Green analytical quality by design RP-HPLC technique with fluorimetric detection for simultaneous assay of cinnarizine and dompridone

A green and sensitive factorial design-assisted reversed-phase high-performance liquid chromatography (RP-HPLC) approach with fluorimetric detection has been developed for simultaneous determination of cinnarizine and domperidone in pure materials, commercial single and combined tablets. Both drugs are co-formulated in one dosage form and are commonly used for the effective treatment of motion sickness in the ratio of 4 : 3 for cinnarizine and domperidone, respectively. To achieve the developed method’s best performance and reliability, a 23 full factorial design was applied during the optimization of chromatographic conditions. Subsequently, isocratic elution on a C18 column with a mobile phase mixture of methanol and 0.02 M potassium dihydrogen phosphate buffer (85 : 15 v/v, pH 3.0) was performed to achieve the best separation. Moreover, the flow rate of 0.8 ml min−1 was applied during the analysis with fluorescence detection at λex 283 nm/λem 324 nm. The elution process was time effective; it consumed less than 6 min for a complete run as retention time for cinnarizine and domperidone, was 4.5 and 3.5, respectively. Good linearity of the proposed method was achieved within the concentration ranges of 0.2–4.5 and 0.4–5.0 µg ml−1 for domperidone and cinnarizine, respectively. The suggested approach was carefully validated according to the International Council for Harmonization (ICH) guidelines. Moreover, green analytical procedure index (GAPI), national environmental methods index (NEMI), analytical greenness (AGREE) and analytical eco-scale methods were applied to assess and confirm method greenness.


Introduction
Over the past few decades, high-performance liquid chromatography (HPLC) replaced various spectroscopic techniques in testing the pre-sale process, drug marketing and drug control [1].Moreover, application of the design of experiments (DoE) approach for optimization of chromatographic conditions guaranteed procedure best performance and produced data reliability [2].
Both drugs are individually available in market and used to treat nausea vomiting/motion sickness.However, it was found that the combination of CRZ and DPN is more effective than individual ones because both of them have synergistic effects on the other [4].
By reviewing the literature, it was found that some analytical methods were reported for the determination of both drugs simultaneously, such as spectrophotometry [22,23], spectrofluorimetry [24] and HPLC with UV-detection [25][26][27].The proposed method is superior to all reported HPLC methods [25][26][27] from many different aspects; the first of them is the use of fluorescence detectors, which are not only significantly more selective than UV detectors for many compounds but also more sensitive, stable and easy to use [1].Furthermore, the development of each of the earlier techniques involved altering a single factor at a time; this made the method laborious, time-wasting, costly, ineffective at fixing errors and liable to produce unpredictable results [28].However, this study depends on the DoE approach which has emerged as an innovative, systematic strategy that leads to scientifically valid results.The application of this approach in chromatographic conditions optimization has several merits in the efficient development of a rugged and robust technique.Additionally, it is a helpful tool for gathering as much information as possible with fewer experiments.DoE is predicated on looking at several variables at once in order to get a particular response and analysing how they interact [29].Moreover, method greenness was evaluated and confirmed with four assessment tools.All these merits made the developed method harmonize with the current direction for developing green, simple, sensitive and valid analytical techniques, with less effort, cost, time and solvents for the determination of CRZ and DPN in pure materials, single and combined tablets.-Minitab Software (USA) was used to perform the steps of DoE.
-Methanol, ethanol and acetonitrile were HPLC grade, and they were obtained from Fisher Scientific, Loughborough, Leics, UK. -Potassium dihydrogen phosphate was obtained from ADWIC Co. (Cairo, Egypt).

Standard and working solutions
CRZ and DPN stock solutions (200 µg ml −1 of each) were prepared individually by dissolving 20 mg of each drug in methanol using a 100 ml volumetric flask.These stock solutions were stable for two weeks when wrapped with aluminium foil and kept in the refrigerator.After that, the mobile phase was used for further dilution to prepare working solutions within the concentration ranges (0.4-5.0 and 0.2-4.5 µg ml −1 for CRZ and DPN, respectively).

Construction of the calibration curves
Working solutions were prepared individually by taking different volumes from each drug standard solution into a set of 10.0 ml volumetric flasks and diluting them with the mobile phase.Each solution was injected in a triplicate manner under the optimized chromatographic conditions.Subsequently, each drug peak area was plotted against its concentration to construct a calibration curve and the regression equation was then obtained.

Assay of CRZ and DPN in their synthetic mixtures
Suitable aliquots of CRZ and DPN stock solutions were mixed and diluted with methanol in a 50.0 ml volumetric flask to get a standard laboratory synthetic mixture solution, maintaining the pharmaceutical ratio of 4 : 3 for CRZ and DPN, respectively.Various volumes of this solution were transferred into a series of 10.0 ml volumetric flasks and diluted to volume with the mobile phase.Then, the procedure outlined under §3.1.was followed.By referring to the matching regression equations or using the corresponding calibration curves, the concentrations of each drug were determined.

Analysis of CRZ and DPN in their single commercial dosage forms
Ten Cinnarzine ® or Dompidone ® tablets were weighed, finely pulverized and mixed well.The accurately measured weight of each drug mixed tablet powder equivalent to 10.0 mg of the respective drug was transferred into a 50.0 ml volumetric flask followed by 20.0 ml methanol.Each flask content was sonicated for approximately 45 min, diluted to the mark with the methanol, then filtered.Furthermore, aliquots containing suitable concentrations within the working range of each drug were then analysed as described in §3.1.Subsequently, each drug concentration was calculated either from a correlative regression equation or the corresponding calibration curve.

Analysis of CRZ/DPN in their co-formulated tablets
Ten Vertigil ® tablets were weighed, finely triturated and mixed well.An accurately calculated weight from mixed tablets powder, equivalent to 20 mg of CRZ and 15 mg of DPN, was quantitively transferred to a 100 ml volumetric flask, 70 ml methanol was then added, and flask content was sonicated for approximately 45 min, completed with methanol to the volume and filtered.Furthermore, different volumes from this combined tablet stock solution were accurately transferred into a set of 10 ml volumetric flasks, completed to the volume with mobile phase, then the procedure in §3.1 was implemented with subsequent determination of each drug concentration via the corresponding calibration curves or regression equations.

Chromatographic method optimization
Recently, DoE has been adopted to decrease the trial number during the optimization stage and enhance separation quality [29].Before employing a factorial design, preliminary investigations should be carried out as follows.
The first tried column was the most suitable one in terms of time of analysis, peak shape and resolution.By using other columns, overlapped, broad, blunted or forked peaks were obtained, which is not desirable.

Choice of suitable wavelength
Ideal excitation and emission wavelengths were chosen by analysing the individual fluorescence spectra of CRZ and DPN.It was observed that CRZ exhibits intrinsic fluorescence at λ ex 253.0 nm/λ em 308.0 nm, while DPN has intrinsic fluorescence at λ ex 307.0 nm/λ em 338.0 nm (figure 2).Subsequently, the ideal combination for the excitation and emission wavelengths was 283.0 and 324.0 nm, respectively, since it offered a good compromise for both medication sensitivity.

Mobile phase optimization
In order to investigate the possibility of enhancing the performance of the chromatographic system, a number of changes to the composition of the mobile phase were made.The type and ratio of the organic modifier, pH and ionic strength of the buffer, and flow rate were among these adjustments.

pH
The pH of buffer affects the sensitivity, resolution, tailing of the peaks and retention times of the studied drugs.pH range of 3.0-3.8was carefully selected for factorial design.During the early trials, it was concluded that decreasing pH below 3.0 led to decreasing in the sensitivity and resolution of both drugs and increasing in tailing factors of both peaks.Whereas, by increasing the pH above 3.8, the sensitivity of both drugs also decreased and the retention time of CRZ increased greatly, which is not desirable during this study.

The organic solvent type and ratio
During early studies in the screening stage, various solvents like methanol, ethanol and acetonitrile were tried.It was noticed that using ethanol led to overlapped peaks, while using acetonitrile led to an undesired increase in retention times, so methanol was chosen as the suitable organic solvent because its use resulted in the ideal retention times, sensitivity, resolution and shape of peaks.Moreover, its ideal ratio (that was selected for the factorial design study) ranged from 75% to 85%, as decreasing its ratio to less than 75% or increasing more than 85% resulted in long retention times and poor resolution, respectively.

Buffer ionic strength
Various ionic strengths of potassium dihydrogen phosphate buffer in the range 0.02−0.06M were tested.It was concluded that the decrease or increase in ionic strength of the used buffer did not significantly affect retention times and resolution of the two studied drugs, but by increasing the ionic strength to more than 0.02 M, CRZ suffered from forked peaks.Therefore, 0.02 M buffer was selected to be used during this study to get a higher theoretical plate number and good peak resolution.Furthermore, this aqueous buffer has some merits, like simple and easy preparation, and the washing of the column after analysis does not consume a long time.On the other hand, using water led to unresolved and forked peaks.

Operating flow rate
The range of flow rate from 0.8 to 1.2 ml min −1 guaranteed the ideal separation in a reasonable time.Therefore, this range was selected for the factorial design.

The 2 3 full factorial design approach
It is a type of design of experiments which is a multivariate optimization.It has a statistics-based approach and offers a number of benefits for developing and optimizing robust methods.The attractiveness of DoE generates from the fact that DoE requires significantly less money, time and resources for optimization than univariate procedures [2].From early studies, it was found that there were three independent factors affected the chromatographic method performance: pH, percentage of methanol and flow rate; and the responses that were observed were the resolution and drug peak tailing.Therefore, a 2 3 full factorial design approach (two levels and three independent factors) was adopted during the optimization stage of chromatographic conditions.Eight experiments were suggested by the 2 3 full factorial design in order to pinpoint the ideal circumstances that produced the  optimum response values (table 1).These studied responses were peak resolution between CRZ and DPN (R s ) and tailing factors of DPN and CRZ peaks.
For the determination of three responses (lower, upper and target values), the Minitab response optimizer was used as shown in table 2. Consequently, optimization of the ideal conditions for the input and desirability values was performed [2,30].Additionally, to make sure that the responses are within acceptable limits, the response optimizer calculates composite desirability (D).Its value ranges from zero to one and it is used to determine whether ideal circumstances are met or not.Maximizing the composite desirability guarantees the establishment of ideal conditions.The Minitab optimization plot displays how each factor affects the composite desirability or responses as well as how those factors interact together to produce the ideal circumstances [2].Moreover, table 2 explains that in response optimizer, lower, target and upper values are defined for dependent responses.The optimal setting for the input variables along with desirability values is calculated by the Minitab response optimizer.To ensure that the optimum conditions are obtained, the Minitab response optimizer calculates the composite desirability (D), which evaluates whether the responses are in their acceptable limits and range from zero to one.Zero is not accepted as it means that many of the responses are out of their accepted limits, while one means that the condition reached is optimum, so its value is better to be one or near one.

Factors that affect resolution (R s ) between CRZ and DPN
According to the Pareto chart of the factor effects (figure 3a), the main effects plot (figure 4a) and the normal plot (figure 5a), the elution buffer pH (A) had a significant positive effect, % methanol (B) had the dominant significant negative influence for a 95% confidence level.From the interaction plots (figure 6a), % methanol had a negative relationship with resolution when it interacted with the pH of elution buffer either at low or high levels.Interactions of other included parameters did not significantly affect resolution.

Factors that affect peak tailing of DPN
Pareto chart (figure 3b) and the normal plot (figure 5b) showed that pH had the highest positive effect on DPN peak tailing, but it was not significant at a 95% confidence level, then % methanol and flow rate.From the main effects plot (figure 4b), pH and flow rate had a strong positive impact, while % methanol had a strong negative impact.Moreover, interaction plots represented that pH had a minimum effect at the high level and strongest negative impact at the low level when interacted with % of methanol, and the flow rate had the strongest positive impact at both levels either interacted with pH or % methanol (figure 6b).

Factors that affect peak tailing of CRZ
As represented by the Pareto charts (figure 3c), the main effects plots (figure 4c) and the normal plot (figure 5c), % methanol had the strongest significant negative influence on CRZ peak tailing at a 95% confidence level.While flow rate affected it positively.The interaction plot (figure 6c) illustrated that % methanol had an inverse correlation on CRZ peak tailing when interacted with the pH at both levels.Furthermore, flow rate had a direct relationship with CRZ peak tailing when interacted with pH or % methanol at both levels for each factor.

Method optimum conditions
Consequently, DoE study concluded that the optimum separation conditions were achieved using a mobile phase consisting of methanol and 0.02 M potassium dihydrogen phosphate buffer (85 : 15, v/v, pH 3.0) and 0.8 ml min −1 flow rate (figure 7; table 3).

Method validation
The suggested technique was validated following ICH Guidelines [31] to guarantee method suitability for its desired purpose.The investigated parameters of validation process were as follows.

Linearity and range
Under the optimized experimental conditions, calibration curves for CRZ and DPN were constructed by plotting peak areas versus concentrations of each drug (µg ml −1 ) and it was concluded that the      where C is the concentration of the studied drug (µg ml −1 ).

Accuracy
The accuracy of the proposed method was confirmed through its application in the quantitative estimation of CRZ and DPN in pure materials and comparing the results with those obtained by comparison method [25] as shown in (table 5).The comparison method recommended HPLC for the simultaneous determination of both drugs using the C18 column.The mobile phase consisted of methanol : 0.05 M phosphate buffer in the ratio (80 : 20 v/v, pH 3.8 adjusted with orthophosphoric acid), flow rate of 1 ml min −1 and UV detection at 275.0 nm.Using Student's t-test and variance ratio F-test [32], the data were statistically analysed, and the results revealed no discernible difference between the two methods, verifying the suggested method's accuracy.

Precision
Three separate analyses of each analyte at three different concentrations were performed on the same day in order to assess the intraday precision.To evaluate the interday precision, the same concentrations were measured three times.Small values for s.d. and % relative standard deviation (RSD), as shown in table 6, pointed out that the evolved method has higher precision.

Specificity
The specificity of the method was ascertained by evaluating the mentioned analytes in both their commercial single and combined formulations and looking for dosage excipient interferences as shown in tables 7 and 8.It turned out that the assay performance was unaffected by these excipients, and there was no discernible difference between the findings of the estimation of CRZ and DPN in pharmaceutical formulations and pure materials.

Robustness
The process parameters, such as the pH (3.0 ± 0.1), percentage of methanol (85% ± 1) and the used buffer ionic strength (0.02 M ± 0.01), were deliberately altered to evaluate the robustness.The findings indicate that the HPLC approach is robust as these variations had little to no effect on its performance (table 9).

System suitability
System suitability assessments were done referring to ICH Guidelines [31] on mixture of CRZ and DPN to calculate the chromatographic parameters, as represented in table 10.

Comparison of analytical performance of the suggested and previously reported methods
The suggested method was compared with the previously reported methods regarding LOD, LOQ, range, experimental conditions, application and greenness assessment as represented in table 11.

Method applications 5.1. Analysis of CRZ/DPN in their synthetic mixtures
The procedure was effectively applied for the analysis of CRZ and DPN in synthetic mixtures (figure 8).A statistical evaluation of the acquired data using the variance ratio F-test, and Student's t-test [32] revealed no discernible distinction in the accuracy and precision of the two approaches' performance as displayed in table 12.

Analysis of CRZ/DPN in their single commercial dosage form
The procedure was successfully used to determine DPN and CRZ in single dosage forms.As demonstrated by t and F values [32], there was no significant distinction between the obtained data and those attained from the comparison method [25] as listed in table 7. Subsequently, there was no evidence of excipient interference.

Analysis of CRZ/DPN in their co-formulated tablets
The investigated medications were analysed in their combined tablets (figure 9), and table 8 shows that the assay produced satisfactory results which agreed well with that of the comparison method [25].

Greenness assessment
Green analytical chemistry has made progress in quantifying how analytical techniques affect the environment.The assessment of the greenness of chemical processes is one of the challenges encountered in green chemistry.It is widely recognized that processes that are unmeasurable are uncontrollable.So, four green chemistry metrics such as green analytical procedure index (GAPI), analytical greenness (AGREE), national environmental methods index (NEMI) and analytical eco-scale score were applied to assuring eco-friendliness of the evolved method.Recently, GAPI has evaluated the green character of an entire analytical methodology.It is applied using pentagrams, which employ a three-level colour system with green, yellow and red denoting low, medium and high environmental effects [33], respectively, as shown in table 13.
Moreover, NEMI is characterized by its simplicity and readability for potential process users.It only takes a quick look at the NEMI pictogram to get a general overview of how a technique will affect the environment [34] as presented in table 13.
The AGREE graph has 12 sectors encircling its circumference and is shaped like a clock.Each sector represents one of the 12 principles of green analytical chemistry (GAC).The method performance with respect to each of the GAC principles is shown by a red, orange and green scale [35].The total   [32].
performance of the method is represented by the colour in the centre of the graph, which has a score between 0 and 1.The score of this suggested method was 0.93 as shown in table 13.The analytical eco-scale is another tool for the assessment of analytical methods' ecological impact [34].For each method, parameter penalty points are assigned.The sum of penalty points should be subtracted from the basis of 100.The closer the score to 100, the higher degree the greenness obtained (table 13).

Conclusion
It was crucial to develop a straightforward, accurate, selective, robust and precise HPLC technique for the simultaneous assay of CRZ and DPN in pure, single and co-formulated pharmaceutical forms.This is the first chromatographic technique using a fluorescence detector to analyse both medications simultaneously in approximately a short time of analysis, less than 6 min.A two-level full factorial design was used to optimize the chromatographic conditions, which cut down on time and trial count.Moreover, greenness, sensitivity, rapidity, appropriate robustness and wide applicability distinguished the approach and made it successfully implemented in actual-life situations by analysing both medications in their single and combined commercial dosage forms, with satisfactory percentages found without any interference.Ultimately, this method can be followed effectively in quality control work.
Ethics.This work did not require ethical approval from a human subject or animal welfare committee.Data accessibility.The datasets generated during and/or analysed during the current study are available in the Dryad repository [36].
Declaration of AI use.We have not used AI-assisted technologies in creating this article.Authors' contributions.N.T.B.: conceptualization, data curation, investigation, methodology, resources, software, validation, visualization, writing-original draft, writing-review and editing; A.M.E.-B.: conceptualization, data curation, formal analysis, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing-original draft, writing-review and editing; M.E.F.: conceptualization, data Table 13.Evaluation results of the greenness of the suggested method.

Figure 3 . 2 3
Figure 3. 2 3 Full factorial design Pareto charts of the effects on the chromatographic responses at alpha = 0.05 (a-c).

Figure 4 . 2 3
Figure 4. 2 3 Full factorial design main effect plots for chromatographic responses by data means type (a-c).

Figure 5 . 2 3
Figure 5. 2 3 Full factorial design normal plots of the effects on the chromatographic responses at alpha = 0.05 (a-c).

Figure 6 . 2 3
Figure 6. 2 3 Full factorial design interaction plots for chromatographic responses by data means type (a-c).

Table 1 .
Suggested eight runs to perform 2 3 experimental factorial designs.

Table 2 .
23full factorial design response optimization for reversed-phase high-performance liquid chromatography (RP-HPLC) separation of cinnarizine and domperidone mixture.

Table 3 .
Optimum chromatographic conditions.approach was linear within the concentration ranges of 0.4-5.0 and 0.2-4.5 µg ml −1 for CRZ and DPN, respectively.Furthermore, linear regression analysis of the data was summarized in table 4 and the following equations were derived: column Promosil C18 (250×4.6 mm, 5 µm) mobile phase methanol and 0.02 M potassium dihydrogen phosphate buffer (85 : 15, v/v) proposed

Table 4 .
Performance data for the proposed HPLC method for estimation of cinnarizine and domperidone.

Table 5 .
Analytical data for the analysis of domperidone and cinnarizine in their pure forms using the developed HPLC method.

Table 6 .
Precision data of the evolved HPLC method for cinnarizine and domperidone.

Table 7 .
Application of the proposed HPLC method for analysis of cinnarizine and domperidone in their single dosage forms.

Table 8 .
[32]ication of the suggested HPLC method for analysis of cinnarizine and domperidone in their combined dosage forms.The tabulated t and F values at (p = 0.05) were 2.44 and 9.27, respectively[32]. b

Table 9 .
The proposed method robustness.

Table 10 .
System suitability parameters of the proposed HPLC method.

Table 11 .
Analytical performance of the suggested technique and other reported methods.

Table 12 .
Application of the evolved HPLC method for analysis of cinnarizine and domperidone in their synthetic mixtures.The tabulated t and F values at (p = 0.05) were 2.44 and 9.27, respectively b