Optimizing Parameters of Strong Oxidizing Free Radicals Application for Effective Management of Wheat Powdery Mildew

Abstract

Wheat powdery mildew is a major fungal disease threatening global wheat production. To develop an effective and environmentally friendly control strategy, this study systematically evaluated the disease-suppressive efficacy of strong oxidative free radicals across a series of treatment parameters, including radical concentrations (3.0–8.0 mg/L), spraying durations (20–60 s), solution pH levels (5–8), spraying heights (0–20 cm), and treatment timings corresponding to different infection stages (0–120 h post-inoculation). Response surface methodology (RSM) was used to optimize these variables with the objective of maximizing disease control efficacy. The results showed that control efficacy increased with radical concentration up to 5.0 mg/L, beyond which a saturation effect was observed. The most effective conditions included a spraying duration of 50 s and a height of 6.5 cm. Maximum suppression was achieved when the treatment was applied within 0–12 h post-infection. Moreover, adjusting the solution pH to a range of 5–7 further enhanced the efficacy. The RSM-based predictive model demonstrated high accuracy (R² = 0.9942), and the optimized parameters—6.65 mg/L radical concentration, 50.84 s spraying duration, and treatment at 15.67 h post-infection—yielded a predicted control efficacy of 97.64%, with a validation error below 0.5%. This study provides a quantitative basis for the precise and sustainable deployment of free radical-based treatments in wheat disease management.

1. Introduction

Wheat (Triticum aestivum L.) is one of the most important staple crops worldwide, It occupies approximately 17% of the world’s cultivated land and contributes essential calories and proteins to more than 35% of the global population annually [1,2,3,4]. Wheat powdery mildew, caused by the obligate biotrophic fungus Blumeria graminis f. sp. tritici (Bgt), is a widespread and economically significant foliar disease [5,6]. The pathogen spreads rapidly through conidial spores and undergoes repeated infection cycles under favorable field conditions [7,8]. Bgt primarily infects wheat leaves and sheaths but may also affect stems and spikes in severe cases, leading to significant yield and quality losses [9]. In epidemic years, powdery mildew can cause yield reductions ranging from 10% to 30%, posing a long-term threat to global food production and security [10].

The traditional control of wheat powdery mildew has predominantly relied on broad-spectrum chemical fungicides such as triazoles, benzimidazoles, and pyridine carboxamides [11]. However, sustained and intensive application of these compounds has accelerated the development of resistance within Bgt populations, with new resistant strains rapidly emerging and undermining fungicide efficacy [12]. Furthermore, excessive chemical use poses broader risks raising serious concerns regarding environmental health and food safety. In response to these concerns, integrated and ecologically based plant protection approaches have gained attention in recent years [13]. These include the use of microbial antagonists such as Bacillus subtilis to inhibit fungal development [14,15,16], the application of immune-priming biostimulants to activate host defenses [17], and the cultivation of genetically resistant varieties harboring Pm resistance genes [18,19,20]. Although promising, these strategies face technical and operational limitations in large-scale field applications and currently lack the consistency and robustness required to replace chemical fungicides entirely. In light of these limitations, there is a pressing need to develop novel, efficient, environmentally friendly, and sustainable strategies for wheat disease control. In recent years, strong oxidative free radicals—such as hydroxyl radicals (•OH), sulfate radicals (SO₄⁻), and ozone-derived radicals—have been widely applied in plant disease management, postharvest preservation, and water treatment due to their broad-spectrum antimicrobial activity and environmental compatibility [21,22,23]. These radicals possess high oxidation potential and rapid reaction kinetics, enabling them to quickly degrade microbial membranes and genetic material, thereby exerting potent bactericidal and fungicidal effects [24,25]. Moreover, they can degrade various organic pollutants, including mycotoxins, highlighting their potential for controlling Fusarium head blight (FHB) and associated toxin contamination in wheat [26,27].

Unlike conventional chemical fungicides, strong oxidative radicals act in a non-selective and rapid manner, reducing the risk of resistance development and enhancing mycotoxin degradation. Numerous studies have demonstrated the significant efficacy of ozone against various pathogens and in food preservation. For example, Agostini et al. reported that gaseous and aqueous ozone at concentrations of 40–60 μg/mL effectively inhibited Staphylococcus aureus and Enterococcus faecalis [28]; Chaidez et al. showed that ozonated water significantly reduced Salmonella contamination in tomatoes, particularly when water turbidity remained at 2 NTU [29]. In food safety applications, ozone has been shown to degrade over 65% of aflatoxins and aflatoxin B1 in peanuts within 30 min at 6.0 mg/L under 5% humidity [30]. Applications in strawberries, poultry, and alfalfa sprouts have also demonstrated its ability to delay pathogen development and extend shelf life [31,32,33]. Additionally, ozone treatments have been used to prolong the shelf life of food products (e.g., buckwheat noodles [34]), enhance the bactericidal activity of disinfectants against Listeria monocytogenes [35], and control Fusarium graminearum and its mycotoxin deoxynivalenol (DON) in wheat [36]. Recent advances include the use of ozone micro–nano-bubbles (O₃-MNBs) to improve fish preservation in chilled systems [37], and combined ozone–ultrasound treatments have been shown to enhance and retain desirable volatiles in tomatoes [38]. Pre-treatment with ozonated water before drying garlic slices significantly increased allicin content (by 7.85%), total phenolics (25.90%), and antioxidant activity (12.31%) compared to untreated controls [39].

Although technologies based on strong oxidative free radicals have shown promise in plant disease management, their antifungal efficacy against Bgt, the causal agent of wheat powdery mildew, remains inadequately explored. This study aims to systematically evaluate the inhibitory effects of oxidative radical solutions on Bgt and to optimize key application parameters. A series of single-factor experiments were conducted to investigate the effects of radical concentration, treatment duration, spraying height, inoculation timing, and solution pH on disease control efficacy. Based on these results, a response surface methodology was employed to develop an efficient and operationally feasible disease management strategy. The findings provide a technical foundation for the practical application of oxidative radical solutions in the green control of wheat diseases and offer a promising, safe, and precise alternative to conventional chemical fungicides, supporting the transition toward sustainable plant protection practices.

2. Materials and Methods

2.1. Materials

2.1.1. Plant Material

The wheat cultivar used in this study was ‘Yangmai 23’, a highly susceptible variety to powdery mildew, developed by the Lixiahe Regional Institute of Agricultural Sciences, Jiangsu Province, China. Seeds were sown in an artificial climate chamber maintained at 22 ± 2 °C with 75% relative humidity. After seven days of growth, seedlings at the fully expanded first-leaf stage were selected for inoculation.

2.1.2. Pathogen Isolate

The Blumeria graminis f. sp. tritici isolate (BgtYZ01) is a highly virulent strain originally collected from an epidemic field in Yangzhou, Jiangsu Province, China [40]. This isolate was used for all inoculation experiments in this study.

2.1.3. Preparation of the Strong Oxidizing Free Radicals (SOR) Solution

The strong oxidizing free radicals (SOR) solution was generated using a plasma-assisted gas–liquid discharge system custom designed by our research team. Under a high-frequency, high-voltage electric field, oxygen molecules were ionized and dissociated to produce various gaseous reactive oxygen species (ROS), including ozone (O₃), atomic oxygen (O), and superoxide anions (O₂⁻). Upon dissolution into water, these species further reacted with water molecules and dissolved oxygen—facilitated by micro–nano-bubble (MNB) gas–liquid conversion technology—to efficiently produce a highly oxidative aqueous solution containing short-lived yet potent radicals such as hydroxyl radicals (•OH), hydroperoxyl anions (HO₂⁻), and ozonide anions (•O₃⁻). These highly reactive species collectively constitute the SOR solution. Among them, ozonated water was identified as the predominant and most stable component. In contrast, radicals such as •OH and •O₃⁻ have extremely short lifetimes (on the order of nanoseconds), rendering their direct quantification technically unfeasible. Therefore, the ozone concentration, measured using a dissolved ozone analyzer (TL-2000, Tonglin Technology Co., Ltd., Guangzhou, China), was adopted as a practical and representative indicator of the overall oxidative strength of the SOR solution.

The SOR generation system included a high-energy ionization discharge unit and a gas–liquid mixing module. The discharge chamber was designed with a 0.5 mm gap between the electrodes, which were fabricated from sintered silver-coated metal materials. A ceramic dielectric plate was used as the insulating layer. The total power of the gaseous ROS generator was 600 W, and the gas–liquid mixing unit operated at 0.75 kW. Ozone gas was introduced at a flow rate of 3.0 L/min.

2.2. Propagation and Collection of BgtYZ01

The purified BgtYZ01 isolate was inoculated onto wheat seedlings using the tapping method. After inoculation, the plants were covered with glass domes to ensure isolation and placed in a greenhouse maintained at 24 ± 2 °C. The photoperiod was set to 16 h of light and 8 h of darkness per day. When abundant conidial colonies developed on the leaf surfaces, the spores were carefully collected into sulfuric acid paper bags and stored for subsequent use.

2.3. Preparation of Powdery Mildew Spore Suspension and Inoculation Procedure

Conidial suspensions of Blumeria graminis f. sp. tritici were prepared using phosphate-buffered saline (PBS, pH 7.0–7.4). The suspension was thoroughly vortexed to ensure uniform dispersion of spores, and the concentration was adjusted to 5.0 × 10⁷ cfu/mL using a Neubauer (Munich, Germany) improved hemocytometer (0650010). The spore suspension was then evenly applied to the surface of wheat leaves using a soft brush to facilitate conidial germination and successful infection.

2.4. Design of Treatments with Different Concentrations of Strong Oxidizing Free Radicals for Powdery Mildew Control

Uniform and vigorous wheat seedlings with fully expanded leaves (7 days after germination) were selected as experimental materials. To ensure uniform conidial distribution, the spore suspension was gently stirred during application, and a soft brush was used to evenly spread the suspension across the adaxial surface of each leaf. The seedlings were maintained under high humidity conditions within 24 h post-inoculation to promote fungal infection.

Immediately following inoculation, the plants were sprayed with strong oxidizing free radicals solutions at five concentrations (0, 3.0, 4.0, 5.0, and 10.0 mg/L) [41,42]. Untreated seedlings, which were inoculated with Bgt but received no oxidative radical treatment, served as the control for evaluating disease control efficacy. A precision sprayer was used to apply the solutions evenly across the leaf surfaces from a fixed distance of 6.5 cm. Each treatment lasted 40 s. The solution pH was maintained at 7.0. For each concentration, three biological replicates were established, with each replicate consisting of a single pot containing 20 wheat seedlings, in order to minimize experimental variability and ensure uniform treatment conditions.

After treatment, the wheat seedlings were maintained under controlled environmental conditions (22 ± 2 °C, 75% relative humidity, and a 16 h light/8 h dark photoperiod) for 7 days to allow for disease development. At the end of the incubation period, pustules caused by BgtYZ01 were counted under a light microscope. All leaves from each treatment group were examined individually. The average number of pustules was used as a quantitative measure of powdery mildew sporulation. Control efficacy was calculated as:

$$\mathrm{Control}\mathrm{efficacy}(\%)={\displaystyle \frac{\mathrm{Number}\mathrm{of}\mathrm{germinated}\mathrm{spores}\mathrm{in}\mathrm{control}-\mathrm{Number}\mathrm{in}\mathrm{treatment})}{\mathrm{Number}\mathrm{in}\mathrm{control}}}\times 100\%$$

To ensure experimental consistency and reproducibility, the following parameters were held constant during the experiment process: the concentration range of strong oxidative free radicals (3–10 mg/L), spraying duration (20–60 s), spraying height (0–20 cm), inoculation time with BgtYZ01 (0–120 h), and the pH range of the oxidative solution (4–8).

2.5. Experimental Setup of Operation Time Treatments Using Strong Oxidizing Free Radicals

Seven-day-old wheat seedlings with uniform growth and leaf development were selected. Following artificial inoculation with BgtYZ01, seedlings were maintained under conditions conducive to pathogen infection. A 4.0 mg/L solution of strong oxidizing free radicals (pH 7.0) was applied immediately post-inoculation, while control plants received no treatment.

Spray application was carried out using a precision atomizer, positioned 6.5 cm above the foliage. The durations of spraying were set at five time intervals: 20 s, 30 s, 40 s, 50 s, and 60 s. For each time point, three biological replicates were established, each consisting of one pot containing 20 wheat seedlings to minimize experimental variation.

2.6. Spraying Height Optimization for Strong Oxidizing Free Radicals Application in Wheat Powdery Mildew Control

Wheat seedlings (7 days old, uniform, and healthy) were artificially inoculated with BgtYZ01, ensuring conidia were evenly deposited on all leaf surfaces. A 4.0 mg/L strong oxidizing free radicals solution (pH 7.0) was applied immediately after inoculation. Control plants received no treatment.

The spray was applied using Electrostatic Sprayer (Mistec Spray Tech Co., Ltd., Dongguan, China) at five height levels: 0, 6.5, 10, 15, and 20 cm above the leaf surface. Each treatment lasted 40 s with consistent spraying pressure to guarantee even distribution across all foliage. In the 0 cm treatment, the atomizer nozzle was held directly above the leaf surface without contact. This setting was used to assess the influence of minimal spray distance on disease control efficacy. Each spraying height included three biological replicates, each composed of a pot containing 20 wheat seedlings, thereby reducing biological variation.

2.7. Treatment Design Based on Pathogen Infection Timing in Powdery Mildew Control

Seven-day-old wheat seedlings, characterized by uniform growth and morphology, were inoculated with BgtYZ01 conidia through uniform spraying to ensure consistent surface infection. Spraying treatments with strong oxidizing free radicals were carried out at seven different post-inoculation time points: 0, 12, 24, 36, 48, 96, and 120 h. The applied solution was maintained at 4.0 mg/L, pH 7.0.

A precision sprayer was used for all applications, maintaining a 6.5 cm nozzle-to-leaf distance and a 40 s spray duration for each treatment. Three biological replicates were included per treatment, with each replicate consisting of 20 wheat seedlings housed in a single pot.

2.8. Effect of Solution pH on Strong Oxidizing Free Radicals-Based Control of Wheat Powdery Mildew

Wheat seedlings with uniform growth and fully expanded leaves (7 days after germination) were selected for the experiment. After inoculation with BgtYZ01 to ensure even pathogen distribution, plants were immediately subjected to foliar spraying with 4.0 mg/L strong oxidizing free radicals solutions.

In order to assess the influence of solution pH on the efficacy of treatment against wheat powdery mildew, the pH of each solution was adjusted accordingly; the pH range (4–8) represents typical environmental conditions suitable for radicals stability [43]. Citric acid was used to adjust the pH to 4.0, 5.0, and 6.0, while sodium bicarbonate (NaHCO₃) was used to adjust the pH to 7.0 and 8.0. The untreated solution had an initial pH of approximately 6.5. All treatments were applied using a precision sprayer positioned 6.5 cm above the leaf surface for a duration of 40 s. Plants in the control group were left untreated.

2.9. Optimization of Strong Oxidative Free Radicals Application Parameters Against Wheat Powdery Mildew Using Box–Behnken Design

Based on results from preliminary single-factor experiments that examined the individual effects of different application variables on disease suppression, a response surface methodology (RSM) approach was employed to systematically optimize the treatment conditions for strong oxidative free radicals application. The optimization focused on three key parameters: (A) concentration of the strong oxidative free radicals (mg/L), (B) operation time (s), and (C) infection time (h post-inoculation). Each factor was tested at three levels, selected based on their significant individual effects on disease suppression observed in earlier trials. In this study, optimization refers to the process of identifying the most effective combination of application conditions that result in the highest level of disease control efficacy.

All experimental data were statistically analyzed using Microsoft Excel 2019. A three-factor, three-level Box–Behnken design (BBD) was constructed using Design-Expert 12.0 software (Stat-Ease Inc., Minneapolis, MN, USA) to evaluate the interactive effects of radicals concentration (3–7 mg/L), spraying time (45–55 s), and infection time (6–24 h post-inoculation) on disease control efficacy. The selection of minimum and maximum levels for each factor was informed by preliminary single-factor experiments, which established the biologically effective and operationally applicable ranges for use in field or laboratory conditions. The design comprised 17 runs, including 5 center points, with each treatment replicated three times. The model’s adequacy was assessed by ANOVA. Residual normality and variance homogeneity were verified using Shapiro–Wilk and Levene’s tests, respectively. Numerical optimization aimed to maximize control efficacy, and optimal conditions were selected based on desirability scores and predicted confidence intervals. The disease control efficacy served as the response variable. The experimental design matrix and corresponding factor levels are presented in

Table 1. Box–Behnken test factors and levels of the control effect of strong oxidative free radicals on wheat powdery mildew.

Factor	Unit	−1	0	1
A: concentration	mg/L	3	5	7
B: operation time	s	45	50	55
C: infection time	h	6	15	24

.

2.10. Data Analysis

All data were measured in triplicate, and outliers were excluded prior to analysis. One-way analysis of variance (ANOVA) was performed using SPSS 16.0 software (IBM Corp., Armonk, NY, USA), with the F-test used to evaluate between-group differences. Statistical significance was defined as p ≤ 0.05. For RSM, Design-Expert software version 12.0 was used at a 95% confidence level. A second-order polynomial regression model was fitted to describe the relationship between the response variable and the experimental factors. The significance of the model and its terms was assessed by ANOVA. To evaluate the validity of the regression model assumptions, diagnostic plots were examined. The normal probability plot of externally studentized residuals was used to test the normality of residuals, while the residuals versus predicted values plot was used to verify the homogeneity of variance. These diagnostics ensured the reliability of model predictions for subsequent optimization.

3. Results

3.1. Effect of Strong Oxidizing Free Radicals Concentration on the Control of Wheat Powdery Mildew

In this experiment, the inhibitory effects of strong oxidizing free radical solutions at different concentrations (3.0 mg/L, 4.0 mg/L, 5.0 mg/L, and 10.0 mg/L) on BgtYZ01 were evaluated. Control efficacy was assessed by calculating the control effect (%). As shown in the corresponding Figure 1, the results exhibited a clear dose–response relationship, with disease suppression increasing significantly as the concentration increased. However, a saturation trend was observed at higher concentrations.

In the concentration range of 0–5.0 mg/L, the control efficacy increased rapidly with rising concentrations, exhibiting a clear nonlinear growth trend. At 3.0 mg/L, the control efficacy reached 62.1%, increasing to 84.3% at 4.0 mg/L, and peaking at 90.3% at 5.0 mg/L. Beyond 5.0 mg/L, the increase in control efficacy began to plateau, reaching 98.4% at 8.0 mg/L. When the concentration was further increased to 10.0 mg/L, the efficacy remained at nearly 100%, indicating that a saturation threshold may have been reached, beyond which a higher concentration did not further improve the control effect.

3.2. Effect of Operation Time on the Control Efficacy of Strong Oxidizing Free Radicals Against Wheat Powdery Mildew

To evaluate how spraying time influences the efficacy of strong oxidizing free radicals, wheat seedlings were treated with a 4.0 mg/L solution for varying durations. As shown in the Figure 2, the control efficacy increased with longer spraying durations and tended to plateau beyond 50 s.

At a spraying duration of 20 s, the control efficacy was only 34.3%, suggesting insufficient solution coverage or contact time with the pathogen, which may have led to incomplete inactivation. When the spraying time increased to 30 and 40 s, the efficacy improved to approximately 57.0% and 72.7%, respectively, indicating that longer exposure facilitates greater interaction between the active ingredients and the pathogen. A further increase to 50 s resulted in near-complete control (close to 100%). However, no significant difference was observed between the 50 s and 60 s treatments, suggesting that maximum efficacy was reached at 50 s and prolonged spraying beyond this duration did not enhance disease control.

3.3. Effect of Spraying Height on the Control Efficacy of Strong Oxidizing Free Radicals Against Wheat Powdery Mildew

The control efficacy of strong oxidizing free radicals applied at different spraying heights (0, 6.5, 10, 15, and 20 cm) was evaluated. As presented in Figure 3, the results showed that control efficacy initially increased and then decreased with increasing spray height, exhibiting a distinct unimodal curve. The highest efficacy (84.7%) was achieved at a height of 6.5 cm, which was significantly higher than that at other heights. This suggests that 6.5 cm is optimal for achieving fine atomization and uniform droplet distribution. The lowest control efficacy was observed at 0 cm and 20 cm, approximately 72.3% and 61.2%, respectively, indicating that both overly short and overly long spray distances are unfavorable for effective droplet deposition. The confidence interval of the trend line indicated high data stability around 6.5 cm, further supporting its reliability. Therefore, 6.5 cm is considered the optimal spray height under the conditions of this experiment.

3.4. Effect of Infection Timing on the Control Efficacy of Strong Oxidizing Free Radicals Against Wheat Powdery Mildew

The efficacy of 4.0 mg/L strong oxidizing free radicals against wheat powdery mildew was tested at different post-inoculation time points (0, 12, 24, 36, 48, 96, and 120 h). As shown in Figure 4, the inhibitory effect of strong oxidative free radicals gradually declined with increasing infection time, indicating that early-stage application leads to higher control efficacy, while delayed treatment becomes significantly less effective. The highest control efficacy (89.7%) was achieved when treatment was applied at 0 h post-inoculation, suggesting that application prior to the establishment of infection structures can effectively suppress disease development. Although the efficacy slightly declined at 12 h, it remained above 80%, indicating that Blumeria graminis f. sp. tritici remains sensitive to oxidative radicals during early colonization. With further delay, control efficacy decreased progressively to 74.4% at 24 h, 69.5% at 36 h, and 54.0% at 48 h, demonstrating that prolonged pathogen residence on leaf surfaces reduces treatment effectiveness. At 96 h and 120 h post-inoculation, control efficacy dropped below 50%, indicating that the pathogen had completed most of its infection process, rendering chemical control considerably less effective.

3.5. Effect of pH on the Control Efficacy of Strong Oxidizing Free Radicals Against Wheat Powdery Mildew

To investigate the influence of pH on the antimicrobial activity of strong oxidizing free radicals, a 4.0 mg/L solution was adjusted to different pH levels (4, 5, 6, 7, and 8) using citric acid. The treatment was applied to inoculated wheat seedlings under standardized spraying conditions. As shown in Figure 5, control efficacy was significantly influenced by the pH.

At pH 4, the efficacy was relatively low (52.2%), while increasing the pH to 5 and 6 markedly enhanced disease suppression to 84.5% and 93.6%, respectively. At pH 7, the efficacy slightly declined to 88.6% but remained high. However, a further increase to pH 8 caused a substantial drop in the efficacy to approximately 43.1%.

3.6. Box–Behnken Experimental Results

3.6.1. Experimental Results and Model Establishment

Based on preliminary single-factor experiments, three key factors—strong oxidizing free radicals, spraying time, and infection time—were selected for optimization due to their pronounced effects on the efficacy of strong oxidizing free radicals in controlling wheat powdery mildew across the tested ranges. Each factor was evaluated at three levels using a Box–Behnken design, which was employed for experimental planning, statistical analysis, and response surface optimization. The experimental data are presented in

Table 2. Results of Box–Behnken test on the control of wheat powdery mildew by strong oxidative free radical solution.

NO.	A/Concentration (mg/L)	B/Operation Time (s)	C/Infection Time (h) (mg/L)	Y/Control Effect (%)
1	3	45	15	91.01
2	7	45	15	93.59
3	3	55	15	92.13
4	7	55	15	95.60
5	3	50	6	91.84
6	7	50	6	95.12
7	3	50	24	88.87
8	7	50	24	89.34
9	5	45	6	93.07
10	5	55	6	95.13
11	5	45	24	89.67
12	5	55	24	90.35
13	5	50	15	98.46
14	5	50	15	99.27
15	5	50	15	99.19
16	5	50	15	99.07
17	5	50	15	99.52

.

The experimental data presented in Table 2 were analyzed using Design-Expert version 12.0. A second-order polynomial regression model was developed to describe the relationship between the control efficacy (%) and the three independent variables: A (concentration), B (operation time), and C (infection time). The resulting regression equation is as follows:

Y = 99.10 + 1.23 A + 0.7338 B − 2.120 + 0.2225 AB − 0.7025 AC − 0.3450 BC − 3.3.

3.6.2. Analysis of Variance (ANOVA) and Significance Test

ANOVA was conducted to assess the significance of the regression model, with the results summarized in

Table 3. Analysis of variance of response surface test results.

Source	Sum of Squares	df	Mean Square	F-Value	p-Value	Significance
Model	232.15	9	25.79	134.27	<0.0001	Significant
A (concentration)	12.01	1	12.01	62.49	<0.0001	**
B (operation time)	4.31	1	4.31	22.42	0.0021	**
C (infection time)	35.83	1	35.83	186.50	<0.0001	**
AB	0.1980	1	0.1980	1.03	0.3438	-
AC	1.97	1	1.97	10.28	0.0149	*
BC	0.4761	1	0.4761	2.48	0.1594	-
A2	48.42	1	48.42	252.03	<0.0001	**
B2	29.09	1	29.09	151.43	<0.0001	**
C2	82.20	1	82.20	427.90	<0.0001
Residual	1.34	7	0.1921	-	-
Lack of fit	0.7209	3	0.2403	1.54	0.3344	Not significant
Pure error	0.6239	4	0.1560	-	-	-
Cor total	233.49	16	-	-	-	-

Note: “*” indicates significance at p < 0.05, and “**” indicates high significance at p < 0.01.

. The overall model was statistically significant (p < 0.001), indicating that at least one of the model terms contributed meaningfully to the prediction of control efficacy. The lack-of-fit test was not significant (p = 0.3344 > 0.05), suggesting that the model adequately describes the observed data.

The model’s goodness of fit was evaluated using the coefficient of determination (R² = 0.9942) and adjusted R² (0.9868), indicating that 98.68% of the variability in control efficacy could be explained by the fitted model. The coefficient of variation (CV) was 0.4653%, reflecting high experimental precision. The adequate precision value of 31.0810, which far exceeds the minimum threshold of 4.0, confirmed that the model had a sufficient signal-to-noise ratio for navigating the design space.

Th analysis of regression coefficients showed that the linear terms (A: concentration, B: operation time, and C: infection time) and quadratic terms (A², B², and C²) were all highly significant (p < 0.01), suggesting substantial individual contributions to the response variable. Among the interaction terms, only AC exhibited a significant interaction effect (p < 0.05). While F-values reflect the relative variance explained by each term, we recognize they should not be solely interpreted as indicators of variable importance. Rather, significance levels and model coefficients were used to assess the relative impact of each factor, with infection time (C) showing the strongest influence on control efficacy, followed by concentration (A) and spraying time (B).

3.6.3. Model Diagnostics and Assumption Validation

To ensure the reliability of the regression model, diagnostic plots were evaluated. The normal probability plot of externally studentized residuals (Figure 6a) confirmed that residuals were approximately normally distributed, as the points followed a near-linear trend. In addition, the residuals vs. predicted values plot (Figure 6b) displayed a random scatter around zero, indicating homoscedasticity and the absence of systematic variance. These results validate the key assumptions of the regression model and support its use for further response surface analysis and optimization.

3.6.4. Interaction Analysis Based on Response Surface Methodology

The interaction effects among the three variables were evaluated using response surface plots and contour maps. In this context, the steepness of the response surface may indicate the degree of interaction between variables, provided the interaction term is statistically significant [44]. Elliptical contour plots typically suggest meaningful interactions, whereas circular contours are generally interpreted as indicative of weak or non-significant interactions [45].

Based on the regression coefficient significance analysis, all three independent variables—concentration of strong oxidizing free radicals (A), spraying duration (B), and infection time (C)—had significant quadratic effects on disease control efficacy. Figure 7 presents the response surface and contour plots for the AC interaction, along with the main quadratic effects of A, B, and C. Each of the main effect plots exhibited a curved (parabolic) surface, indicating the existence of optimal values for each factor. The AC contour plot displayed an elliptical shape, consistent with the significance of this interaction term. These findings underscore the importance of accounting for both individual and interactive effects—particularly between concentration and infection timing—when optimizing treatment efficacy. The response surface model thus provides a robust analytical framework for identifying optimal application conditions and supports the practical development of more precise and efficient disease control strategies.

3.6.5. Model Validation

The data presented in Table 3 were subjected to regression fitting and optimization analysis using Design-Expert Version 12.0, with the control efficacy against wheat powdery mildew set as the response target. The response surface optimization module identified the optimal parameter combination as follows: strong oxidative free radical concentration of 6.65 mg/L, spraying time of 50.84 s, and infection time of 15.67 h. Under these optimized conditions, the model predicted a control efficacy of 97.64%. To validate the model’s predictive performance, three independent experiments were conducted using the optimized parameters. The average observed control efficacy was 97.23%, which closely aligned with the predicted value, demonstrating the model’s robustness, accuracy, and practical applicability.

4. Discussion

The current findings provide compelling evidence for the dose-dependent antifungal efficacy of strong oxidative free radicals against wheat powdery mildew. A significant increase in disease control was observed as the concentration of oxidants rose from 3.0 to 5.0 mg/L, followed by a saturation phase beyond this threshold, with no statistically significant improvement noted at 10.0 mg/L. This saturation behavior is consistent with the oxidative ceiling reported in studies on ozone and its derived reactive oxygen species (ROS), particularly hydroxyl radicals (OH). The predominant mechanism is presumed to involve ·OH-mediated oxidative disruption of pathogen integrity, including non-specific attacks on the fungal cell wall, plasma membrane, and genetic material [46].

Operational parameters were found to significantly affect the control efficacy of the free radical-based treatment system, in addition to its concentration. Specifically, operation time and height were identified as key determinants of droplet behavior, influencing both atomization uniformity and deposition efficiency. A contact duration of 50 s and an application height of 6.5 cm yielded the most favorable interaction conditions. Temporal analysis further revealed that early-stage intervention (0–12 h post-pathogen inoculation) significantly outperformed late-stage treatments, likely due to the enhanced disruption of initial pathogen colonization processes and improved synchrony with host–pathogen interaction windows. These spatiotemporal dynamics critically affect radical–pathogen interaction interfaces, mediating effective delivery, reactivity, and biocidal performance [47].

Response surface modeling using the Box–Behnken design confirmed that a radicals concentration of 6.65 mg/L, operation time of 50.84 s, and application at 15.67 h post-infection constitute the optimal parameter set, achieving a disease control efficacy of 97.64%. The model showed high predictive accuracy (R² = 0.9942) and minimal deviation (<0.5%) between observed and predicted values, indicating strong generalizability under variable field-relevant conditions.

The present findings are well corroborated by previous studies across multiple dimensions. For instance, ozone water containing 4.0 mg/L ozone was shown to effectively suppress the spread of cucumber powdery mildew without adversely affecting leaf morphology or photosynthetic performance [48]. In the case of strawberry powdery mildew, ozone spraying achieved a lower disease index (18.52%) compared to conventional spraying (22.96%) and electrostatic spraying (21.48%) [49]. Greenhouse trials further demonstrated that regular ozone release can effectively reduce the incidence of cucumber powdery mildew, with a combined application of ozone and chemical fungicides achieving 81.9% control efficacy, significantly higher than either ozone alone (53.6%) or chemical agents alone (57.7%) [50]. Electrolyzed ozone water has also been reported to maintain initial infection levels of cucumber powdery mildew without causing physiological damage to the plants, highlighting its potential for precision disease management in early-stage interventions [51].

Beyond disease control, ozone treatment has shown potential in improving both yield and quality in greenhouse crop production. Field trials in tomato and pepper indicated that ozone application not only reduced disease incidence but also promoted crop productivity [52]. Response surface analysis of ozone treatments for tomato leaf mold and gray mold revealed that optimal combinations of concentration, duration, and frequency significantly improved both disease control and physiological indices, with gray mold control efficacy reaching 99.49% under optimal conditions, substantially outperforming conventional treatment strategies [53].

In conclusion, the strong oxidative free radicals system demonstrated promising efficacy, moderate dosage requirements, and a well-defined operational window for the control of wheat powdery mildew. Coupled with modeling-based optimization strategies, this approach offers the potential for precise and efficient field application. Future research should focus on multi-location field validation and the integration of automated spraying technologies to evaluate the adaptability and scalability of this control strategy across diverse agro-ecological zones and management regimes.

5. Conclusions

This study established an efficient and environmentally friendly disease control strategy centered on strong oxidative free radicals. Through multi-parameter optimization, the combined effects observed of key variables—including radicals concentration, application timing, pathogen infection stage, and system stability—on disease suppression were elucidated. The response surface regression model confirmed the accuracy and practical utility of the optimized protocol, providing a valuable reference for the future development and field-scale application of free radical-based plant protection products. In addition to disease suppression, we plan to conduct further experiments to systematically analyze the physiological and biochemical effects of strong oxidizing free radicals on wheat. These studies will assess plant growth, chlorophyll content, enzyme activity, and oxidative stress markers, providing a comprehensive understanding of the broader impact of the treatment on plant health. This future work will help optimize the application of strong oxidizing free radicals for sustainable agricultural management, minimizing potential plant damage while maximizing disease control efficacy.