Be wary of losses during the refining of *Vernicia fordii* oil: 100% loss of β-carotene and 95% loss of β-sitosterol. How to choose the appropriate deacidification process?

To screen high-quality processing techniques for *Vernicia fordii* oil, this study used the same batch of *Vernicia fordii* fruits as raw material to investigate the effects of drying methods (hot air, vacuum microwave, infrared radiation), oil extraction methods (pressing, solvent extraction, aqueous enzymatic method), and refining processes (degumming, deacidification, decolorization, deodorization, dewaxing) on the content of beneficial lipid byproducts such as polyphenols, flavonoids, tocopherols (α/β/δ types), squalene, β-sitosterol, and β-carotene in the oil. Principal component analysis (PCA) and cluster heatmap analysis revealed the intrinsic relationship between processing techniques and component retention rates. The results showed that infrared radiation drying facilitated the dissolution of polyphenols, flavonoids, squalene, and β-carotene, while vacuum microwave drying better preserved β-sitosterol and tocopherol. Among the oil extraction methods, the retention rates of lipid byproducts were significantly higher in the aqueous enzymatic method and the pressing method than in the solvent extraction method. The aqueous enzymatic method showed excellent retention effects for polyphenols (541.62% higher than the pressing method), flavonoids (91.35% higher), and squalene (19.17% higher). The refining process resulted in varying degrees of loss of each component, with β-carotene completely lost, β-sitosterol loss reaching 95.42%, and δ-tocopherol loss at 62.56%. PCA showed that the oil samples extracted by vacuum microwave drying and pressing had high overall scores, while the overall score decreased with increasing refining, indicating a significant impact from the deacidification process. Cluster heatmaps confirmed the interaction between flavonoids and β-sitosterol, and between β-sitosterol and β-carotene. Multiple components, including carotene, showed a highly significant positive correlation (p<0.01). This study provides a theoretical basis for the targeted selection of processing technology for *Vernicia fordii* oil.

1. Introduction

*Idiospermum australiense*, an important woody oilseed plant in my country, has seeds with high oil content. The oil is rich not only in unsaturated fatty acids but also contains various beneficial lipid byproducts such as polyphenols, flavonoids, tocopherols, squalene, phytosterols, and β-carotene. These components possess antioxidant, anti-inflammatory, and lipid-regulating physiological functions, serving as the core carriers of the nutritional value and functional properties of *Idiospermum australiense* oil. Processing technology is a key factor affecting the retention of functional components in the oil. Differences in parameters during drying, oil extraction, and refining can lead to the degradation, loss, or transformation of lipid byproducts, directly impacting product quality.

Current research on *Idiospermum australiense* oil processing mainly focuses on oil extraction efficiency and fatty acid composition. Systematic research on the impact of processing technology on beneficial lipid byproducts is still incomplete. Therefore, this study, based on the same batch of raw materials, controlled for single variables to explore the influence of different processing stages on target components. Combined with multivariate statistical analysis, it clarified the correlation between key process parameters and component retention, aiming to provide a scientific basis for optimizing and selecting high-quality *Idiospermum australiense* oil processing technology.

2. Materials and Methods

2.1 Experimental Materials

Mature *Vernicia fordii* fruits from the same batch were used after removing impurities.

2.2 Processing

2.2.1 Drying Treatment

Three drying methods were used: ① Hot air drying: drying at a constant temperature of 60℃ until the moisture content was below 8%; ② Vacuum microwave drying: vacuum degree 0.08MPa, microwave power 300W, drying until the moisture content was below 8%; ③ Infrared radiation drying: radiation temperature 70℃, drying until the moisture content was below 8%. After drying, the shells were removed to obtain *Vernicia fordii* kernels.

2.2.2 Oil Extraction Methods

① Pressing Method: Oil is extracted by pressing with a screw press (temperature 120℃, pressure 30MPa), and then filtered to obtain crude oil;

② Solvent Extraction Method: Hexane is used as the solvent (solid-to-material ratio 1:8. extraction at 60℃ for 4 hours), and the solvent is removed by rotary evaporation to obtain crude oil;

③ Aqueous Enzymatic Method: A compound enzyme (cellulase: pectinase = 1:1) is added at 0.5%, enzymatic hydrolysis is performed at pH 4.5 at 50℃ for 2 hours, followed by centrifugation to extract oil, and then filtered to obtain crude oil.

2.2.3 Refining Process

Crude oil underwent degumming (85℃, pH adjusted to 2.5 with phosphoric acid), deacidification (neutralization of free fatty acids with sodium hydroxide solution), decolorization (3% activated clay added, decolorization at 105℃ for 30 min), deodorization (240℃, vacuum 0.095 MPa, deodorization for 2 h), and dewaxing (refrigerated at 4℃ for 24 h, then filtered). Oil samples were collected at each refining stage.

2.3 Determination Indicators and Methods

High-performance liquid chromatography (HPLC) was used to determine the contents of polyphenols, flavonoids, tocopherols (α/β/δ types), and β-sitosterol; gas chromatography-mass spectrometry (GC-MS) was used to determine the squalene content; and ultraviolet spectrophotometry was used to determine the β-carotene content.

2.4 Data Processing

Principal component analysis (PCA) was performed using SPSS 26.0. and cluster heatmaps were generated using GraphPad Prism 9. Pearson correlation analysis was used to test the associations between components; p < 0.05 was considered statistically significant, and p < 0.01 was considered highly statistically significant.

3. Results and Analysis

3.1 Effects of Different Drying Methods on Beneficial Lipid Concomitants

The three drying methods showed significant differences in the dissolution effects of lipid concomitants in *Vernicia fordii* oil (Table 1): The infrared radiation drying group showed significantly higher contents of polyphenols, flavonoids, squalene, and β-carotene, which were 32.15%, 28.76%, 25.34%, and 30.12% higher than the hot air drying group, respectively; the vacuum microwave drying group showed excellent contents of β-sitosterol and total tocopherol, which were 41.23% and 35.67% higher than the hot air drying group, respectively, with α-tocopherol and β-tocopherol contents being particularly prominent. This may be related to the penetrability of infrared radiation and the effect of the mild drying environment of vacuum microwaves on the cell structure of the raw material, reducing the degradation of heat-sensitive components.

3.2 Effects of different oil extraction methods on beneficial lipid byproducts

The oil extraction method significantly affected the retention rate of components (Table 2): the retention rates of lipid byproducts in the aqueous enzymatic method and the pressing method were significantly higher than those in the solvent extraction method. Among them, the aqueous enzymatic method had the best retention effect on polyphenols, flavonoids, and squalene, with contents of 128.64 mg/kg, 95.32 mg/kg, and 87.56 mg/kg, respectively, which were 541.62%, 91.35%, and 19.17% higher than those in the pressing method. The retention rates of tocopherol (especially δ-tocopherol) and β-sitosterol in the pressing method were slightly higher than those in the aqueous enzymatic method. In the solvent extraction method, due to the dissolution of organic solvents and the subsequent desolventizing process, most of the functional components were lost, and the contents of each indicator were very low. The advantage of the aqueous enzymatic method lies in the mild enzymatic hydrolysis process, which can destroy the cell wall structure of the raw material and promote the release of components, while avoiding the damage to components caused by high temperature or chemical solvents [4]. 3.3 Impact of Refining Processes on Beneficial Lipid Byproducts

Each refining process resulted in varying degrees of loss of lipid byproducts (Table 3): the overall loss rate was ranked as follows: β-carotene (100%) > β-sitosterol (95.42%) > δ-tocopherol (62.56%) > polyphenols (46.05%) > squalene (38.39%) > flavonoids (40.54%) > β-tocopherol (33.23%) > α-tocopherol (23.34%). The deacidification process (high-temperature neutralization reaction) and the deodorization process (high-temperature vacuum treatment) were the main sources of component loss. β-carotene was completely degraded during deodorization due to its high heat sensitivity, while β-sitosterol was significantly lost due to precipitation with fatty acids during deacidification. The degumming and decolorization processes had relatively smaller impacts on the components, but still resulted in partial loss of polar components such as polyphenols and flavonoids.

3.4 Principal Component Analysis (PCA) Results

PCA concentrated eight lipid adjuncts into two principal components, with a cumulative variance contribution rate of 89.67%, comprehensively reflecting the differences in oil sample quality (Figure 1): Among different drying methods, the vacuum microwave drying group had the highest overall score (2.86), followed by the infrared radiation drying group (1.52) and the hot air drying group (-4.38); among different oil extraction methods, the pressing method had the highest overall score (3.21), followed by the aqueous enzymatic method (1.89), while the solvent extraction method (-5.10) had the lowest; in the refining process, the overall score of the oil sample gradually decreased with increasing refining degree, with the crude oil score being high (4.53), and dropping sharply to 1.27 after deacidification. After deodorization and dewaxing, the score further decreased to -2.35. This indicates that the deacidification process has a significant impact on the functional quality of the oil sample.

3.5 Cluster Heatmap and Correlation Analysis

The cluster heatmap showed significant correlations among the lipid concomitants (Figure 2): flavonoids and β-sitosterol (p<0.01. r=0.83), β-sitosterol and β-carotene (p<0.01. r=0.89), α-tocopherol and β-tocopherol (p<0.01. r=0.82), and β-tocopherol and δ-tocopherol (p<0.01. r=0.90) showed highly significant positive correlations, indicating that these components are related in their form or metabolic pathway, and may be simultaneously lost or retained during processing. Although polyphenols and squalene showed no significant correlation, both exhibited high content in the aqueous enzymatic oil extraction group, suggesting that this process has a synergistic effect on the retention of both types of components.

4 Discussion

This study confirms that the three major processing steps—drying, oil extraction, and refining—significantly affect the beneficial lipid concomitants in *Vernicia fordii* oil. In the drying stage, infrared radiation and vacuum microwave drying are significantly superior to traditional hot air drying, which is consistent with the research conclusions of woody oils such as walnut oil and tea seed oil [5]. The core mechanism is that mild drying conditions can reduce cell structure damage and component oxidative degradation. Among the oil extraction methods, the aqueous enzymatic method, as a green processing technology, performs well in retaining functional components. This is closely related to the specificity and mildness of the enzymatic hydrolysis process. However, the solvent extraction method is not suitable for the production of high-quality tung oil due to environmental issues and component loss defects. The refining process is the main link in component loss, especially the high-temperature treatment of deacidification and deodorization processes, which is consistent with the general law of degradation of heat-sensitive components in oil refining [6]. The complete loss of β-carotene and the high loss rate of β-sitosterol suggest that if such components need to be retained, the refining process parameters should be optimized (such as reducing the deodorization temperature and shortening the processing time) or a mild refining technology should be adopted. The PCA composite score indicates that the combined process of vacuum microwave drying and pressing yields high-quality *Vernicia fordii* oil with excellent functional properties, while the degree of refining needs to be dynamically adjusted according to product requirements.

The highly significant positive correlations between components provide a new perspective for process optimization. For example, retaining β-tocopherol can simultaneously increase the content of δ-tocopherol without separate regulation. However, this study did not address the gradient optimization of processing parameters (such as drying temperature and enzymatic hydrolysis time). Future research could conduct response surface methodology experiments to clarify the quantitative impact of each process parameter on lipid byproducts.

5. Conclusions

Different processing methods significantly affect the content of beneficial lipid byproducts in *Vernicia fordii* oil: ① Among drying methods, infrared radiation drying facilitates the dissolution of polyphenols and flavonoids, while vacuum microwave drying better preserves tocopherol and β-sitosterol; ② Among oil extraction methods, the component retention rates of enzymatic and pressing methods are significantly higher than those of solvent extraction, with enzymatic methods showing excellent retention of polyphenols, flavonoids, and squalene; ③ The refining process leads to varying degrees of loss of components, with β-carotene being completely lost, and the deacidification process significantly impacting overall quality; ④ The combined process of vacuum microwave drying and pressing yields excellent functional qualities, with multiple highly significant positive correlations among the components.

In actual production, appropriate processes can be selected based on product requirements: if high antioxidant activity (polyphenols, flavonoids) is desired, infrared radiation drying + aqueous enzymatic oil extraction is preferred to simplify the refining process; if nutritional balance is emphasized (tocopherols, phytosterols), vacuum microwave drying + pressing is recommended to optimize deacidification and deodorization parameters; if large-scale, low-cost production is required, hot air drying + solvent extraction can be used, but subsequent functional component enhancement technologies are necessary. This study provides a theoretical basis for the optimization and standardization of high-quality *Vernicia fordii* oil processing technology. CHOCOMACH's fully automatichydraulic oil press empowers the high-quality development of the *Vernicia fordii* processing industry.