.28 ± 0.13 n-Hexane 2.23 ± 0.18 4.12 ± 0.13 Absolute de enfl eurage 3.72 ± 0.15 2.74 ± 0.13 n-hexane concrete. In the absolute de enfl eurage, farnesene was the most prominent, fol- lowed by Z-3-hexenyl tiglate, E-3,7-dimethyl-1,3,6-octatriene, Z-3-hexenyl benzoate, in- dole, and E-9-octadecene. These aroma extracts were thereafter submitted for preliminary evaluation for sensory response by a small group of volunteers. The preliminary preference tests of the gardenia fl ower oil extracts were conducted with 43 Thai volunteers using the Likert scale (Table III). All of the concretes received lesser approval from the volunteers because of their high intensities. Although the absolute ethanol concrete was best among the solvent-extracted fragrant oils with the intensity close to the absolute de enfl eurage, the preference was signifi cantly (p 0.05) lower. In contrast, the approval for the absolute de enfl eurage was obvious. The nuance of the aro- matic fl oral odor of the enfl eurage may come from other aroma compounds that were not detected in the concretes. Therefore, the odor similarity of the absolute de enfl eurage was compared with fresh gardenia in a larger group of 152 volunteers that included a variety of age groups (Table IV). Absolute de enfl eurage using spermaceti wax mixed with palm oil was evaluated for odor quality in terms of preference and difference compared with the fresh gardenia fl ower. Thai volunteers (male and female, 76 each) with the different age ranges (Table IV) were included in the study. Fragrant sample 4 (absolute de enfl eurage from palm oil) gained high preference in male and female volunteers, and this preference was consistent across all ages. While female volunteers liked sample 4 more (p = 0.01) than the male volun- teers, no signifi cant difference in preference was observed for different age ranges. The aroma extract was also compared with fresh gardenia odor, but volunteers insignifi cantly Table IV Odor Quality of Fardenia Aroma Extracts Demography Preference Difference Gender n Means p Means p Male 76 3.74 ± 0.07 0.01 3.12 ± 0.09 0.23 Female 76 4.01 ± 0.06 3.26 ± 0.08 Age (years) 16–25 38 3.87 ± 0.13 0.61 3.55 ± 0.09 0.00 26–35 37 4.00 ± 0.13 3.46 ± 0.11 36–45 39 3.79 ± 0.10 3.00 ± 0.12 45 38 3.84 ± 0.10 2.76 ± 0.11

JOURNAL OF COSMETIC SCIENCE 376 Figure 1. Psychological characters of (A) the gardenia absolute de enfl eurage and (B) optimized products incorporated with the aroma extract. (p 0.05) noticed the differences between the extract and the fresh fl ower. However, vol- unteer age clearly infl uenced the odor differences, with the younger volunteers rating higher scores than older volunteers. Volunteers were also asked about the psychological effects of the aroma extract prepara- tion. Relaxing and refreshing sensations were characteristic traits of the gardenia odor, followed by calming, as shown in Figure 1. The most notable personal care products that incorporate the gardenia aroma extract are body care and cleansing products, which is in accordance with the relaxing and refreshing characters of the extract. Surprisingly, only 62 volunteers recommended using this fragrant extract for perfume creation. Although the extract provides a relaxing sensation, massage oil was not strongly suggested. It is possible that the mild scents with fl oral note are less preferred than the top note citrus scents in massage oil (12).