Optimizing germination conditions of Ghaf seed using ZnO nanoparticle priming through Taguchi method analysis

Ghaf, a resilient tree in arid environments, plays a critical role in ecological restoration, desertification mitigation, and cultural heritage preservation. However, the seeds’ inherent challenges, notably their hard outer coating restricting germination, emphasize the pressing need for innovative strategies. This work aimed to investigate the optimization of Ghaf seed germination process through seed priming with ZnO nanoparticles treatment (duration (t), concentration (c), temperature (T), and agitation (a), employing the Taguchi method for efficient experimental design. Furthermore, the study includes Analysis of Variance (ANOVA), analysis for the regression model to assess the significance of predictor variables and their interactions, thereby strengthening the statistical validity of our optimization approach. Notably, it revealed that concentration is a pivotal influencer in optimization of Ghaf seed germination. The results showed that the concentration of ZnO nanoparticles has no linear relation with any other parameters. To verify the outcomes, validation tests were performed utilizing the predicted optimal parameters. The observed low error ratio, falling within the range of 1 to 6%, confirmed the success of the Taguchi methodology in identifying optimal levels of the factors chosen. Significantly, ZnO-primed seeds showcased a remarkable enhancement in Ghaf seed germination, increasing from 15 to 88%. This study introduces a novel approach utilizing ZnO nanoparticle treatment optimized through the Taguchi method, significantly enhancing seed germination rates of Ghaf seeds and offering a promising avenue for sustainable agricultural practices in arid environments.

For this experiment, a sample of Ghaf seeds weighing ten grams, equivalent to approximately (350 ± 6) seeds in triplicates, was used.Following the imbibition process, significant alterations were observed in the shape, weight, and color of the Ghaf seeds.The primary response parameter studied is the percentage of responded seeds.The numbers of responded seeds are counted in each run to calculate the percentage of responded seeds.

Identification of optimum factors and their levels
The experimental design, structured in accordance with the Taguchi orthogonal array (L16), was executed to investigate the impact of varied physico-chemical parameters on seed germination.Table 2 encapsulates the response values for mean responded seed and Signal to Noise (S/N) ratio, with all experimental trials meticulously conducted in triplicates to ensure result accuracy and reliability.Observations across the experimental runs revealed notable fluctuations in responded seed percentages.Specifically, experimental run 16 (C4T4t1A2) exhibited the minimum responded seed percentage, while conversely, experimental run 10 (C3T2t4A2) demonstrated the highest percentage at 88% (Table 3).These findings underscore the discernible influence of combined physico-chemical parameters during incubation on the percentage of responded seeds and subsequent seed germination.The analysis of these outcomes provides valuable insights into the nuanced effects of experimental conditions on critical response variables.Concentration consistently stood out as the most favorable at Level 3, showcasing the highest S/N ratio in the main effect plot (Fig. 3).Similarly, temperature displayed optimal performance at Level 3, emphasizing its significance, while time peaked at Level 2, and Agitation at Level 1.
The main effect plot and response table for S/N ratio provide a focused assessment of the optimization study outcomes 15 , specifically examining concentration, time, temperature, and agitation.In the S/N ratio table, where larger values are preferable, each factor is evaluated at four levels (1 to 4), presenting the corresponding S/N   www.nature.com/scientificreports/ratio.The Delta values represent the difference between the maximum and minimum S/N ratio for each factor, indicating the degree of improvement across levels 16 .Concentration emerges as the top performer with a Delta of 2.00 Table 4, showcasing the most significant improvement among factors.Temperature follows with notable improvement, while time and agitation show marginal changes.These response tables emphasize concentration's pronounced influence on the optimization study, seen in both S/N ratio and mean response values.The findings offer valuable guidance for prioritizing and refining optimization parameters, with concentration identified as a key driver of improvement.

ANOVA and regression analysis
Analysis of Variance (ANOVA) was employed to evaluate the effects of incubation parameters on the improvement of Ghaf seed germination rates.Results indicated that the concentration of ZnO nanoparticles played a pivotal role, contributing 92% to the observed variability in germination rates.Temperature exhibited a moderate impact at 6%, whereas both time and agitation showed minimal effects of 1% each.The residual variance, attributed to errors in measurement, accounted for the remaining 1%.These findings underscore the critical influence of ZnO nanoparticle concentration in optimizing the germination process of Ghaf seeds (Fig. 4).Table 5 presents the results of the ANOVA and regression analysis for the optimization study, offering a breakdown of the contributions of different factors to the observed variability in the response variable.The Regression analysis, comprising concentration, time, temperature, and agitation, is statistically significant (p ≤ 0.014) and accounts for 98.86% of the total variability.Concentration stands out as the primary influencer, demonstrating substantial significance (p ≤ 0.002) and contributing 91.88% to the model's explanatory power.This underscores concentration's pivotal role in the optimization process.On the other hand, time, temperature, and agitation exhibit comparatively lower impacts.Temperature contributes significantly to the model, explaining 6.12% of the variability (p ≤ 0.100), while time and agitation show minimal influence.Table 5 presented F-test results that validates the statistical significance of the regression model, highlighting concentration as the major driver and emphasizing its crucial role in explaining the observed variability in the response variable within the optimization study.

Analysis of interelation between the factors
The analysis of the contour plot (Fig. 5) reveals that concentration exhibits a lack of interaction with other parameters, such as temperature, time, and agitation.The contour lines corresponding to concentration remain consistently parallel, indicating that changes in concentration do not significantly impact the response variable differently at various levels of temperature, time, or agitation.On the contrary, a robust interaction is evident between the factors of time, temperature, and agitation.The contour lines associated with the combinations of these variables intersect and exhibit non-parallel patterns.This indicates that alterations in one of these factors have a varying impact on the response depending on the levels of the other two factors 17 .The observed strong interaction among time, temperature, and agitation underscores the complexity of their combined effects and emphasizes the need for a comprehensive understanding of their interplay in optimizing the system.Optimizing seed germination outcomes, thereby enhancing the precision and efficacy of our experimental approach.

Predicting the best factor for increased % of responded Ghaf seeds
The regression model, with a low standard error (S = 1.79699) and a high R-squared value (98.86%), effectively explains nearly 99% of the variability in responded seeds based on concentration, time, temperature, and agitation.
The regression analysis reveals a predictive model for the number of responded seeds based on the factors of concentration, time, temperature, and agitation.The regression Eq. (1): The regression equation represents the relationship between the responded seeds and the specified levels of concentration (125, 250, 500, and 1000) ng/mL, time (2, 4, 8, and 16) hrs., temperature (25, 30, 35, and 40) ͦ C, and agitation (0, 50, 100, and 150) RPM.Each coefficient indicates the contribution of the corresponding factor level to the number of responded seeds.
The Normal Probability Plot (Fig. 6) assesses the normality of the residuals from the regression analysis by comparing the residuals to their expected values under a normal distribution.In this analysis, the residuals closely www.nature.com/scientificreports/follow the red line, suggesting that the assumption of normality is reasonably met.This validation of normality enhances the credibility of our regression model, which has a high R-squared value (98.86%), indicating that the model's predictions and the statistical inferences drawn are reliable.The normal distribution of residuals is crucial for ensuring the accuracy and robustness of the Taguchi regression analysis applied in optimizing the germination conditions for Ghaf seeds.Regression equations were used to compute the anticipated percentages of responded seeds for each trial run and predict the impact of alterations in parameters on the dependent variables.The corresponding predicted values for the percentage of responded seeds in certain experimental runs are detailed in Table 4.An alternative approach for forecasting the dependent variables involves the formulation of equations through the Taguchi method.Consequently, the Taguchi-predicted values for multiple S/N ratios at the optimal factor level (ε0) were also determined using Eq. ( 2).
Here, ε0 represents the prediction, εm denotes the total mean S/N ratio, εi signifies the mean S/N ratio at the optimal level, and j corresponds to the number of input process parameters 18 .

Validation of the predicted equations and confirmatory analysis
Verification of the proposed experimental design's validity is crucial to substantiate the expected enhancement in the process response through the optimal parameters identified by the matrix test.Five runs (1, 4, 8, 12 & 16) were randomly selected from the L16 Taguchi Orthogonal Array (OA), in addition to a run utilizing the optimized levels of incubation parameters (C3t2T4a2), for the purpose of validating both the regression and Taguchi models.The predicted values from both Taguchi and regression analyses are same, demonstrating consistency in their predictions.Each experimental run was conducted in triplicate, and the average percentage of responded seeds was compared with the predicted values.Table 6 presents the comparative data between actual experimental values and predicted values for particle size and PDI, as determined by the Taguchi method and regression prediction equation.The minimal error observed (less than 6% of responded seeds) in the comparison between (2) ε0 = εm + i = 1j (εi − εm).experimental and predicted values underscores the efficacy of mathematical modeling in forecasting optimal incubation parameters for enhanced seed germination.A confirmatory analysis given in Fig. 7A illustrates the emergence of responded seeds following treatment with the optimized parameter combination of 500 ng/ml concentration, 40 °C temperature, 4 h duration, and 50 RPM agitation (C3t2T4a2).In contrast, Fig. 7B portrays the seeds treated with a water control.

Discussion
The chemically synthesized ZnO nanoparticles exhibited a crystalline structure, confirmed by XRD analysis.The Debye-Scherrer equation determined an average grain size of 42.64 nm, while scanning TEM and SEM analyses revealed a spherical morphology and a particle size range of 50-100 nm.The effectiveness of ZnO nanoparticle in enhancing seed germination is evident, yet the concentration plays a pivotal role.While these nanoparticles have been demonstrated to promote germination, excessive concentrations pose potential harm.
The experimental design is founded on the Taguchi approach, utilizing the orthogonal array to efficiently explore the interactions between factors and levels while minimizing the number of experiments 19 .The Taguchi method's orthogonal arrays enable the exploration of various parameter combinations in a relatively concise set of experimental runs.This efficiency is paramount when dealing with complex interactions among factors 20 .The method's ability to identify the most influential factors and their optimal levels is pivotal for targeted and resource-effective optimization 21 .Furthermore, Taguchi analysis provides statistical rigor, allowing for the quantification of parameter effects and the determination of statistically significant factors, enhancing the credibility of the findings.Additionally, the robustness testing inherent in the Taguchi method ensures that the identified optimal conditions remain effective in the face of minor variations or uncertainties.Ultimately, the Taguchi method serves as a powerful tool not only for understanding the factors influencing Ghaf seed germination but also for guiding the development of practical and optimized seed priming protocols with ZnO nanoparticle for real-world applications.Utilizing a full factorial design would entail conducting 256 runs in triplicate (i.e.768 runs) to pinpoint the optimal combination of selected incubation parameters.In contrast, the Taguchi orthogonal array (OA) design offers a more streamlined approach to configuring experimental parameters compared to the factorial method, significantly reducing the number of experimental runs to 16.This method not only minimizes the experimental cost and time but also maintains efficiency in exploring key factors for optimization 22 .
The S/N Ratios play a pivotal role in assessing key factors in this study, revealing essential insights into performance 23 .A higher S/N ratio signifies superior performance, and our analysis identified distinct trends 24 , aligning with our goal of enhancing the germination percentage of Ghaf seeds.This formula encapsulates the essence of our approach, providing a quantitative measure of the success of our optimization efforts.
The Delta values quantified the variation within each factor, elucidating their impact on system performance 25 .The delta (Δ) value was determined by subtracting the lowest S/N from the highest S/N for each factor.Higher delta value signifies higher variation.Hence based on the delta value the concentration of ZnO nanoparticle is the most influential parameter followed by temperature, time and agitation.The Analysis of Variance (ANOVA) , a robust statistical approach allowed for the systematic exploration of key incubation parameters, enabling a comprehensive understanding of their individual and collective impacts on the germination process 26 .ANOVA results emphasize the statistical significance of the regression model.Concentration stands out as the major influencer, explaining 91.88% of the model's explanatory power (p ≤ 0.002), underscoring its pivotal role in www.nature.com/scientificreports/ the optimization process.Conversely, time, temperature, and agitation exhibit lower impacts, with temperature contributing 6.12% (p ≤ 0.100).The high F-value and low p-value for concentration suggest a strong and statistically significant relationship between concentration levels and the variability in the response variable.Studies suggests that nanoparticle may enhance seed germination mechanisms by potentially increasing water absorption 27 , improving seeds' capacity to absorb and utilize water and nutrients 28 , elevating nitrate reductase enzyme levels 29 , bolstering seed antioxidant systems to mitigate oxidative stress by reducing superoxide radicals and malonyldialdehyde content 30 , and boosting the activities of certain enzymes such as superoxide dismutase, ascorbate peroxidase, guaiacol peroxidase, and catalase 31 .Nanoparticle concentration significantly impacts seed germination by altering biochemical processes, making it the most influential factor in the study 32 .The concentration of ZnO nanoparticles emerges as the pivotal influencer in optimizing Ghaf seed germination due to its direct and profound impact on biochemical processes critical to seed viability and growth.Unlike factors such as temperature, time, and agitation, which influence external conditions during germination, nanoparticle concentration directly affects the quantity of nanoparticles available for interaction with seed tissues.This interaction can modulate physiological processes such as water uptake, nutrient absorption, and enzymatic activities within the seed 10,33 .Higher nanoparticle concentrations typically provide more active sites for these interactions, thereby potentially enhancing seed germination rates 34 .Consequently, the concentration of ZnO nanoparticles assumes centrality in this study as it fundamentally alters internal seed processes necessary for successful germination, distinguishing it as the most influential parameter among the factors investigated.
The intricate relationships between concentration, time, temperature, and agitation are further elucidated through the developed regression equation.The linear regression analysis aimed to model these relationships, capturing any linear or polynomial dependencies among the variables.The predicted values obtained through Taguchi OA experiments, executed with Minitab 19.0 software, were systematically compared with the outcomes derived from the linear regression equations 35 .This comparison was further validated through experimental observations, and the efficacy of the developed regression models was evaluated through statistical measures.
The potential of the regression models to explain the variability observed in Ghaf seed germination rates was quantitatively assessed using the coefficient of determination (R 2 ) 36 .This methodological approach not only identified the key parameters influencing the enhancement of Ghaf seed germination but also validated the predictive capabilities of the developed regression models through rigorous statistical and experimental analyses.
The coefficients in the equation not only quantify the individual contributions of each factor level but also capture the collective impact of these factors on the responded seed count.The positive and negative coefficients signify the directionality and magnitude of these effects, offering a nuanced understanding of the optimization process 37 .The constant term (64.56) serves as a baseline, providing context for seed response when all factors are at their reference levels 38 .
The contour plot analysis adds depth to the discussion, revealing distinct patterns of interaction between factors.Notably, concentration demonstrates a lack of interaction with other parameters, maintaining parallel contour lines.This suggests that changes in concentration do not significantly alter the response variable in conjunction with variations in temperature, time, or agitation.In contrast, a robust interaction is observed among time, temperature, and agitation, as indicated by non-parallel contour lines.This emphasizes the need for a comprehensive understanding of the combined effects of these factors for effective optimization.
The predictive capabilities of the regression model are highlighted in the discussion, emphasizing its utility in anticipating the percentage of responded seeds for each experimental run.The comparison of predicted values with actual experimental outcomes, as depicted in Table 4, underscores the model's accuracy and reliability.The minimal error observed (< 6% of responded seeds) signifies the effectiveness of mathematical modeling in forecasting optimal incubation parameters, further validating the predictive power of the developed regression equation.
Moreover, the application of the Taguchi method offers a systematic and efficient approach to optimizing germination conditions.The method's ability to explore multiple factors simultaneously, using orthogonal arrays, facilitates a comprehensive understanding of the individual and collective impacts of the chosen parameters.The "larger the better" S/N ratio criterion ensures the quality of experimental outcomes, and the Delta values provide insights into the degree of improvement across factor levels.These aspects collectively contribute to the robustness and reliability of the optimization study.
This study focuses on optimizing parameters to enhance Ghaf seed germination using the Taguchi method, with the goal of establishing a green canopy in desert regions.Ghaf trees are uniquely adapted to harsh climates, making them suitable for combating desertification and increasing oxygen production.However, seed dormancy poses a significant challenge to achieving consistent germination rates and uniform plant growth.Addressing these issues is crucial as inconsistent germination patterns can lead to uneven plant growth.Given Ghaf trees' potency in harsh conditions, once established, they require minimal maintenance.This resilience underscores their potential as a sustainable solution for enhancing green cover in desert areas.By tackling seed dormancy and improving germination consistency, this initiative aims to promote a robust Ghaf tree population, contributing to environmental stability and biodiversity conservation in arid regions.

Conclusion
In conclusion, this study successfully synthesized ZnO nanoparticles using a chemical reduction method and characterized their crystalline structure and morphology.Through Taguchi method analysis, it identified concentration as the most influential factor in enhancing Ghaf seed germination, followed by temperature, time, and agitation.The regression model developed based on these factors accurately predicted the percentage of responded seeds.These findings underscore the importance of nanoparticle concentration in optimizing seed germination and highlight the efficacy of the Taguchi method in systematically exploring and optimizing incubation parameters.Moreover, the experimental validation confirms the reliability and accuracy of the developed regression model, offering valuable insights into the interplay of various factors in seed germination enhancement.Ultimately, this research contributes to advancing seed priming protocols using ZnO nanoparticles, with implications for sustainable agricultural practices in arid environments.However, it is essential to acknowledge the study's limitations.Firstly, while the Taguchi method efficiently explores interactions among factors, further studies could incorporate additional variables that may influence seed germination, such as light conditions or soil pH.Secondly, the focus on ZnO nanoparticles warrants consideration of potential ecological impacts and long-term effects on soil health.Future research could delve into these aspects to ensure sustainable agricultural practices.Additionally, expanding the study to different geographical locations and varying environmental conditions would enhance the generalizability of our findings.By addressing these aspects, future studies can build upon our findings and contribute to further advancements in seed priming methodologies using nanotechnological approaches.

Method of ZnO nanoparticle synthesis
ZnO nanoparticle was synthesized via a chemical reduction method 39 .Zinc nitrate tetrahydrate (Zn(NO 3 ) 2 •4H 2 O) served as the precursor, with trisodium citrate dihydrate (Na 3 C 6 H 5 O 7 •2H 2 O) (TSC) as the capping agent and NaOH as the reducing agent.A 0.2 M Zinc nitrate solution was prepared in 100 mL ultrapure water, and a corresponding 0.2 M TSC solution was created.The two solutions were mixed, heated to 80 °C, and stirred.In order to induce the formation of a white precipitate, 0.2 M NaOH was gradually add under stirring (50 rpm).After 3 h of stirring, the solution cooled to room temperature.The resulting precipitate underwent washing with ultrapure water three times, centrifugation at 5000 rpm for 5 min, and an additional rinsing with absolute ethanol.
Drying occurred at 70 °C for 8 h, followed by grinding for uniformity.Crystalline ZnO nanoparticle was obtained through calcination at 450 °C for 2 h.This nanoparticle was then subjected to, X-ray diffraction analysis was conducted using a Bruker diffractometer (AXS Kappa APEX II CCD X-ray), operating at 40 mA and 40 kV, with Cu Kα radiation (1.54 Å) as the source.The obtained XRD patterns were compared using JPCDS standard card No: 043-0002.Transmission electron microscopy (TEM JOEL 2100) and scanning electron microscopy (JEOL JSM-7600F FEG-SEM) analysis.The average ZnO grain size was calculated using the full width at half maximum (FWHM) of diffraction curves by the Debye-Scherrer formula (3).where D is the particle size; λ-wavelength of X-ray radiation (0.15406 nm); K -Scherrer constant (0.9); β-full width at half maximum (FWHM) of the XRD peak; θ -Bragg's angle 40 .

Plant material
The 10 g of Ghaf seeds were surface sterilized by using 100 mL of 0.1% sodium hypochlorite for 2 min, followed by a thorough triple rinse with distilled water and were gently air-dried on blotting paper at a consistent room temperature of 22 ± 1 °C.These prepared seeds were subjected to runs designed by Taguchi OA L16 according to Morfidan et al. 41 .

The experimental design using Taguchi approach
Taguchi approach was employed to select combinations of multiple factors for simultaneous experimentation on Ghaf responded seeds.The experimental design, utilizing Minitab 19.0 software, takes a systematic and multifactorial approach to comprehend the influence of ZnO nanoparticles on seed responsiveness.

Identification of factors and their respective levels
A Taguchi orthogonal array experimental design 41 was employed, incorporating four key factors (concentration, temperature, time, and agitation), each at four distinct levels as given in Table 7.For study the concentration (125, 250, 500, 1000 ng/ml).Temperature (25, 30, 35 40 °C).Time durations selected were (2, 4, 8, 16 Hrs) and agitation were (0, 50,100, and 150 RPM).These factors were chosen to encompass a range of conditions that could significantly influence the responsiveness of Ghaf seeds to the ZnO nanoparticle treatment.Four level four factor leads to the L16 runs in triplicates given in Table 2.
(3) D = k βcosθ , Table 7. Incubation parameters and their levels selected for the optimization of Ghaf seeds germination using Taguchi method.www.nature.com/scientificreports/

Incubation parameters Abbreviation
The optimization criterion for the experiments is the larger-the-better S/N ratio, aligning with the objective of maximizing the germination response 42 .The formula utilized for this assessment is defined (4) 43 .
Y represents the measured response or output for each specific experimental run, while n denotes the total number of observations.

Analysis of variance
Analysis of variance (ANOVA) was employed using the Taguchi orthogonal array (OA) methodology to identify the most and least influential parameters affecting the enhancement of Ghaf seed germination rates 44 .

Regression model development
Regression analysis, a pivotal component in discerning the relationships between the response variable (germination rate) and selected incubation parameters, was carried out to develop predictive mathematical models 45 .In this investigation, linear regression analysis was utilized, implemented through statistical software (e.g.Minitab 19.0) 46 .The chosen input parameters for the regression models included concentrations of ZnO nanoparticles, treatment duration, temperature, and agitation levels during the seed priming process.

Contour plot and confirmatory analysis
Contour plots were created using the Taguchi approach to explore the intricate relationships between each pair of conditions, providing a visual representation of their interrelation 47 .To validate the findings, a confirmation test was executed by conducting an experiment with the predicted optimum levels for each condition, as determined by the program.The experiment was replicated three times, and the results were analyzed to calculate the error rate of the conducted experiment.

Figure 2 .
Figure 2. (a) SEM analysis of chemically synthesized ZnO nanoparticle using JEOL JSM-7600F FEG-SEM at 20 kV with scale bar of 200 nm (b) TEM analysis of chemically synthesized ZnO nanoparticle by JOEL 2100 at 10kv with a scale bar of 50 nm.

Figure 3 .
Figure 3. Main effect plot for S/N ratio of concentration, time, temperature and agitation for on percentage of responded Ghaf seed.

Figure 4 .
Figure 4. ANOVA analysis for percentage of contribution of concentration, time, temperature, and agitation on percentage of responded Ghaf seeds.

( 1 )Figure 5 .
Figure 5. Two-dimensional contour plots illustrate the influence of concentration, time, temperature and agitation on percentage of responded Ghaf seeds.

Figure 6 .
Figure 6.Normal probability plot of residuals for the regression analysis of the percentage of responded seeds to concentration, time, temperature, and agitation.

Table 1 .
Crystallite size of synthesized ZnO nanoparticles as determined from XRD analysis.

Table 2 .
The selected experimental trials designed using the Taguchi method for enhancing the germination of Ghaf seeds using ZnO nanoparticle.

Table 3 .
Experimental design using Taguchi L16 orthogonal array design (OA) and the percentage of responded Ghaf seeds and Signal-noise ratio for each factor level combination designed.

Table 4 .
Mean signal-noise ratio by factor level for concentration, time, temperature and agitation on percentage of responded Ghaf seeds.

Table 5 .
ANOVA output on contribution of concentration, time, temperature, and agitation on percentage of responded Ghaf seeds (at 95% confidence level).

Table 6 .
Predicted values and confirmation test results by Taguchi method for random runs and optimized combination levels for the percentage of responded Ghaf seeds.C3t2T4a2: combination of 500 ng/ml concentration, 40 °C temperature, 4 h duration, and 50 RPM agitation.