In this study, the sweet osmanthus (Osmanthus fragrans) and Shui Xian white tea (Camellia sinensis) were purchased from Zhenghe Yungen White Tea Co. Ltd. (Fujian, China). The rutin standard (purity ≥98%) was acquired from Shanghai Yuanye Biotechnology Co. Ltd. (Shanghai, China). The 1,1-diphenyl-2-picrylhydrazyl and 2,2′-azinobis-3-ethylbenzthiazoline-6-sulphonate were obtained from Sigma Co. Ltd. (Shanghai, China). The β-cyclodextrin, erythritol, and citric acid purchased were of food grade (purchased from Henan Wanbang Chemical Technology Co. Ltd., Henan, China). Other chemicals used for analysis were analytical grade (purchased from Sinopharm Group Chemical Reagent Co. Ltd., Shanghai, China).

Sweet osmanthus extract was prepared using the method described by Meng et al. (22). Sweet osmanthus was mixed with water at a 1:30 solid-to-liquid ratio (g/mL), and the mixture was subjected to ultrasonication at 120 W, 50°C for 35 min. The resulting liquor was filtered through a 200-mesh sieve to obtain the sweet osmanthus extract. Concurrently, white tea extract was prepared following the procedure outlined by Zhang et al. (23). A mixture of white tea and water at a solid-to-liquid ratio of 1:40 (g/mL) was prepared. This mixture was boiled for 5 min and subsequently strained through a 200-mesh sieve to obtain the white tea extract.

To prepare the IOWT, the sweet osmanthus and white tea extracts were mixed with varying proportions of β-cyclodextrin, erythritol, and citric acid. The mixtures were thoroughly stirred until all the dry ingredients were completely dissolved. Subsequently, the mixtures were centrifuged (Lishen Scientific Instrument Equipment, Neofuge-15R, Shanghai, China) at 2,795 × g for 10 min at −5°C. The supernatants were concentrated using a rotary evaporator (Shanghai Yarong, RE-52AA, Shanghai, China) at 45°C. Finally, concentrated extracts were obtained by spray (Huihetang Bioengineering Equipment, Bioq-8000, Shanghai, China) with a feed temperature of 150°C and a feed flow rate of 3.0 L/h to produce IOWTs (24).

The preparation of IOWTs is divided into two distinct phases: (1) the extraction of sweet osmanthus and white tea extracts and (2) the formulation of IOWTs, as outlined in Section 2.2 (Figure 1).

Based on the single-factor experiments (Section 2.3), the optimal range of single factors was preliminarily screened, resulting in the selection of four factors: sweet osmanthus-to-white tea ratio, β-cyclodextrin concentration, erythritol concentration, and citric acid concentration. An L₉ (3⁴) orthogonal experimental design was used with sensory scores and TFC as the evaluation indicators (Table 1).

Sensory evaluation was conducted following the methodology outlined by Lin et al. (25). The prepared IOWT was brewed with water at an optimal 1:300 (g/mL) ratio at 85°C. A panel of 10 (five males and five females) trained professional tea experts evaluated the sensory quality of the brewed IOWT. The evaluation encompassed four sensory attributes: color (maximum score of 20), aroma (maximum score of 30), taste (maximum score of 30), and solubility (maximum score of 20) (Table 2).

The TFC in the IOWT samples was assessed using the method outlined by Czechowski et al. (26). In brief, the IOWTs were brewed with water at the optimal tea-to-water (1:300, g/mL) at 85°C. Subsequently, 0.2 mL of the brewed tea samples were transferred to 10 mL test tubes, and the volume was brought up to 5 mL using 80% ethanol solution. The mixtures were thoroughly mixed before measuring the absorbance at 510 nm. The TFC (expressed in %) of the IOWT was calculated using the rutin standard curve (y = 10.82x + 0.0123, R² = 0.997) and Equation 1.

where, C (in mg/mL) represents the concentration of the total flavonoids, V₁, V₂, and V₃ (in mL) represents the volume of the sample solution during measurement, the total volume of the sample solution, and the volume of the sample solution after dilution, respectively, and M (in g) represent the mass of the sample.

The antioxidant activities of the IOWT powder at various concentrations were compared with those of Vit. C (ascorbic acid) and instant white tea. In the preparation step, 0.1 g of Vit. C, instant white tea, and IOWT powder were accurately weighed. Using distilled water, ascorbic acid, instant white tea, and IOWT solutions were prepared at varying concentrations of 0.2, 0.4, 0.6, 0.8, and 1.0 mg/mL. These sample solutions were utilized for the following in vitro antioxidant assays.

All assays were conducted in triplicate. The data is presented in mean ± standard deviation. Analysis of variance (ANOVA) and Duncan’s multiple range test were employed to analyze the data utilizing the SPSS statistical software (IBM, Chicago, United States). A p-value less than 0.05 was deemed statistically significant.

As β-cyclodextrin content increases from 4 to 8%, sensory evaluation scores exhibit a slight upward trend, indicating that higher β-cyclodextrin content enhances sensory acceptability (Figure 2A). Conversely, flavonoid content decreases consistently, suggesting a negative correlation between β-cyclodextrin levels and flavonoid retention. This phenomenon may be attributed to the increase in β-cyclodextrin content leading to the inclusion interaction between β-cyclodextrin and flavonoid compounds, thereby altering the physicochemical properties of flavonoids and resulting in a decline in TFC (23). Sensory scores were improved when β-cyclodextrin was added in concentrations ranging from 4 to 7%. This improvement can be attributed to the minimal impact of varying proportions of β-cyclodextrin on taste and aroma compared to its more pronounced effects on color and solubility (29). With increasing concentrations of β-cyclodextrin, the IOWT exhibited brighter colors and improved solubility. Since β-cyclodextrin possesses a sweet taste, the sensory score peaked at 77.17 with a 7% addition. Within the scope of this study, the TFC was at its highest, 31.07%, when the β-cyclodextrin concentration was 4%, resulting in a sensory score of 75. Considering both parameters, the influence of β-cyclodextrin on sensory scores was negligible. Consequently, β-cyclodextrin was added at 4% for subsequent single-factor experiments.

As the sweet osmanthus-to-white tea ratio increases from 1:10 to 5:10, sensory evaluation scores peak at 2:10, followed by a sharp decline (Figure 2B). Conversely, flavonoid content rises steadily with the rising ratio, indicating a trade-off between sensory acceptability and flavonoid retention as the ratio changes. The higher the proportion of the sweet osmanthus extract, the higher the TFC and the more pronounced the sweet osmanthus fragrance. However, increasing TFC can also lead to a less desirable taste (17). The taste became bland with a 1:10 ratio, and only a faint scent of sweet osmanthus was detectable. At a 2:10 ratio, the taste became mellow and sweet, with the unique aromas of white tea and sweet osmanthus present, reaching the highest sensory score of 83.1 points and a TFC of 27.74%. At a 3:10 ratio, the tea flavor became strong and exhibited bitterness, though the sweet osmanthus fragrance was highlighted. Different solid-to-liquid ratios had a notable impact on the sensory attributes of the final product, including taste, color, and concentration (30). Considering all factors, the sweet osmanthus-to-white tea ratio of 2:10 (or 1:5) offers the best balance and was used for subsequent single-factor experiments.

As erythritol content increases from 2 to 10%, sensory evaluation scores peak at ~6%, followed by a rapid decline (Figure 2C). This observation suggested that moderate levels of erythritol are optimal for sensory acceptability, while higher levels diminish consumer preference. Similarly, flavonoid content decreases gradually with increasing erythritol content, indicating a negative impact on flavonoid retention at higher erythritol concentrations. Consistent with previous findings, Kamkaew et al. (31) demonstrated that sweeteners significantly influence the stability of flavonoid solutions. In their study, Wojtyś et al. (32) investigated the osmotic dehydration of Japanese quince fruit in erythritol solution. Their findings revealed a notable impact of erythritol on the flavonoid content of the fruit, suggesting that sweeteners may mask the bitterness of flavonoids. The product lacks sufficient sweetness at a concentration of 2%, resulting in a sensory score of 83.1 points. At a concentration of 6%, the product exhibits a rich tea flavor that is sweet and sour, making it delectable, with the highest sensory score of 87.1 points and a TFC of 25.83%. The product’s sweetness became excessive at a concentration of 10%, leading to the lowest sensory scores and TFC. Considering all factors, the optimal taste is achieved with 6% erythritol. This concentration was used for subsequent single-factor experiments.

As the citric acid content increases from 0.4 to 1.2%, the sensory evaluation scores peak at 0.6% and decline sharply, indicating that higher citric acid concentrations adversely affect sensory acceptability (Figure 2D). Conversely, flavonoid content progressively decreases with increasing citric acid content, suggesting a potential degradation of flavonoids at elevated acidity levels. The study by Santosh et al. (33) proposed that the reduction in flavonoid content may be attributed to multiple H⁺ ions in citric acid. At a citric acid concentration of 0.6%, the sensory score was 88.2 points (the highest), devoid of any sour taste, and the TFC was 25.20%. At a citric acid concentration of 1.0%, the taste began to exhibit sourness, and at 1.2% citric acid, the taste became notably sour and astringent, with the lowest TFC. The outcomes were comparable to those reported in a recent study conducted by Wang et al. (34). The study demonstrated that reducing citric acid levels alleviated sourness and astringency, enhancing the mouthfeel and releasing total flavonoid content. Notably, the antioxidant activity exhibited a notable increase. Considering all factors, the optimal taste is achieved with a citric acid concentration of 0.6%. This concentration was used for subsequent single-factor experiments.

From the sensory range values (R_I), it is evident that the factors influencing the sensory scores of the product were ranked in the order of A > C > D > B (Table 3). This indicates that the factors affecting the product’s sensory scores are the sweet osmanthus-to-white tea ratio (A), erythritol concentration (C), citric acid concentration (D), and β-cyclodextrin concentration (B), respectively. According to the flavonoid range analysis (R_II) results, the primary factors affecting the flavonoid content of the product were ranked as A > C > B > D, which translates to the sweet osmanthus-to-white tea ratio > erythritol > β-cyclodextrin > citric acid. Among these, the sweet osmanthus-to-white tea ratio has the most significant impact on the IOWT’s sensory scores and the TFC. A study by Jiang et al. (19) demonstrated that various experimental factors had varying effects on the process optimization, physicochemical properties, and antioxidant activity of instant black tea.

The combination A₂B₂C₂D₁ was the optimal formulation for the IOWT (Table 3), which contradicts the highest-scoring experiment number 5 in the orthogonal test. Consequently, this formulation was retested for validation. The results showed that the sensory score and TFC of A₂B₂C₂D₁ were slightly higher than those of A₂B₂C₃D₁, with a sensory score of 89.5 (characterized by a unique aroma of white tea and sweet osmanthus fragrant, a rich and sweet-tasting, a clear and bright color, and a rapid dissolution rate). Furthermore, the TFC of A₂B₂C₂D₁ was 25.76%. Therefore, the optimal IOWT formulation was determined to be A₂B₂C₂D₁, which corresponds to a sweet osmanthus-to-white tea ratio of 1:5, 4% β-cyclodextrin, 6% erythritol, and 0.5% citric acid. Recently, Jiang et al. (19) employed sequential inoculation with golden flower (Eurotium cristatum) and Aspergillus niger in liquid-state fermentation to produce instant black tea and similarly conducted validation experiments to investigate the influencing factors.

Compared to the TFC in the white tea extract and IOWT, the TFC was found to have increased from 6.75 to 25.76%, a rise of 3.82 times (Table 4). Simultaneously, compared to the TFC of instant white tea, the TFC of IOWT was 36.30% higher. This increase in the TFC indicated that after adding sweet osmanthus extract under the same processing conditions, there was a significant increase in the TFC of the instant white tea, eliminating the possibility of factors related to the processing technique.

Chen et al. (41) investigated ultrasonic-assisted extraction of total flavonoids in sweet osmanthus, resulting in an extraction rate of 17.7%. Hu et al. (42) analyzed white tea’s physical and chemical composition and reported a flavonoid content ranging from 0.51 to 1.43%. Wang et al. (35), Zhou et al. (36), and Ning et al. (43) also analyzed the flavonoid components in sweet osmanthus and white tea, identifying several similarities, including quercetin, catechins, and rutin. However, the specific concentrations and other components vary between the two. Sweet osmanthus contains a higher variety of flavonoid glycosides and hesperidin, while white tea primarily comprises catechin compounds, particularly epicatechin gallate (EGCG). Consequently, combining the flavonoids from white tea and sweet osmanthus can substantially enhance the antioxidant effect through complementary species and synergistic scavenging of free radicals.