Sensory Testing of Australian Fragrant, Imported Fragrant, and Non-fragrant Rice Aroma

Abstract

A large consumer panel of mixed age, genders, and cultural background participated in a series of triangle tests involving the aroma of rice samples obtained from retail outlets. The group could, as a whole, distinguish between Australian fragrant (AF) and non-fragrant (NF) rice and between AF and imported fragrant (IF) rices based on aroma. The ability to discriminate between the samples varied with age, gender, and cultural background. In addition descriptions of the aromas by participants implied that the aroma of AF rice was preferable to that of both the IF and Australian NF rice. The NF rice contained approximately twice the amounts of 2-hexenal[E], nonanal, 2-pentylfuran, and 2-octenal[E], and significantly more 2-nonenal[E] and hexanal, compared to the fragrant rices. 2-Acetyl-1-pyrroline (2AP) was present in both fragrant rices but not in the NF sample. The AF rice contained twice the amount of 2-heptenal[E] and three times the amount of 2-decenal[E] and 2,4-decadienal[E,E] compared to the IF rice. The concentrations of significant aroma volatiles were linked to aroma descriptions of the rices.

Keywords:

• Fragrant rice
• Aroma volatiles
• GS/MS
• Sensory testing

Introduction

Fragrant rice is recognised and characterised by its particular aroma, which is an integral part of its overall sensory properties. There are two major types of fragrant rice: Jasmine which is of Thai origin and Basmati which is of Indian origin. At present Australia produces only a Jasmine type commercially. The Australian rice industry produced 1.4 million tonnes of rice (rough) in 1999 compared to1.2 million tonnes in 1997.1 Of this, about 5% is locally grown fragrant rice of the Jasmine type which is largely consumed within Australia. Australian rice consumption increased from 6.7 kg cap⁻¹ year⁻¹ in 1990 to 9.3 kg cap⁻¹ year⁻¹ in 1998.1

Fragrant rice is predominantly consumed in countries of origin, including Basmati in India and Jasmine in Thailand. Western markets tend to consume non-fragrant (NF) rice but fragrant types are becoming more generally recognised and accepted. There is a possibility of exporting fragrant rices to such traditional markets. Also an increase of Asian communities within western countries has resulted in a growing awareness of them in western markets. A consequence of this is the breeding of fragrant rice cultivars to suit agronomic conditions in countries such as America and Australia. Fragrant rices from traditional producers are seen as the benchmark of aroma quality. Therefore the aroma of fragrant rice developed in western countries needs to be comparable to that of traditional varieties.

For successful development of fragrant rices, suitable for commercial production in non-traditional regions, consumer ability to distinguish between fragrant and NF rice, and between newly developed and traditional varieties, and consumer preferences, will dictate market acceptability of any new varieties. In addition, understanding human perception of specific rice aroma constituents is essential since the characteristic aroma of fragrant rice needs to be readily perceivable, characteristic of fragrant type, distinguishable from that of NF rice and acceptable to consumers.

This study aimed to determine if consumers, of various ages, gender and cultural background could detect differences in the aroma of Australian fragrant (AF) rice and NF rice, and between AF and imported fragrant (IF) rice. In addition through consumers' descriptions of rice aroma, preferences for the rices tested will be established. Results from sensory evaluation were compared with levels of significant volatile aroma compounds in the rice samples and links between perceived odor characteristics and aroma compounds investigated. In order to achieve this, we exploited an opportunity provided by a large group of visitors at a Science Expo in Sydney. This group gave us a consumer panel whose range of age, cultural background, and gender provided a cross-section representative of contemporary Australian consumers. It was also deemed especially important that retail samples of rice were used for this investigation in order to represent the way in which consumers would normally obtain the rices for domestic consumption.

Materials and Methods

The rice samples used in the study were all commercial rice varieties purchased from a local supermarket. Subjects participated in two tasks, each involving a triangle test procedure. Task 1 involved comparing AF Koala brand Jasmine and NF Sunrice brand rice. Task 2 involved comparing AF Koala brand Jasmine and IF Kumarnthong brand Jasmine rice. The odd sample for the two triangle tests was varied throughout testing.

Cooking was standardised by using rice cookers. A 1–1.5 ratio of rice/water was used with an approximate cooking time of 16 min. The warm cooked rice was filled into 300 mL sniffing bottles to 1/4 of the volume at a temperature of 70°C. Sniffing bottles containing the rice samples were labeled with a three digit code. Fresh samples were presented to the subjects who were required to sniff the samples and to select the odd sample.

The sensory trial was run at an annual Science Expo in Sydney, which attracted students from the age of 6–18, teachers, parents, and other members of the public. This venue was perceived to be an opportunity to utilise a large consumer panel, consisting of various ages, gender, and cultural backgrounds. The sensory test was performed in a controlled manner such that one subject at a time performed the test. A total of 254 subjects completed task 1 and 226 completed task 2. Most subject's (218) completed both tasks. Subjects were required to complete a questionnaire that involved stating their age, gender, and cultural background. The two triangle tests followed the Australian standard for triangle tests.2 Minimum numbers of correct responses to establish significance were determined according to3 to determine the minimum number of correct judgements to establish significance. Chi-square analyses were performed on data relating to age, gender, and cultural background. The volatile aroma compounds of the rice were determined by simultaneous distillation extraction, following the method of Wilkie et al.4

Results and Discussion

Overall Panel Responses

For the 254 subjects participating in the triangle test comparing AF and NF rice aroma, the minimum number of correct judgements required to establish significant differentiation between them was 103. A total of 139 subjects recorded a correct result, establishing that the consumer group as a whole found a significant difference (p<0.05) between the aroma of AF and NF samples. For the 226 subjects participating in the second task involving AF and IF rice aroma, the minimum number of correct judgements required to establish significant differentiation between them was 92. A total of 115 recorded a correct result, establishing that the consumer group as whole found a significant difference (p<0.05) between the aromas of the two fragrant rices.

The relative ability of subjects to distinguish between samples in tasks 1 and 2 was also examined. A total of 218 subjects undertook both tasks with 31% identifying the odd sample in both tasks compared to 25% incorrectly identifying the odd sample in both tasks. The remaining subjects (44%) identified only one odd sample correctly out of the two tests. Of these subjects, more subjects (54%) were able to distinguish between AF and NF rice aroma (task 1) than between AF and IF rice aroma (46%, task 2). The results suggest that this consumer group found it easier to distinguish between the aroma of NF and fragrant rice than between the two different types of fragrant rices.

A perceivable difference in the aromas of AF and IF rice by local consumers is also important for the development of AF rice. This implies that the aroma of AF rice is either more or less preferred than the imported fragrant rice. Overall, subjects preferred AF rice and this aspect is discussed later in this paper.

The human nose has a theoretical odor detection limit of about 10⁻¹⁹ moles, making it a valuable and sensitive tool for the detection of odor active volatiles.5 Therefore it is not surprising that subjects can distinguish between the aroma of fragrant and NF rice and between the aroma of the fragrant rices. However, since a large proportion of subjects did not correctly identify the odd samples in the triangle tests, it appears that there are large variations in odor sensitivities between individuals. At least part of the reason for this is that some of the population are born smell blind or anosmic and cannot smell several types of compounds,6 while odor perception itself varies between individuals.

Effects of Age, Gender, and Cultural Background

Differences in the ability of subjects to distinguish between the aroma of fragrant and NF rice and between the aroma of AF and IF rice based on age, gender, and cultural background were investigated. Cultural backgrounds were categorised into Australian (135 and 127 subjects for tasks 1 and 2, respectively), non-Australian (105 and 83) and Asian (49 and 40) subjects. Non-Australians were those from cultural backgrounds other than Australian and a further subset of these were “Asian” subjects, who were from Thailand, Taiwan, Burma, Korea, China, Sri Lanka, India, Malaysia, Hong Kong, Japan, Singapore, and Vietnam. These were categorised separately because of the greater likelihood that they would be familiar with fragrant rices. The ages were categorised into ranges of 6–12 years (168 and 148 subjects), 13–18 years (58 and 52) and 19+ years (10 and 12). The influence of gender was also investigated with male (79 and 67) and female (163 and 147 subjects) participating. The majority of subjects who participated in the sensory trials were children aged between 6 and 12 years. This was not surprising since the Science Expo, where the sensory test was performed, was targeted at school students. However the use of children subjects was seen as acceptable since they readily understood what they were required to do to perform the tasks of the sensory test and children represent a major sector of rice consumers. Sensory studies that have included children suggest that there was no difference in olfactory acuity between age groups ranging from 6 to 94 years.7 Similarly a study involving sensory perception of orange beverages reported no significant differences in the abilities of taste perception of orange flavor between the age groups of 6 and 65 years.8 In addition children have a significant influence on food purchase decisions in the home so their sensory perceptions of foods is of great interest.9 Therefore there should be no concerns about the validity of utilising children for olfactory sensory studies.

Chi-square statistics indicated that Australians were significantly better (χ ²=9.17, p<0.05), at perceiving differences between the aromas from the AF and NF rices than non-Australians. When Australian subjects were compared to Asian subjects there was no significant difference between the groups. It was expected that Asians, being more experienced with fragrant rice, would be better at distinguishing between the rices. However, Australian society is culturally diverse and this is reflected in a wider variety of cuisines, leading to widespread consumption of traditional Asian dishes containing fragrant rices.

Females were significantly better, (χ ²=9.17, p<0.05) at distinguishing between AF and NF rice than males, as were subjects between 13–18 years of age compared to the other age groups (χ ²=11.47, p<0.05). When comparing AF and IF rice aroma data, subjects 19 years and over were able to distinguish these better than the other, younger, age groups (χ ²=54.3, p<0.05).

Australia's cultural diversity offers the opportunity to assess acceptability of locally developed fragrant rice varieties in-country to suitable consumer groups although it must be recognised that the tastes of expatriate consumers may change and it is advisable to test the product in the country of interest for export.10

Volatile Components of the Rice Samples

Figures 1, 2, and 3 present the chromatograms of volatile components isolated from AF, IF, and NF rice respectively. Each variety has an individual pattern of volatile aroma compounds. The two chromatograms of fragrant rice (Fig. 1 and 2) are similar and that of NF rice (Fig. 3) is different in several respects.

Figure 1. Chromatogram of AF rice (Koala brand).

Figure 2. Chromatogram of IF rice (Kumarnthong brand).

Figure 3. Chromatogram of Australian NF rice (Sunrice brand).

In this study, in excess of 100 volatile aroma compounds were identified in each of the rice samples. Table 1 lists the selected aroma compounds, their odor descriptions, odor thresholds, concentrations, and odor units. However, these results are only for compounds previously identified in the literature as being significant aroma compounds in fragrant and NF rices.11 12 13 14 15 16 17

Table 1 Concentration, T odor unit, and odor description of significant volatile aroma compounds in AF, IF, and NF rice

			Rice grain samples
			Imported fragrant		Australian fragrant		Non-fragrant
Significant aroma compounds11 12 13 14 15 16 17	T 11 (ppb)	Odor13 14 15 16 17 18	Concentration (ppb)	Odor unita	Concentration (ppb)	Odor unit	Concentration (ppb)	Odor unit
Hexanal	5	Green, grass	854	171	998	199.6	1,507	301.4
Butanol	500	Medicinal	5.3	0.01	4	0.008	9.4	0.02
2-Heptanone	140	Fruit, spicy	23.2	0.17	24.7	0.18	39.5	0.3
Heptanal	3	Fruity, fatty	24.7	8.2	20.4	6.8	25.8	8.6
2-Hexenal [E]	17	Green, fruity	7.0	0.4	7.4	0.4	15.2	0.9
2-Pentyl furan	—	Nutty, beany	34.8	—	47.1	—	77.9	—
1-Pentanol	4,000	Sweet, strong	83.7	0.02	102.3	0.03	103.7	0.03
Octanal	0.7	Citrus, fatty	26.1	37.3	23.9	34.1	29.0	41.4
2-Heptenal [E]	13	Fatty, green	45.4	3.5	92.8	7.1	80.2	6.2
2-Acetyl-1-pyrroline	0.1	Sweet, popcorn	49.0	490	33.8	338	0	—
6-Methyl-5-hepten-2-one	50	Herby, green	10.8	0.2	6.8	0.14	3.3	0.07
1-Hexanol	2,500	Sweet, green	50.7	0.02	76.6	0.03	58.9	0.02
Nonanal	1	Floral, fatty	28.2	28.2	28.5	28.5	41.9	41.9
2-Octenal [E]	3	Green, herby	47.4	15.8	59.0	19.7	94.4	31.5
1-Octen-3-ol	1	Herby, earthy	33.6	33.6	39.9	39.9	57.5	51.5
2-Ethyl hexanol	—	Oily, sweet	0	—	0	—	44.4	—
Benzaldehyde	350	Nutty, sweet	35.6	0.1	22.7	0.06	48.5	0.14
2-Nonenal [E]	0.08	Fatty, waxy	6.4	80	5.9	73.8	28.0	350
2-Decenal [E]	0.4	Sl.fatty, green	12.4	31.1	38.0	95.0	15.3	38.3
2,4-Decadienal[E,E]	0.07	Fatty, citrus, powerful	13.4	191	36.3	518.6	30.6	437.1
2-Phenyl ethyl alcohol	1,100	Sweet, floral	5.2	0.005	3.3	0.003	0	—
4-Vinylguaicol	3	Spicy, fruity	14.9	5.0	13.8	4.6	42.4	14.1
4-Vinylphenol	10	Unpleasant	3.5	0.4	4.1	0.4	4.5	0.5
Indole	140	Faecal, floral	12.3	0.09	2.2	0.02	16.9	0.12

^aOdor unit (U_o) calculated by concentration/T.

In general foods contain many volatile compounds, only a few of which are of sensory significance.5 The characteristic aroma of fragrant rice is attributed mainly to the volatile aroma compound, 2-acetyl-1-pyrroline (2AP). The concentration of 2AP was previously reported to vary from 40–90 ppb in fragrant rice varieties whereas only 8 ppb or less was found in NF rice.19

The significance of a volatile compound to the total aroma is determined by calculation of the odor unit (U_o) for that compound. The odor unit is calculated by dividing the concentration (C) of the compound in food by the odor threshold (T ) of the compound in water (U_o=C/T ). Assuming that T is the same for the food, a compound is considered to have a significant contribution to the total odor when U_o is greater than one. For example,16 determined the odor unit of 2AP in a fragrant rice to be 250 compared to 0 in a NF rice. From Table 1, compounds expected to contribute to the aroma of both fragrant and NF rice included, hexanal, heptanal, octanal, 2-heptenal[E], nonanal, 2-octenal[E], 1-octen-3-ol, 2-nonenal[E], 2-decenal[E], 2,4-decadienal[E,E], and 4-vinylguaicol.

2-Acetyl-1-pyrroline was identified as a significant aroma compound in fragrant rice but not in NF rice. The compound, 4-vinyl phenol, was found to be a significant aroma volatile in NF rice only and 2-ethyl hexanol was present only in NF rice. Therefore these two compounds would contribute to the characteristic aroma of the NF rice. In addition to these major differences, NF rice also contained approximately twice the amount of 2-hexenal[E], nonanal, 2-pentylfuran, and 2-octenal[E] and significantly more 2-nonenal[E] and hexanal than the fragrant rices.

The AF and IF rice had the same significant volatile aroma compounds. However the concentrations of these compounds differed between them. It is these subtle differences that give the characteristic aromas to each of these rices. The AF rice was found to contain twice the amount of 2-heptenal[E] and three times the amount of 2-decenal[E] and 2,4-decadienal[E,E] compared to the IF rice. In addition the IF rice contained more indole than the AF rice.

Correlation of Sensory Descriptions with Individual Aroma Compounds

Subjects were asked to comment on how the aroma of the chosen odd sample in the triangle test was different. Only comments from subjects who identified the odd samples correctly were analysed. The comments were categorized as positive or negative. Examples of positive comments included, sweeter, popcorn, nicer, hot chips, better, did not stink, less stinky, and cornier. Examples of negative comments included, eggy, off smell, unlike rice, musty, funny, worse, disgusting, unpleasant, bad smell, rotten eggs, revolting, rancid, more pungent, sour, and sulfur. Chi-square statistics were performed on the proportions of negative and positive comments for the different rices to determine if there was a significant positive or negative correlation between the aromas and the subjects' aroma preferences. For the AF vs. NF and AF vs. IF rices it was determined that there was a significant difference between the proportion of negative and positive comments, (χ ²=24.4 and 23.1, p<0.05 respectively). It was considered that the AF rice aroma was preferred over both NF and IF rice aroma.

The sensory findings on aroma differences and descriptions were compared with the concentrations of significant volatile aroma compounds in the rice samples (Table 1). The concentration of the volatile aroma compounds, including, 2-heptenal[E], 2-decenal[E], and 2,4-decadienal[E,E] were much greater in the AF rice compared to the IF rice. The common odors contributed by these compounds include, green and fatty odors. Therefore comments such as nicer, better and fresher, for the AF rice, may be attributed to the higher concentrations of these compounds compared to the IF rice. The IF rice contained extensively more indole compared to the AF rice. Indole contributes faecal, putrid, musty, unpleasant odors but exhibits floral odors at low concentrations. Therefore aroma descriptions of the imported rice such as, off smell, rotten, sulfur, disgusting, and pungent may be due to the high indole levels.

The most significant difference in volatile components between fragrant and NF rice was the concentration of 2AP. Its contribution to the aroma of AF and IF rice included odors that were sweeter, cornier, and popcorn-like. Another notable difference between fragrant and NF rice was that NF rice contained 2-ethylhexanol while the fragrant rice did not. This compound contributes oily, sweet, and slightly rosy odors. In addition NF rice also contained significantly more 2-hexenal[E], nonanal, 2-pentylfuran, 2-octenal[E], 2-nonenal[E], and hexanal compared to the fragrant rices. These volatile compounds contribute common odors such as green, grassy, and fatty. Therefore the presence of these compounds in NF rice may explain aroma descriptions like rancid and hot chips. The NF rice also had a significant number of negative odor descriptions such as unpleasant, bad smell, rotten eggs, unlike rice, and revolting. This may have been due to the high level of 4-vinyl guaicol that can contribute an “unpleasant” odor.

Conclusions

Consumers, as a group, could distinguish between the aroma of fragrant and NF rice and between that of AF and IF rice. The sensory trial indicated a preference for the AF rice over the other two. Females were better than males, and Australians were better than non-Australians, at distinguishing between the aroma of fragrant and NF rices were. These effects were similar, but less pronounced, for aroma differentiation between the two fragrant rices. Subjects in the 13–18 year old category were better than younger and older age groups at distinguishing between AF and NF rice while those 19 years and older were better at distinguishing between AF and IF rice. Differences in consumer perception between the three rice samples could be explained in the context of differences in their volatile components.