Integrated LC–MS/MS assessment of cyprodinil and fludioxonil residues in greenhouse strawberries: Dissipation, dietary risk and washing removal under Saudi Arabian conditions

2.1. Chemicals and reagents

Cyprodinil and fludioxonil reference standards, each with a purity of 99.5%, were obtained from Chem Service Inc. (West Chester, PA, USA). The organic solvents used for chromatographic analysis, namely acetonitrile and methanol, together with glacial acetic acid, were purchased from Fisher Scientific (Loughborough, UK). Anhydrous magnesium sulfate, sodium acetate, and sodium bicarbonate were obtained from Chem-Lab NV (Zedelgem, Belgium). Sample clean-up was performed using PSA sorbent (Chromabond Diamino, primary–secondary amine), supplied by Macherey-Nagel (Düren, Germany), and multi-walled carbon nanotubes, supplied by Shilpa Enterprises (Shilpent, Maharashtra, India). A ceramic homogenizer from Agilent Technologies Inc. (Wilmington, DE, USA) was used during sample preparation. The formulation applied in the field experiment was a commercially available 62.5% WG water-dispersible granule product containing 37.5% cyprodinil and 25% fludioxonil. The formulation was purchased from the local market and produced by Syngenta (Basel, Switzerland). Ultrapure water was prepared in-house using an Ultra-Clear purification system manufactured by Evoqua Water Technologies GmbH (Günzburg, Germany).

2.2. Standard solutions

Individual primary standard solutions of cyprodinil and fludioxonil were prepared in acetonitrile at 1000 mg L⁻¹. Briefly, 50.2 mg of each reference standard was weighed separately, transferred to a 50 mL volumetric flask, diluted to volume with acetonitrile, and purity-corrected. A mixed intermediate standard solution containing both analytes at 100 mg L⁻¹ was prepared from the corresponding primary solutions. Subsequent dilution of this mixture with acetonitrile provided a working standard solution containing 10 mg L⁻¹ of each analyte. This working mixture was used to prepare both solvent-based calibration solutions and matrix-matched calibration solutions in blank strawberry extract. The calibration range covered eight levels from 1 to 200 µg L⁻¹. All solutions were stored at −20°C before analysis.

2.3. Extraction and cleanup

Extraction and clean-up were performed using the acetate-buffered QuEChERS method, with minor modifications [16]. Briefly, 10.0 g of homogenized frozen strawberry sample was weighed into a 50 mL polypropylene centrifuge tube, and 10 mL of acetonitrile containing 1% (v/v) acetic acid was added. The tube was shaken manually for 2 min. Subsequently, 4.0 g of anhydrous MgSO₄ and 1.0 g of anhydrous sodium acetate were added, together with a ceramic homogenizer to facilitate mixing, and the tube was immediately shaken vigorously for 1 min. The extract was then centrifuged at 5000 rpm for 5 min.

For dispersive SPE clean-up, a 2.0 mL aliquot of the supernatant was transferred into a 15 mL centrifuge tube containing 300 mg MgSO₄, 50 mg PSA, and 5 mg MWCNTs. The mixture was vortex-mixed for 30 s and centrifuged at 5000 rpm for 5 min. The cleaned extract was filtered through a 0.22 µm syringe filter into an LC vial before LC–MS/MS analysis.

2.4. LC–MS/MS analysis

A Dionex UltiMate 3000 RS UHPLC system coupled to a TSQ Altis triple-quadrupole mass spectrometer (Thermo Fisher Scientific, Austin, TX, USA) was used for the determination of cyprodinil and fludioxonil. Chromatographic separation was achieved on an Accucore™ Q C18 analytical column (2.6 µm, 2.1 × 150 mm), with the column oven maintained at 40°C. The chromatographic system employed two mobile phases: mobile phase A consisted of deionized water/methanol (95:5, v/v) containing 5 mM ammonium formate and 0.1% formic acid, whereas mobile phase B consisted of methanol/deionized water (95:5, v/v) containing the same additives. Analysis was performed at a flow rate of 0.30 mL min⁻¹ with an injection volume of 2 µL. The elution program was as follows: 10% B for the first 1 min, followed by a linear increase to 90% B from 1 to 4 min, held at 90% B from 4 to 9 min, returned to 10% B between 9.0 and 9.1 min, and re-equilibrated at 10% B until 18 min.

MS detection was performed in multiple-reaction monitoring (MRM) mode with automatic polarity switching. Cyprodinil was recorded under positive electrospray ionization (ESI⁺) conditions, whereas fludioxonil was recorded under negative electrospray ionization (ESI⁻) conditions. Two transitions were monitored for each compound. The source settings were as follows: sheath gas, 40 arb; auxiliary gas, 10 arb; sweep gas, 10 arb; ion-transfer tube temperature, 325°C; vaporizer temperature, 350°C; spray voltage, 3.5 kV in positive mode and 2.5 kV in negative mode. TraceFinder™ software, version 4.1, was used for instrument control, signal acquisition, and quantification. Detailed transition data, collision energies, dwell times, ion ratios, and retention times are presented in Table S1.

2.5. Method validation

Method validation of the LC–MS/MS method in strawberry fruit was performed according to the requirements of SANTE/11312/2021 [12]. Blank strawberry extracts were analyzed to confirm method selectivity and to ensure the absence of interfering peaks at the retention times corresponding to cyprodinil and fludioxonil. Identification reliability was established by matching retention times and verifying that qualifier/quantifier ion ratios complied with the SANTE acceptance criteria. For quantification, calibration was performed using matrix-matched standards prepared in blank strawberry extract.

An eight-level matrix-matched calibration curve was used to assess linearity over the range of 0.001–0.20 mg kg⁻¹, and linearity was evaluated using the coefficient of determination (R²). The limit of detection (LOD) was estimated from low-level matrix-matched extracts as the concentration producing an identifiable chromatographic peak with a signal-to-noise ratio of approximately 3 for the quantifier transition. The limit of quantification (LOQ) was defined as the lowest validated fortification level that fulfilled the SANTE requirements for identification, trueness, and precision. In this study, LOQ verification was performed at 0.005 mg kg⁻¹ using six replicate fortified samples for each analyte. Trueness was assessed through recovery experiments using blank strawberry homogenate fortified at 0.005, 0.01, 0.1, and 1 mg kg⁻¹; six replicate samples were analyzed at each fortification level after the complete extraction and clean-up procedure. Precision at the LOQ was expressed as intra-day repeatability (n = 6) and inter-day within-laboratory reproducibility (n = 18 across three days). Matrix effect was determined from the difference between the slopes of the matrix-matched and solvent-based calibration curves according to Eq. (1):

where $S_{\text{matrix}}$ and $S_{\text{solvent}}$ are the slopes of the matrix-matched and solvent-based calibration curves, respectively.

2.6. Greenhouse dissipation trial

A supervised greenhouse dissipation trial was conducted on strawberry (Fragaria × ananassa) in Al-Baha, Saudi Arabia, in January 2026. A commercial water-dispersible granule formulation containing cyprodinil and fludioxonil (62.5% WG; 37.5% cyprodinil and 25% fludioxonil) was applied using a calibrated backpack sprayer. Two application rates were tested: 0.8 and 1.2 kg formulation ha⁻¹. The spray volume was 800 L ha⁻¹. Based on this spray volume, the corresponding spray concentrations were 375 mg L⁻¹ cyprodinil and 250 mg L⁻¹ fludioxonil at 0.8 kg ha⁻¹, and 562.5 mg L⁻¹ cyprodinil and 375 mg L⁻¹ fludioxonil at 1.2 kg ha⁻¹. These application rates were equivalent to 0.30 and 0.45 kg a.i. ha⁻¹ for cyprodinil and 0.20 and 0.30 kg a.i. ha⁻¹ for fludioxonil, respectively.

Each treatment was conducted in three separate greenhouses, and each greenhouse was considered an independent experimental replicate. The treatments were therefore physically separated to minimize the possibility of spray drift or cross-contamination between replicates. An additional greenhouse was sprayed only with water and served as the untreated control to assess background contamination and obtain control samples. Strawberry plants were grown on raised beds, with approximately 0.30 m spacing between plants within rows and 1.0–1.2 m between beds. Greenhouse temperature was maintained at 20–25°C during the day and 12–15°C at night, with relative humidity of 60–80%; these conditions were monitored throughout the trial. Fruit samples, 2.0 kg from each plot at each sampling time, were collected at 0 days, approximately 2 h after application, and at 1, 3, 7, 10, and 14 days after spraying. At each sampling time, fruits were randomly collected from multiple plants within each plot and combined into a single composite sample per plot. The samples were placed in polyethylene bags, transported to the laboratory under temperature-controlled conditions, stored at −20°C, and extracted the following day.

To relate the single-application dissipation data to residue compliance, an indicative pre-harvest interval (PHI) was calculated from the single first-order (SFO) model according to Eq. (2):

where $C_{0,\text{measured}}$ is the measured initial residue concentration and k is the first-order dissipation rate constant.

2.7. Terminal residues

Terminal-residue trials were conducted using the same commercial formulation of cyprodinil + fludioxonil under two spray regimes: either 2 or 3 applications. For each regime, the formulation was applied at 0.8 and 1.2 kg formulation ha⁻¹. A 14-day interval was maintained between consecutive applications in all repeated-application treatments.

Strawberry fruit samples, 2.0 kg from each plot at each sampling time, were collected at 3, 7, and 10 days after the second and third applications at both dose levels. At each sampling time, fruits were randomly collected from multiple plants within each plot and combined into a single composite sample per plot, as described for the dissipation trial. The samples were placed in polyethylene bags, transported to the laboratory under temperature-controlled conditions, stored at −20°C, and extracted the following day. The resulting terminal-residue data were used as inputs for the dietary risk assessment.

2.8. Dietary risk assessment

Chronic and acute dietary exposure to cyprodinil and fludioxonil from strawberries was assessed deterministically using pre-washing terminal-residue data generated in the greenhouse trial. For chronic exposure, supervised-trial median residues (STMRs) were calculated at 3, 7, and 10 days after the second and third applications using replicate residue values obtained at each sampling time (n = 3). These STMR values were used to assess chronic dietary exposure in adults and children.

Two Saudi consumption scenarios were applied: 70.9 g day⁻¹, based on consumption estimate reported by Moradi-Lakeh et al. [17], and 112 g day⁻¹, derived from KSAWHS 2019 fruit servings per day and converted using an 80 g standard serving size [18,19]. Consumption values were converted to kg day⁻¹ by dividing by 1000. Body weights were set at 70 kg for adults and 12 kg for children, in accordance with EFSA default values for dietary exposure assessment [20]. The child body weight was used as a conservative screening assumption for young consumers; using a higher child body weight would proportionally reduce the estimated exposure values.

Chronic risk was expressed as a percentage of the acceptable daily intake (%ADI) according to Eq. (3):

where C is the STMR (mg kg⁻¹), IR is the ingestion rate (kg day⁻¹), BW is body weight (kg), and ADI is the acceptable daily intake (mg kg⁻¹ bw day⁻¹). The ADI values used were 0.03 mg kg⁻¹ bw day⁻¹ for cyprodinil and 0.37 mg kg⁻¹ bw day⁻¹ for fludioxonil (European Commission, 2006, 2007).

Acute dietary exposure was additionally assessed using the highest-residue (HR) values from the terminal-residue dataset. Acute exposure was expressed as estimated short-term intake (ESTI) and as a percentage of the acute reference dose (%ARfD). ESTI was calculated according to Eq. (4):

where HR is the highest residue concentration (mg kg⁻¹), IR is the ingestion rate (kg day⁻¹), and BW is body weight (kg). Acute risk was then expressed as %ARfD according to Eq. (5):

where ARfD is the acute reference dose (mg kg⁻¹ bw). The ARfD values used were 2.0 mg kg⁻¹ bw for cyprodinil [21] and 1.0 mg kg⁻¹ bw for fludioxonil [22]. Chronic and acute exposure estimates below 100% of the corresponding toxicological reference value were considered acceptable. For contextual comparison, EFSA PRIMo revision 3 was used to benchmark dietary exposure under predefined European dietary assumptions using the same residue inputs [23].

2.9. Washing treatments

Washing experiments were performed to evaluate the removal of cyprodinil and fludioxonil residues from strawberries after application. Samples collected 1 day after spraying from the high-rate treatment were used. The washing media were tap water, acetic acid solutions at 1% and 5% (v/v), and sodium bicarbonate solutions at 1% and 5% (w/v). For each treatment, 500 g of strawberries were immersed in 5 L of washing medium at a solution-to-sample ratio of 10:1 (v/w) for either 1 or 5 min, with gentle manual agitation. Each treatment was performed in triplicate, and each replicate was prepared independently.

Ultrasonic-assisted washing was also evaluated in tap water using the same solution-to-sample ratio and contact times. Strawberries (500 g) were immersed in 5 L of tap water and treated for 1 or 5 min in an ultrasonic water bath (model 3510E-DTH, Branson Ultrasonics Corporation, Danbury, CT, USA) operated at 42 kHz and room temperature (25°C).

The washing treatments were selected to represent household-relevant removal conditions with different mechanisms. Tap water represented neutral rinsing; acetic acid and sodium bicarbonate represented acidic and mildly alkaline washing media, respectively; and ultrasonic washing represented a physically assisted treatment previously used to reduce pesticide residues in strawberries and other fresh produce [13,24]. The 1- and 5-min contact times were used to compare a short, rinse-like exposure with a longer household soaking period. The 5% acetic acid treatment was included as an upper-acid scenario for comparison, but its practical interpretation was approached with caution because acidic treatments may affect the sensory quality of strawberries [25].

After washing, fruits were drained and surface-dried under laboratory conditions before extraction. Removal efficiency was calculated by comparing residue concentrations in washed samples with those in the corresponding unwashed control collected on the same sampling day and from the same treatment. Residue concentrations are reported as mean ± SD (n = 3). Washing removal (%) was calculated according to Eq. (6):

where $C_{unwashed}$ and $C_{washed}$ are the mean residue concentrations measured in unwashed and washed strawberries, respectively. Treatment effects were evaluated using a two-way ANOVA with washing condition and contact time as fixed factors, separately for each pesticide. Tukey’s HSD test was applied when significant effects were observed. Statistical significance was set at $p<0.05$.

3.1. MS/MS optimization

MS/MS optimization was performed by direct infusion of individual standard solutions using a Harvard syringe pump. For each pesticide, a 1 mg L⁻¹ solution in methanol–water (50:50, v/v) was infused to assess ionization polarity and to optimize precursor/product ion transitions, collision energies, and RF lens settings.

Both ionization modes were examined to identify the conditions producing the strongest and most informative product ions. In ESI+, fludioxonil gave a weaker response and was therefore not selected for routine analysis (Figure S1). In contrast, fludioxonil showed a markedly better response in ESI−, producing abundant product ions at m/z 180.00 and 126.00 from the precursor ion at m/z 246.95 (Figure S1). The transition 246.95 → 180.00 gave the highest response and was selected as the quantifier transition, with a collision energy of 29.9 V, whereas 246.95 → 126.00 was selected as the qualifier transition, with a collision energy of 32.1 V. The optimized RF lens voltage was 103.0 V (Table S1; Figure S2).

For cyprodinil, the most reliable fragmentation was obtained in ESI⁺, where the precursor ion at m/z 226.09 generated intense product ions at m/z 118.00 and 93.07 (Figure S1). The transition 226.09 → 118.00 showed the highest response and was selected as the quantifier transition, with a collision energy of 31.0 V, whereas 226.09 → 93.07 was selected as the qualifier transition, with a collision energy of 35.5 V. The optimized RF lens voltage was 72.1 V (Table S1; Figure S1). Under the tested conditions, cyprodinil did not provide a suitable response in ESI⁻ and was therefore quantified in ESI⁺ mode (Figure S2).

Transitions were selected based on strong signal intensity while retaining a robust qualifier ion for confirmation. The final LC–MS/MS method was operated in polarity-switching mode within a single injection, monitoring cyprodinil in ESI⁺ and fludioxonil in ESI⁻ with the optimized transitions, collision energies, RF lens settings, ion ratios, and retention times (Table S1; Figures S1 and S2).

3.2. LC optimization

Mobile-phase optimization was carried out in two steps: selection of the organic modifier, followed by evaluation of additives in the selected system (Figure S3). In the initial comparison, methanol–water produced higher responses than acetonitrile–water for both analytes, with 1.71-fold and 1.25-fold increases for fludioxonil and cyprodinil, respectively. Methanol was therefore selected for further work. Despite the higher signal, methanol–water without additives showed noticeable injection-to-injection variation in peak area, reducing method robustness across replicate injections. This behavior is consistent with LC–ESI–MS, where electrospray response can be influenced by small changes in eluent composition [26] while co-eluting matrix components may cause ion suppression within the analyte elution window [11].

Additive screening in the methanol–water system showed that 0.1% formic acid alone reduced the response of both analytes relative to methanol–water without additives, whereas the buffered system containing 0.1% formic acid and 5 mM ammonium formate substantially improved the signal compared with formic acid alone, with 2.48-fold and 11.48-fold increases for fludioxonil and cyprodinil, respectively (Figure S3). This buffered condition was therefore selected as the final mobile phase because it provided a more stable ionic environment and improved run-to-run consistency under polarity-switching conditions, with cyprodinil monitored in ESI⁺ and fludioxonil in ESI⁻. This behavior is consistent with established electrospray ionization principles, whereby mobile-phase pH, ionic composition, and volatile additives can influence ion formation and signal stability [26,27], while co-eluting matrix components may further affect analyte response through ion suppression or enhancement [11].

3.3. Method validation

Method validation was performed on strawberry fruit in accordance with SANTE/11312/2021, and the main performance characteristics are summarized in Tables S2 and S3. Representative MRM chromatograms of fludioxonil and cyprodinil in spiked strawberry extract at 1 µg kg⁻¹ and in recovery samples fortified at 5 µg kg⁻¹ are shown in Figure 2.

Close

Figure 2. Representative MRM chromatograms of fludioxonil and cyprodinil in spiked strawberry extract at 1 µg kg⁻¹ and in recovery samples at 5 µg kg⁻¹ during method validation.

Export to PPT

3.4. Dissipation of cyprodinil and fludioxonil in strawberries

3.4.2. Dissipation kinetics

Field dissipation reflects the combined effects of chemical transformation and physical loss processes, including volatilization, wash-off, tissue penetration, and growth dilution, rather than a single degradation pathway [29,30]. In the present greenhouse trial, concentration–time data for cyprodinil and fludioxonil were fitted to the single first-order (SFO) model, $C_t=C_0{e}^{-kt}$. The SFO model provided a good description of the observed dissipation patterns, as indicated by the linear relationship between log₁₀-transformed residue concentrations and time and by the high coefficients of determination (R² = 0.9583–0.9727) (Table 2; Figures 3a and b).

Table 2. Single first-order (SFO) kinetic parameters for the dissipation of cyprodinil and fludioxonil in strawberries under two treatment levels.

Pesticide	Dose (g a.i. ha⁻1)	C₀ (mg kg 1)	k (d⁻1)	Slope	Intercept	R2	t1/2 (days)	MRL (mg kg -1)	PHI* (days)
Cyprodinil	300	0.593	0.3170	−0.1376	−0.1254	0.9711	2.187	0.100	5.615
	450	0.955	0.2940	−0.1277	+0.0300	0.9727	2.358	0.100	7.675
Fludioxonil	200	0.371	0.2660	−0.1155	−0.3357	0.9583	2.606	0.050	7.535
	300	0.510	0.2702	−0.1173	−0.2150	0.9658	2.565	0.050	8.595

* Kinetic PHI estimate (days)

Close

Figure 3. Single first-order (SFO) dissipation plots for cyprodinil ((a); 300 and 450 g a.i. ha⁻¹) and fludioxonil (b); 200 and 300 g a.i. ha⁻¹) residues in strawberries.

Export to PPT

The selection of the SFO model was further supported by comparison with the biphasic DFOP model (Table S4). Across all four dissipation datasets, DFOP did not reduce RMSElog relative to SFO. In contrast, AICc values were consistently lower for SFO, with ΔAICc values of 40.0 in all cases. These results indicate that the additional parameters of the biphasic model were not supported by the available dissipation data. Therefore, SFO was retained as the most appropriate parsimonious model for describing residue decline under the tested greenhouse conditions.

Table S4

Download

Cyprodinil dissipated more rapidly than fludioxonil under the tested conditions. The dissipation rate constant (k) ranged from 0.2940 to 0.3170 d⁻¹ for cyprodinil, corresponding to half-lives ($t_{1/2}$) of 2.187–2.358 days. For fludioxonil, k ranged from 0.2660 to 0.2702 d⁻¹, with corresponding half-lives of 2.565–2.606 days (Table 2). These kinetic results are consistent with the residue profiles presented in Table 1, where cyprodinil dissipation reached 76.95–84.25% by day 7, compared with 71.16–73.75% for fludioxonil.

The slower decline of fludioxonil relative to cyprodinil is consistent with differences in physicochemical and use-related properties, while recognizing that field dissipation integrates several concurrent transformation and transfer processes [29,30]. Cyprodinil has a higher vapor pressure than fludioxonil is a low-volatility, predominantly contact fungicide [31], These differences may have contributed to the more rapid apparent decline of cyprodinil from exposed fruit surfaces under warm, well-ventilated greenhouse conditions. However, the individual contributions of volatilization, tissue penetration, and other dissipation processes cannot be distinguished from residue-concentration data alone. Despite their different decline rates, the dissipation of both fungicides was adequately described by the SFO model (Figure 3; Table 2).

The half-lives obtained in this study fall within the range reported for strawberries and other crop systems, although persistence varies with crop type, formulation, and environmental conditions. Previous studies on strawberries have reported dissipation rates and half-lives for cyprodinil and fludioxonil under field and greenhouse production conditions [14,15,32]. Longer persistence of cyprodinil has been reported under open-field conditions than under greenhouse conditions, indicating an important role of microclimate in dissipation behavior [32]. Differences in the persistence of cyprodinil and fludioxonil have also been reported in other crops, including greenhouse-grown tomato leaves [33]. These comparisons indicate that dissipation behavior should be interpreted in relation to site conditions, crop characteristics, formulation, and greenhouse microclimate. Accordingly, the fitted kinetic models should be regarded as empirical descriptions of the observed decline in residue rather than as direct evidence of a single degradation mechanism [29,30].

3.5. Terminal residues

Terminal residues, expressed as mean concentrations at 3, 7, and 10 days after the second and third applications, are summarized in Table 3. These residues represent the decline phase most relevant to fixed pre-harvest intervals and dietary exposure assessment [23]. After the second application, cyprodinil residues declined from 0.338–0.584 mg kg⁻¹ at day 3 to 0.033–0.046 mg kg⁻¹ at day 10 across the two treatment levels. Over the same interval, fludioxonil residues declined from 0.513–0.549 mg kg⁻¹ at day 3 to 0.094–0.113 mg kg⁻¹ at day 10. After the third application, cyprodinil residues ranged from 0.428 to 0.757 mg kg⁻¹ at day 3 and declined to 0.054–0.092 mg kg⁻¹ by day 10. Fludioxonil residues ranged from 0.563 to 0.924 mg kg⁻¹ at day 3 and decreased to 0.130–0.186 mg kg⁻¹ at day 10. Across both spray regimes, measurable residues remained at day 10, and higher terminal residues were observed at the higher treatment level for both compounds.

At the same post-application interval, fludioxonil usually showed higher terminal residues than cyprodinil, particularly after the third application. This pattern is consistent with its slower apparent decline observed in the dissipation study and with previous reports describing the comparatively persistent behavior of fludioxonil in strawberries and other greenhouse-grown crops [14,15,33]. The increase in residues after the third application, relative to those at the same treatment level in the second application, indicates residue carryover between sequential sprays. Nevertheless, residues declined substantially between days 3 and 10 in all treatments. Under repeated applications, cyprodinil residues declined below the applied MRL by day 10 at both dose levels. In contrast, fludioxonil residues remained above the applied MRL at day 10, indicating that the kinetic PHI derived from the single-application trial was shorter than the interval suggested by the repeated-application terminal-residue data for fludioxonil.

The SFO-derived PHI and the repeated-application terminal residues describe different aspects of residue behavior. The PHI calculated from SFO kinetics was based on a single-application dissipation curve and should therefore be considered an indicative kinetic estimate. By contrast, terminal residues after two and three applications better represent practical greenhouse use because they include repeated deposition and possible residue carryover between applications. Therefore, the terminal-residue data were considered more relevant for practical PHI interpretation and dietary exposure assessment, while the SFO-derived PHI was used to describe the dissipation trend after a single application.

3.6. Dietary risk assessment

Chronic and acute dietary risk outcomes for cyprodinil and fludioxonil are presented together in Table 4. The assessment was based on pre-washing terminal-residue data; therefore, the exposure estimates reflect residue intake before any household-washing mitigation. For chronic exposure, STMR values were used, whereas HR values were used for acute exposure. Under both Saudi intake scenarios, all estimates of chronic and acute exposure remained well below 100% of the corresponding toxicological reference values.

For cyprodinil, chronic exposure was higher at the higher application rate and decreased with increasing post-application interval. Under Scenario 1, adult %ADI values ranged from 0.10 to 2.38%, whereas child %ADI values ranged from 0.58 to 13.86%. Under Scenario 2, the corresponding values ranged from 0.16 to 3.75% in adults and from 0.92 to 21.90% in children. The highest chronic exposure to cyprodinil was observed 3 days after the third application at 450 g a.i. ha⁻¹, corresponding to an STMR of 0.704 mg kg⁻¹ and %ADI values of 2.38 and 13.86% for adults and children under Scenario 1, and 3.75 and 21.90% under Scenario 2.

For fludioxonil, chronic exposure estimates were markedly lower than those for cyprodinil, primarily due to its higher ADI. Under Scenario 1, adult and child %ADI values ranged from 0.02 to 0.23% and from 0.13 to 1.36%, respectively. Under Scenario 2, the corresponding values ranged from 0.04 to 0.37% in adults and from 0.21 to 2.14% in children. The highest chronic exposure to fludioxonil was recorded 3 days after the third application at 300 g a.i. ha⁻¹, with an STMR of 0.849 mg kg⁻¹ and %ADI values of 0.23 and 1.36% for adults and children under Scenario 1, and 0.37 and 2.14% under Scenario 2.

Acute dietary exposure was also low for both fungicides when HR values were used. For cyprodinil, the highest acute exposure occurred 3 days after the third application at 450 g a.i. ha⁻¹, where the HR was 0.954 mg kg⁻¹. The corresponding %ARfD values were 0.05 and 0.28% for adults and children under Scenario 1, and 0.08 and 0.45% under Scenario 2. For fludioxonil, the highest acute exposure was observed 3 days after the third application at 300 g a.i. ha⁻¹, where the HR was 1.203 mg kg⁻¹. The corresponding %ARfD values were 0.12 and 0.71% for adults and children under Scenario 1, and 0.19 and 1.12% under Scenario 2.

Across both compounds, child exposure estimates were consistently higher than adult estimates because of the lower body weight used in the calculation. Nevertheless, even under this conservative young-child scenario, all estimates of chronic and acute exposure remained below the corresponding toxicological reference values. The higher Saudi intake scenario also produced higher exposure estimates than Scenario 1, as expected from the higher consumption rate. Nevertheless, all %ADI and %ARfD values remained far below 100%, indicating low chronic and acute dietary concern for cyprodinil and fludioxonil residues measured at 3, 7, and 10 days after the last application under the tested greenhouse conditions. For contextual comparison, EFSA PRIMo benchmarking using the same residue inputs gave exposure estimates within the same low-risk interpretation, with differences reflecting the predefined consumption and body-weight assumptions used in the model [23].

3.7. Washing removal of cyprodinil and fludioxonil from strawberries

Washing reduced cyprodinil and fludioxonil residues to different extents (Figure 4). Cyprodinil showed relatively limited removal, ranging from 20.96 to 42.41%. The highest reduction was obtained after 5 min of tap-water washing (42.41 ± 5.17%). Similar reductions were observed with 1% sodium bicarbonate for 1 min (40.49 ± 8.19%) and 5% acetic acid for 5 min (39.29 ± 5.79%). These results indicate that a considerable proportion of cyprodinil residues was not easily removed by short-contact washing.

Close

Figure 4. Removal (%) of cyprodinil and fludioxonil residues from strawberries by household washing treatments. Error bars represent SD (n = 3). Different letters indicate significant differences among washing treatments within each pesticide at p < 0.05 according to Tukey’s HSD test. Treatments sharing the same letter are not significantly different.

Export to PPT

Fludioxonil was removed more efficiently than cyprodinil under all tested conditions, with removal ranging from 65.10 to 94.22%. Tap-water washing increased fludioxonil removal from 65.10 ± 9.17% after 1 min to 93.70 ± 11.14% after 5 min. High removal was also obtained with acidic and alkaline washing solutions, reaching 94.22 ± 13.36% after treatment with 1% acetic acid for 5 min. Turkey’s HSD grouping showed no significant differences among fludioxonil washing treatments, as all treatments shared the same significance letter (Figure 4). This suggests that fludioxonil residues were readily removable under most of the tested conditions.

The washing treatments were selected to compare simple household rinsing with acidic, alkaline, and physically assisted washing. Previous studies have shown that bicarbonate- or vinegar-based washing can improve the removal of surface pesticide residues from fresh produce, although the effect varies with the pesticide, commodity, contact time, and the extent of residue penetration into surface tissues [24]. Ultrasonic cleaning has also been reported to reduce pesticide residues in strawberries, but it is less representative of routine household practice than tap-water washing [13]. In the present study, 5% acetic acid and 5% sodium bicarbonate were included to assess more aggressive washing conditions. These treatments should be interpreted with caution for strawberries, as they are soft and perishable, and acidic treatments may affect sensory quality, particularly odor [25]. Therefore, the washing results should be considered in terms of both residue reduction and practical suitability for fresh strawberries.

The statistical analysis confirmed the compound-dependent washing response. For cyprodinil, removal was significantly affected by contact time and by the interaction between washing treatment and contact time, indicating that the effect of washing duration depended on the washing medium used. Tukey’s HSD test showed that tap-water washing for 5 min, 1% acetic acid for 5 min, 5% acetic acid for 5 min, 1% sodium bicarbonate for 1 min, and 5% sodium bicarbonate for 1 min belonged to the higher removal group. In contrast, tap-water washing for 1 min and ultrasonic-assisted washing for 1 min were among the least effective treatments. For fludioxonil, no significant differences were observed among washing treatments at p < 0.05, although removal was consistently high across most treatments.

The different washing responses are consistent with the physicochemical behavior and surface location of the two fungicides. Both compounds are hydrophobic, with log P values of approximately 4.0 for cyprodinil and 4.12 for fludioxonil [31], which favors their association with the waxy cuticle. However, washing efficiency cannot be predicted solely from aqueous solubility. Cyprodinil is a weak base (pKa 4.44) with an aqueous solubility of approximately 13 mg L⁻¹ at pH 7 and 20°C, whereas fludioxonil has a lower aqueous solubility of 1.8 mg L⁻¹ under the same conditions and is essentially non-ionized at food-relevant pH [31]. The partly systemic behavior of cyprodinil may allow a greater fraction to penetrate or redistribute into the cuticle and shallow fruit tissues, thereby reducing the amount readily available for removal by washing. In contrast, the predominantly contact behavior of fludioxonil may leave a larger proportion of the residue at or near the fruit surface. Its greater removal in the present study is therefore more plausibly attributed to the physical detachment of surface-associated deposits than to dissolution in the washing medium. This interpretation is supported indirectly by radiolabel studies in grapes, in which approximately 57% of the total radioactive residue remained on the external fruit surface at maturity [34]. However, because residue distribution was not directly measured in strawberries in the present study, this explanation should be regarded as a plausible mechanistic interpretation rather than definitive evidence.

Using the pre-washing residues measured 1 day after application at the higher treatment level, cyprodinil and fludioxonil were detected at 0.710 and 0.398 mg kg⁻¹, respectively (Table 1). Based on the EU Pesticides Database entry for strawberries, the applicable MRLs were 0.10 mg kg⁻¹ for cyprodinil and 0.05 mg kg⁻¹ for fludioxonil [35]. According to the observed removal ranges, cyprodinil residues remained above the MRL even after the most effective washing treatment. In contrast, fludioxonil residues could be reduced below the MRL under the most effective washing treatments, although this was not achieved under all tested conditions. In this study, dietary exposure assessment was based on pre-washing residues, whereas washing was treated as a post-harvest mitigation step rather than as an input to the exposure calculation.