JOURNAL OF COSMETIC SCIENCE 528 heated to 40°C, and agitated for 5 min at 200 rpm with a helix agitator. Then, pepper- mint oil was added to the emulsions. In order to study the emulsions’ stability, centrifugation was used. After being centri- fuged at 3000 rpm for 15 min, no phase separation was observed in the emulsions, even after 6 months of storage at 25°C. SENSORY PANEL A sensory panel consisting of six trained assessors with previous experience in the evalu- ation of cosmetic products evaluated the freshness of the refreshing creams. Before the test, a mixture of isopropyl alcohol and water (45:55) was applied to the forearms of the asses- sors in order to clean the zone, and it was allowed to dry for 5 min. Then, 4-cm-diameter circles were drawn on the internal side of the non-dominant forearms of the assessors. Assessors were asked to place a controlled preweighted amount of the emulsion (0.05 g) in the center of one of the circles. The assessors rubbed the sample with their fi ngers within the circle in a circular way 30 times at a rate of two times per second. Freshness was evaluated 5 and 10 min after application using a structured 11-point scale anchored by 0 (no freshness) and 10 (burning sensation). In this scale, a rating of 8 corresponded to the description “very fresh.” Samples were presented using white polyethylene containers labeled with three-digit random numbers. Two sessions were held over two consecutive days for duplicate evalua- tion of the samples. In the two sessions, the four samples were presented following a randomized balanced rotation (multiple orthogonal Latin square) in monadic sequence. The tests were carried out in a sensory laboratory designed in accordance with ISO 8589 (24). The evaluations were performed under artifi cial daylight-type illumination, tem- perature control (between 22° and 24°C), and air circulation. CONSUMER STUDIES Two studies were carried out in two different cities: one in Buenos Aires (Argentina), and the other in Montevideo (Uruguay). In each city fi fty female consumers of refreshing creams, ages ranging from 20 to 60, evaluated the four samples. A preweighted amount of each emulsion (0.3 g) was presented to consumers in plastic containers. Samples were presented following a randomized balanced rotation in monadic sequence, as in the trained assessors’ panel evaluations. In this case, consumers evaluated each sample only once. First, consumers were asked to smell the samples and to score the freshness sensation of each and their odor liking, using a structured ten-point scale (1–10). Then, consumers applied the samples on their forearms, and 5 min after application they were asked to evaluate their freshness sensation and their liking using a structured ten-point scale. DATA ANALYSIS In order to establish differences between the freshness sensation of the evaluated creams, an analysis of variance (ANOVA) was carried out on trained assessors’ data, consider- ing type of peppermint oil, emulsion formulation, time, assessor, repetition, and the

FRESHNESS EVALUATION OF REFRESHING CREAMS 529 fi rst-order interactions “type of peppermint oil × emulsion formulation,” “time × type of peppermint oil,” and “time × emulsion formulation” as fi xed sources of variation. Differ- ences were considered signifi cant at p 0.05. When differences were signifi cant, honestly signifi cant differences were calculated using Tukey’s test. An ANOVA was also carried out on consumers’ data, considering type of peppermint oil, emulsion formulation, city, and the interaction “type of peppermint oil × emulsion for- mulation” as fi xed sources of variation. Differences were considered signifi cant at p 0.05. When differences were signifi cant, honestly signifi cant differences were calculated using Tukey’s test. RESULTS AND DISCUSSION TRAINED ASSESSORS’ PANEL According to ANOVA, assessor and repetition did not have a signifi cant effect on fresh- ness scores (Table II), suggesting homogeneity and reproducibility of the trained asses- sors’ panel. Moreover, as shown in Table II, according to the trained assessors’ panel, type of pepper- mint oil, emulsion formulation, and time signifi cantly affected freshness sensation (p 0.05). Moreover, the interactions “time × emulsion formulation” and “time × type of pepper- mint oil” were signifi cant (p 0.05), indicating that the evolution of the freshness sensa- tion with time depended on the refreshing agent and the formulation of the emulsion that contained the refreshing agent. As shown in Table III, 5 min after application freshness sensation was signifi cantly higher when formulation B was considered. This could be explained by considering that penetra- tion of peppermint oil into the skin is necessary in order to reach the nerve terminals as- sociated with freshness sensation (8). Thus, the low proportion of apolar components in formulation B could have enhanced peppermint oil liberation and penetration into the skin, increasing freshness perception. On the other hand, the type of peppermint oil did not signifi cantly affect freshness per- ception after 5 min of application (Table III). Thus, menthol removal did not affect fresh- ness sensation at this evaluation instance, suggesting that other components different from Table II Results of the ANOVA Performed on Data from the Trained Assessors’ Panel Source of variation ANOVA results F p-value Assessor 1.40 0.2312 Repetition 0.02 0.8820 Emulsion formulation 272.92 0.001 Type of peppermint oil 8.00 0.0059 Time 20.35 0.001 Type of peppermint oil × emulsion formulation 67.01 0.001 Time × type of peppermint oil 30.32 0.001 Time × emulsion formulation 17.32 0.001