Enhancing the Safety and Quality of Coconut Residue from Virgin Coconut Oil Wet Processing Through Thermal Pretreatment: A Preliminary Study ^†

Abstract

Coconut residue (CR) is a major by-product generated during the wet processing of virgin coconut oil (VCO). Despite its potential as a raw material for value-added products such as dietary fiber, it remains underutilized due to its perishable nature, highlighting the need for appropriate pretreatment to improve safety and quality prior to valorization. This study evaluated the effects of thermal pretreatments, namely pan-roasting at 65–70 °C, hot-air drying at 50 °C and 60 °C, and their combinations, on the microbiological and physicochemical properties of CR. Microbiological quality was assessed through aerobic plate count, yeast and mold count, and total coliform analysis, while physicochemical properties were evaluated using pH, titratable acidity (TA), and instrumental color measurements. Results showed that CR subjected to pan-roasting, either alone or followed by drying at 60 °C, maintained acceptable microbial counts and generally exhibited lower TA and higher pH compared to other treatments, suggesting improved stability and reduced acidity development. However, pan-roasting caused color changes as reflected by a significant reduction in lightness (L*) values relative to the control. Overall, pan-roasting could serve as a promising thermal pretreatment step to enhance microbiological safety and preserve the physicochemical quality of CR. While these results indicate its potential for preparing CR for dietary fiber valorization, confirmation through analysis of fiber content, techno-functional properties, and validation using CR from commercial VCO processing facilities is still required.

1. Introduction

The growing demand for virgin coconut oil (VCO) has resulted in an increased generation of coconut residue (CR), a solid by-product obtained after the mechanical pressing of coconut meat to extract coconut cream [1,2]. CR is known to be nutrient-dense, containing approximately 50.1–60.5% dietary fiber, 5.1–15.9% protein, and 16.3–29.2% oil on a dry basis [3,4]. It is considered a fiber-rich material, consisting primarily of insoluble dietary fibers, making it a promising raw material for value-added food applications [5]. Several studies have demonstrated the potential of CR for food product development. These include its partial substitution for wheat flour in bakery products [6], fortification of cassava starch in tapioca-based foods [7], and its use as a source of crude polysaccharides with prebiotic effects on selected probiotic strains [8]. Additionally, dietary fiber extracted from CR has been successfully incorporated as a prebiotic source in probiotic ice cream formulation [3]. Despite these demonstrated applications, the utilization of CR remains limited as it is often diverted to low-value use such as animal feed or is disposed of as waste [9].

A key constraint in the valorization of CR is its rapid microbial deterioration due to its high moisture content (50–55%), particularly when derived from the wet method of VCO processing [10]. This susceptibility to spoilage was evident in our initial collection of CR from a local VCO processing facility, where even samples that were immediately frozen after collection exhibited high microbial loads and acidic aroma, making them unsuitable for further processing. These observations highlight the need for an effective pretreatment method that can be applied immediately after sample collection to ensure microbiological safety and maintain the quality of CR, supporting its successful valorization. Thermal pretreatments like conventional hot-air drying have previously been applied to reduce the moisture content of fresh CR; however, this method has been reported to exhibit a low drying rate and cause significant whiteness loss [11]. Thus, there is a need for a simple and low-cost thermal pretreatment that can be applied immediately after sample collection to preserve quality with reduced processing time. Pan-roasting presents a viable alternative, as it allows direct heat transfer through contact with a hot surface and consequently shorter processing times compared to conventional hot-air drying [12]. Although pan-roasting has not yet been explored as a pretreatment method for CR, its successful application to other agro-industrial by-products, such as rice bran, to enhance stability after processing into flour [13] or before oil extraction [14] suggests its potential as a CR pretreatment.

Hence, this study explored the effects of pan-roasting and hot-air drying on the microbiological safety and physicochemical quality of CR as pretreatment strategies toward its value-added utilization. This work served as our preliminary experiment to establish appropriate pretreatment conditions prior to our large-scale collection of CR from a local VCO processing facility.

2. Materials and Methods

2.1. Coconut Residue Preparation

Freshly ground coconut meat (10.1 kg) was procured from a local vendor in Brgy. Krus na Ligas, Quezon City, Philippines. The grinder was thoroughly cleaned to ensure hygienic processing, and the sample was immediately transported to the laboratory. Coconut milk was extracted using a screw press, yielding 3.9 kg of coconut residue (CR) for further processing.

2.2. Coconut Residue Pretreatment

The collected CR was equally divided into six groups: five pretreatment conditions and one untreated control, as summarized in Table 1. For the pan-roasting treatment, CR samples were pan-roasted until the material temperature reached 65–70 °C for 10 min, with constant stirring to ensure uniform heating. Samples subjected to drying alone were dried in a hot-air incubator (IN260 Memmert, Büchenbach, Germany) with air temperature maintained at 50 °C or 60 °C and a sample layer thickness of 10 mm, until a final moisture content of <10% was reached. For the combination treatments, CR samples were first pan-roasted at 65–70 °C for 10 min with constant stirring, followed by hot-air drying at 50 °C or 60 °C. The temperature ranges for the pan-roasting and hot-air drying treatments were selected to minimize the risk of overheating or scorching the samples and to prevent thermal damage to the CR. All CR samples were stored frozen (−20 °C) until analysis.

Table 1. Pretreatment conditions for coconut residue.

Treatment Code	Pan-Roasting	Drying 1
Control	–	–
PR	65–70 °C for 10 min	–
PR-D50	65–70 °C for 10 min	Hot-air drying at 50 °C
PR-D60	65–70 °C for 10 min	Hot-air drying at 60 °C
D50	–	Hot-air drying at 50 °C
D60	–	Hot-air drying at 60 °C

¹ Drying was performed until a final moisture content of <10% was reached.

2.3. Microbial Analysis

Samples were analyzed for aerobic plate count (APC), yeasts and molds count (YMC), and total coliforms. A 25 g portion of each sample was aseptically homogenized with 225 mL of sterile peptone water (PW) (HiMedia, Thane, Maharashtra, India). For APC, serial dilutions (10⁻¹ to 10⁻⁴) were pour-plated on Plate Count Agar (TM Media, Bhiwadi, Rajasthan, India) and incubated at 35 °C for 24 h. For YMC, the same dilutions were pour-plated on Potato Dextrose Agar (TM Media, Bhiwadi, Rajasthan, India) acidified with 10% tartaric acid, followed by incubation at 28 °C for 5 days. Results were presented as CFU/g.

Total coliforms were determined by the MPN method, wherein 1 mL of 10⁻¹–10⁻³ dilutions was inoculated into three Lauryl Tryptose Broth (TM Media, Bhiwadi, Rajasthan, India) tubes per dilution and incubated at 35 °C for 24 h. Gas- and turbidity-positive tubes were confirmed in Brilliant Green Lactose Bile Broth (TM Media, Bhiwadi, Rajasthan, India) after a further 24 h at 35 °C. Results were calculated using an MPN table [15] and reported as MPN/g. All microbial analyses were performed in duplicate.

2.4. Physicochemical Analysis

Physicochemical properties, including pH, titratable acidity (TA), and instrumental color attributes, were evaluated. For pH and TA determination, 10 g of each sample was homogenized in 60 mL of distilled water using a stainless-steel blender. Sample pH was determined using a pH meter (Eutech™ pH 700 Meter, Eutech Instruments Pte Ltd., Singapore), while TA was determined via titration with 0.1 N NaOH to an endpoint pH of 8.2 and expressed as % malic acid, calculated using Equation (1) [16,17].

$$TA \left(\%\right)={\displaystyle \frac{{mL}_{NaOH}\times N_{NaOH}\times milliequivalent weight of malic acid }{{g }_{sample}}}\times 100$$

(1)

The milliequivalent weight of malic acid is 0.067.

Instrumental color attributes, including L* (lightness), a* (redness–greenness), and b* (yellowness–blueness), were recorded using a colorimeter (CR-20 Color Reader, Konica Minolta, Tokyo, Japan). All physicochemical analyses were performed in triplicate.

2.5. Data Analysis

Results were reported as mean ± standard deviation (SD). Data were analyzed using one-way ANOVA, followed by Duncan’s Multiple Range Test to identify significant differences among pretreatments. Dunnett’s test was applied to determine differences relative to the control. Statistical analyses were performed using SPSS Statistics v22 (IBM Corp., Armonk, NY, USA), with significance set at p < 0.05.

3. Results and Discussion

3.1. Effect of Coconut Residue Pretreatment on Microbial Quality

The microbial quality of coconut residue (CR) samples subjected to different pretreatments is summarized in Table 2. The Philippine National Standards for Coconut Flour (PNS-BAFPS 75-2010) was used as a guide due to the lack of specific standards for CR [18].

Table 2. Microbial quality of coconut residue samples subjected to different pretreatment conditions.

Treatment	Aerobic Plate Count (CFU/g)	Yeast and Mold Count (CFU/g)	Total Coliform (MPN/g)
Control	2.5 × 106 EAPC	5.2 × 102	>1.1 × 103
PR	7.9 × 102	<1.0 × 101	<3.0
PR-D50	3.7 × 104	1.0 × 103	<3.0
PR-D60	4.8 × 103	<1.0 × 101	<15
D50	2.5 × 106 EAPC	9.8 × 103	>1.1 × 103
D60	2.5 × 106 EAPC	2.1 × 103	>1.1 × 103

CFU: colony forming unit; EAPC: Estimated Aerobic Plate Count.

The Philippine National Standard specifies microbial limits for coconut flour as follows: aerobic plate count (APC) not exceeding 1.0 × 10⁴ CFU/g, yeasts and molds (YMC) not exceeding 1.0 × 10² CFU/g, and total coliforms (TC) not exceeding 5.0 × 10¹ CFU/g [18].

Pan-roasting treatments (PR and PR-D60) reduced microbial loads, with APC, YMC and total coliform falling within acceptable limits. PR-D50 also decreased microbial counts relative to the control, but APC and YMC remained above the acceptable limits, indicating only partial effectiveness. In contrast, drying alone (D50 and D60) did not substantially reduce microbial loads, with APC, YMC, and total coliforms remaining high. This suggests that drying alone may be insufficient to inactivate microorganisms, likely due to the relatively mild heat exposure compared with the direct heating achieved during pan-roasting. These results suggest that pan-roasting has potential to reduce microbial loads in CR. However, it should be noted that frozen storage after pretreatment and prior to analysis may have also contributed to limiting microbial proliferation in the samples [19,20]. Moreover, further studies are needed to confirm the effectiveness of pan-roasting under scaled-up operations, as the current study was conducted only on CR samples obtained at the laboratory-scale.

3.2. Effect of Coconut Residue Pretreatment on Physicochemical Quality

3.2.1. Titratable Acidity and pH

Titratable acidity (TA) and pH differed significantly among treatments (p < 0.05) (Table 3). The D50 sample exhibited the highest TA (0.29%) and the lowest pH (5.53), both significantly different from the control. TA and pH are interrelated indicators of food acidity, wherein an increase in TA is typically associated with a decrease in pH. While pH measures the concentration of free hydronium ions and reflects overall acidity or alkalinity, TA represents the total acid content [21].

Table 3. Titratable acidity and pH of coconut residue samples subjected to different pretreatment conditions.

Treatment Code	Titratable Acidity (% Malic Acid)	pH
Control	0.13 ± 0.02	6.14 ± 0.10
PR	0.11 ± 0.00 a	6.40 ± 0.05 d*
PR-D50	0.10 ± 0.00 a	6.26 ± 0.08 c
PR-D60	0.14 ± 0.01 b	6.42 ± 0.03 d*
D50	0.29 ± 0.01 c*	5.53 ± 0.01 a*
D60	0.15 ± 0.02 b	6.16 ± 0.02 b

Values are reported as mean ± SD (n = 3). Values within the same column with different superscript letters differ significantly (p < 0.05). Values with asterisk (*) differ significantly from the control sample (p < 0.05).

Malic and citric acids are naturally present in mature coconut flesh and may contribute to the initial acidity of CR samples [22]. However, the increased acidity observed in the D50 sample may be partly associated with microbial activity and the production of organic acids [23,24]. In contrast, PR samples (PR, PR-D50, and PR-D60) generally showed lower TA (0.10–0.14%) and higher pH values (6.26–6.42) than the control, suggesting reduced formation of spoilage-related acids. Although the changes in acidity are consistent with microbial results, other chemical or enzymatic processes may also have contributed.

3.2.2. Instrumental Color

Instrumental color of the CR samples was evaluated using L* (lightness), a* (redness–greenness), and b* (yellowness–blueness) parameters. The effects of different pretreatment conditions on these color attributes are presented in Table 4.

Table 4. Instrumental color attributes of coconut residue samples subjected to different pretreatment conditions.

Treatment	L*	a*	b*
Control	81.10 ± 0.53	−1.03 ± 0.15	4.20 ± 0.20
PR	75.83 ± 0.74 a*	−1.00 ± 0.10 a	4.03 ± 0.06 a
PR-D50	81.37 ± 0.61 c	−0.97 ± 0.12 a	5.87 ± 0.42 b*
PR-D60	78.07 ± 0.42 b*	−0.80 ± 0.2 0 a	6.60 ± 0.60 bc*
D50	85.13 ± 0.59 e*	−0.90 ± 0.00 a	6.17 ± 0.32 b*
D60	83.3 ± 0.70 d*	−0.5 ± 0.20 b*	7.17 ± 0.42 c*

Values are reported as mean ± SD (n = 3). Values within the same column with different superscript letters differ significantly (p < 0.05). Values with asterisk (*) differ significantly from the control sample (p < 0.05).

Generally, samples subjected to thermal pretreatments exhibited significant color changes. PR samples (PR and PR-D60) appeared darker than the control, as indicated by significantly lower L* values (p < 0.05), while D50 and D60 appeared lighter, with significantly higher L* values (p < 0.05). All pretreated samples, except PR, showed a significant increase in b* values (p < 0.05), suggesting increased yellowing compared with the control. The observed color changes in CR may be related to enzymatic and non-enzymatic Maillard reactions induced by thermal treatments [11,25,26]. Because color is an important quality attribute that can influence consumer preference and the potential use of CR in value-added products, the effects of thermal processing on visual quality should be considered.

4. Conclusions

Pan-roasting effectively improved the microbiological safety and physicochemical quality of CR, indicating its potential as a stabilization step prior to valorization, such as for dietary fiber extraction. However, given the preliminary nature of this study, further investigations involving dietary fiber compositional analysis, evaluation of techno-functional properties, and validation using CR from commercial VCO processing facilities are needed to confirm its suitability for industrial applications.