Descriptive Aroma Changes in Selected Philippine Virgin Coconut Oil (VCO) during Storage at Elevated Temperatures

: Virgin coconut oil (VCO) is known to have functional properties. It is important to maintain its quality, such as its sensory properties, especially during storage. This study evaluated the effects of elevated temperature storage (i.e., 35 °C, 40 °C, and 45 °C) on the aroma of three differently processed (i.e., fermented, centrifuged, and expeller-pressed) VCO. Stored samples were evaluated by eight (8) trained panelists at various sampling days, based on a Q 10 of 2 for hydrolytic rancidity. Freshly prepared fermented and centrifuged VCO samples had predominantly acid and nutty aromas, respectively. Expeller-pressed VCO was perceived to have latik (aroma associated with cooked sweet, coagulated coconut milk) notes. Changes in the distinguishing aroma characteristic of each VCO sample exhibited polynomial behavior during storage in all elevated temperatures. The results imply that, during the initial stages of storage, aroma perception increased, followed by a decline, which may be due to the volatilization of the aroma compounds. Further, the rancid aroma intensity of samples surprisingly decreased, except for expeller-pressed VCO stored at 35 °C. This observation may also be attributed to the volatilization of the free fatty acids generated during storage. This should be investigated further, as this has an implication on the storage requirements of VCO.


Introduction
The global demand for virgin coconut oil (VCO) is expected to grow by roughly 2.5% over the next five years, as the commodity established a market niche of its own as a functional food [1]. The Philippines is one of the major exporters of VCO-a premium export commodity of the country-which allowed the growth of the local VCO industry, as it was proven profitable and promising as a low investment microenterprise with high market demand [2]. Increasing numbers of health-conscious consumers boosted the growth of VCO market due to its various health benefits and vast usage in manufacturing consumable and medical products to maintain good health and lifestyle [3]. Further, ongoing clinical study in the Philippines shows the potential of VCO to lower coronavirus load by 60-90% in mild cases [4]. The rising global demand for this commodity challenges the need to maintain quality parameters, such as sensory properties, especially during storage of VCO.
Virgin coconut oil is obtained from fresh, mature coconut kernel processed by mechanical or natural means, with or without the application of heat, while maintaining the natural state of the oil. It can be processed through fermentation, centrifuged, expeller-  pressing, etc. [5]. The quality of VCO is generally determined by physicochemical properties [6], however, understanding the descriptive quality of VCO in terms of its sensory properties is necessary to better distinguish it from refined coconut oil and differentiate its various processes. Sensory properties of VCO reflect the type of process the oil is produced with, as the production retains its natural state [7]. Moreover, sensory profiles of VCO, specifically aroma, are mostly attributed to the volatile organic compounds present in the oil which can be affected by external factors, such as variation of storage conditions, especially at the consumers' end. For VCO, storage at high temperatures may affect the stability of volatile compounds present in the headspace of the oil, which may cause a higher threshold of undesirable smell [8]. Sensory properties, such as aroma profiles, may help determine the stability and acceptability of the oil; hence, their need to be maintained.
Elevated temperature storage is commonly used in accelerated shelf-life testing (ASLT) to hasten the rate of deterioration of a product without modifying the order of changes usually seen in a product under normal storage conditions [9]. The chosen commodity is subjected to kinetic modeling with an appropriate time and temperature combination. With oil products, results of ASL may determine how oxidation changes in rates and product distributions under the catalyzed conditions (how oils or foods handle test stresses), providing suitable specifications for storage conditions for oils such as VCO [10].
With the growing VCO industry, there is a need to assess and maintain the export quality of VCO under varying storage conditions. There is a scarcity of reports on the evaluation of VCO aroma at elevated storage temperatures. Thus, this study determined the changes in the descriptive aroma profile of three (3) differently processed VCO, stored under elevated temperatures.

Ethical Consideration
Ethical clearance (NEC Code: 2019-001-Villarino-VCO) was approved by the National Ethics Board of Department of Science and Technology-Philippine Council for Health Research and Development.

Experimental Design
A randomized complete block design was used in the study. A total of three (3) samples with two (2) replicates were evaluated by the panelists.

Panelists
A total of 8 trained panelists (5 females and 3 males), aged 24 to 56 years old, participated in the tests. The panelists were recruited and selected from a three-phase screening process, which included a taste recognition test, an odor recognition test, and an intensity ranking test. The final list of panelists was randomly selected from the pool of participants who obtained a total score of 80% from the screening process.

Training
The panelists were trained, and samples were evaluated using the generic descriptive method [11], a method of quantitative descriptive method (Tragon Corp., Redwood City, CA, USA) and a Spectrum™ Analysis Method (Sensory Spectrum, Inc., Chatam, NJ, USA), which was also employed in the descriptive sensory analysis of VCO and RBD oil [5]. A total of 30 h, including the orientation of panelists, was spent for training 2 replicates of the sample.
In the first session of training, each panelist received VCO samples and validated perceivable aroma profiles based on the established VCO profiles [5]. The panel decided whether the generated descriptors and the existing list of VCO descriptive attribute were redundant and should be refined, or if there were terms that should be added. The list of terms was finalized, and the panelists defined each descriptor (Table 1). Panelists also identified possible reference standards from which the rating of the generated descriptor would be based on. Succeeding training sessions (i.e., 15 training sessions) were performed to refine reference standards, techniques for evaluation, and calibration of the panel.

Descriptor
Definition Nutty The aroma associated with the 2nd layer of fresh coconut kernel with testa Latik The aroma associated with cooked, sweet, coagulated coconut milk Acid The aroma associated with acetic acid solution Rancid The aroma associated with old, unpleasant, soapy, acrid oil

Sample Evaluation
A total of 7 mL sample was presented in 30 mL-capacity glass containers with screw caps, coded with 3-digit random numbers, and maintained at room temperature (30 ± 2 °C). Samples consisting of 6 VCO samples were presented to the panelists in a balanced random and monadic order. Figure 1 shows three randomly coded VCO samples in glass bottles. The samples were presented in a clean tray lined with white bond paper, together with reference standards for aroma, and a tablet device for the developed web application for the answer sheet of the panelists.
Usage of scented products (i.e., perfume, hand sanitizer, rubbing alcohol, etc.) was restricted prior to the evaluation session, and the panel was instructed to use the provided unscented cleansing soap (Cetaphil, Dallas, TX, USA) to wash hands to remove unnecessary scents. The aroma of the oil samples was evaluated by smelling the back of the palm first, to clear their nose. Then, the panelists proceeded to swirl the sample bottle five times clockwise and five times counterclockwise, making sure that oil samples reached the neck of the bottle. After which, panelists unscrewed the cap, tilted the bottle 45° towards noselevel, and performed three quick sniffs and evaluated the samples within five seconds. •

Elevated Temperature Storage
The 3 differently processed VCO samples (F, C, E) were stored in 3 varying temperatures with 6 different sampling points per temperature, based on a Q10 of 2.0 for lipid oxidation [12]. These varying temperatures were used to determine possible changes in the sensory properties of VCO, specifically the aroma profile of the samples. Table 2 presents the sampling point matrix for the elevated storage temperature of VCO.

Aroma Profile
Four (4) descriptors were identified by the panel which described the aroma of the VCO samples. Table 3 provides the list of descriptors with reference standards and corresponding intensity ratings used in the descriptive evaluation of the aroma of VCO samples. Figure 2 shows the standard references for each descriptor. VCO produced via centrifugation (F) was initially described to have a predominantly nutty aroma. Latik was the major aroma descriptor for expeller VCO (E), while an acid aroma was prevalent in samples produced through fermentation (F). A rancid aroma was also apparent in stored samples.

Sample Evaluation
As presented previously, the VCO samples initially had different predominant aromas, and developed a rancid aroma as storage progressed. Results are presented based on the prevalent aroma, with the development of a rancid aroma.

•
Elevated storage temperature: 35 °C The nutty aroma of sample C, the latik perception of sample E, and the acid aroma of sample F displayed polynomial behaviors during storage (Figure 3a-c). All three samples were observed to have a decreasing rancid aroma by the end of the storage period (day 210) compared to the initial evaluation. However, sample E obtained a higher rancid aroma on the last sampling day compared to the initial evaluation.  (Figure 4a,c). On the other hand, the latik aroma of sample E illustrated a decreasing perception, with a slight increase towards the end of the storage period (Figure 4b). Similarly, the rancid aroma of VCO samples was observed to be decreasing for all samples, which slightly rose at the end of storage (Figure 4d).

Elevated Storage Temperature: 35 °C
The decreased perception of both nutty (in C) and latik (in E) aromas can be attributed to the volatilization of the volatile organic compounds (VOCs) present in the oil samples, specifically ethyl acetate and 2-heptanone, respectively. Meanwhile, the increased perception of acid aroma in F samples can be attributed to the detection of acetic acid, which is produced during fermentation by endogenous microflora in coconut [8,13], and is known for its pungent smell which imparts undesirable odor to VCO [8].
Moreover, the increased perception of rancid aroma in the final sampling point of E may be due to the "unmasking" of the predominant latik aroma, which allowed the perception of the natural coconut aroma (i.e., nutty) 2-heptanone, which is responsible for the nutty aroma, together with ethyl acetate, and may have resulted the rancid aroma of E, as this volatile organic compound can be perceived as off odor on a higher threshold [8].

Elevated Storage Temperature: 40 °C
VCO produced through fermentation, F, displayed an increasing acid perception over the storage period. As discussed previously, an increasing perception of acid aroma can be attributed to the higher threshold of acetic acid produced during fermentation. Meanwhile, latik aroma displayed a decreasing intensity against sampling days. This may be attributed to the volatilization of highly volatile lactone responsible for the latik aroma [14].
The decreasing perception of a rancid aroma in VCO samples may be attributed to the inactivation of microorganisms at the 40 °C responsible for the hydrolysis of free fatty acids, which deplete at temperatures above 37 °C [15], as well as the volatilization of free fatty acids responsible for the rancid aroma profiles of VCO samples.

Elevated Storage Temperature: 45 °C
Similar to observations from samples stored under 40 °C, the increased perception of nutty aroma in sample C can be attributed to the high threshold of ethyl acetate and 2heptanone-volatile organic compounds responsible for the nutty aroma of VCO [8]. As mentioned, volatile organic compounds responsible for a latik aroma are easily volatized in the headspace of the VCO samples [8]. This observation can be attributed to the high storage temperature, which can cause depletion of microbial activity responsible for the hydrolysis of free fatty acids.

Conclusions
The changes in the aroma of the VCO samples stored at varying elevated temperatures exhibited polynomial behavior during storage. Results indicate that, during the initial stages of storage, aroma perception increased, followed by a decline, which may be due to the volatilization of the volatile organic compounds responsible for the aroma perceived in VCO samples. Further, a rancid aroma intensity of samples surprisingly decreased, except for expeller-pressed VCO stored at 35 °C, which can be due to the volatilization of free fatty acids responsible for the rancid aroma. This observation may also be attributed to the depletion of microbial activity which, at higher temperatures, is responsible for hydrolysis of free fatty acids which can result in the detection of a rancid aroma. This should be investigated further, as this has an implication for the storage requirements of VCO.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Data available on request due to restrictions (e.g., privacy or ethical).