^{1}

^{*}

^{2}

^{1}

^{3}

^{1}

The aim of this study was to develop a simple and derivatization free method for the Quantification of S-Epichlorohydrin in R-Epichlorohydrin by using a gas chromatography coupled with flame ionization detector (FID). Enantiopure epichlorohydrin was a valuable epoxide key starting material for preparing optically active Rivaroxaban. The enantiomeric separations of S-Epichlorohydrin and R-Epichlorohydrin were achieved on Gamaa-Dex-225 (30 meters × 0.25 mm I.D, 0.25 μm) column with a total run time of 30 min. Nitrogen was used as a carrier gas with constant pressure 25.0 psi. The critical experimental parameters such as, column selection, flow rate, injection volume and diluent were studied and optimized. Excellent correlation coeffient between peak responses and concentrations was >0.9998. The recoveries of S-Epichlorohydrin spiked in R-Epichlorohydrin were in the range from 98.2% to 102.8%. Limit of quantitation for S-Epichlorohydrin was sufficiently lower than limits specified by ICH. The method has validated as per International Conference on Harmonization (ICH) guidelines. A precise, accurate, linear and robust Gas Chromatography method was developed for the quantification of S-Epichlorohydrin in R-Epichlorohydrin for Rivaroxaban.

Separation of the enantiomers of chiral drugs has become an important issue in analytical chemistry in recent years, because of differences in the biological activity and pharmacokinetic properties of drug enantiomers [

Rivaroxaban is chemically 5-Chloro-N-({(5S)-2-oxo-3-[4-(3-oxo-4-orpholinyl) phenyl]-1,3-oxzolidin-5-yl}methyl)-2-thiophene-carboxamide with molecular formula C_{19}H_{18}ClN_{3}O_{5}S. Rivaroxaban was used for potent anticoagulant and antithrombotic effects [

R-Epichlorohydrin and S-Epichlorohydrin were purchased from Sigma-Aldrich Dich

loromethane was purchased from Merck, Germany.

A calibrated electronic single pan balance Mettler Toledo. All analysis performed on Agilent 6890 and 7890 modules equipped with FID detectors. Empower-3 software was used for signal monitoring and data processing. Microsoft Excel 2007 was used for analysis of validation results.

The method was developed by using Gamaa-Dex-225 (30 meters × 0.25 mm I.D, 0.25 µm) column. The separation was achieved using an isothermal oven program 50˚C for 30 min. The injector temperature was maintained at 250˚C. Nitrogen was used as a carrier gas with constant pressure 25.0 psi. The detector temperature was maintained at 250˚C. The injection volume was 1.0 μL. Split ratio 1:50 and Runtime was 30.0 min.

Dichloromethane was used as diluent during the standard and test samples preparation. Chromatograms were summarized in Figures 2(a)-(d).

Transfer accurately 0.1 mL of S-Epichlohydrin standard in 50 mL volumetric flask, containing 10.0 mL diluent and made up to volume with diluent.

Transfer accurately 1.0 mL of standard in 50 mL volumetric flask, containing 10.0 mL diluent and to this add accurately 0.5 mL of above S-Epichlorohydrin stock solution and made up to volume with diluent.

Transfer accurately 1.0 mL of standard in 50 mL volumetric flask, containing 10.0 mL diluent and made up to volume with diluent.

Transfer accurately 1.0 mL of test sample in 50 mL volumetric flask, containing 10.0 mL diluent and made up to volume with diluent.

The method has been validated by GC as per ICH guidelines [

The precision of the method was verified by repeatability and by intermediate preci- sion. Repeatability of the method was checked by (Agilent 7890 module equipped with

FID detector) injecting six individual preparations of R-Epichlorohydrin sample spiked with 0.10% of S-Epichlorohydrin (0.10% of S-Epichlorohydrin isomer with respect to 20 μL/mL R-Epichlorohydrin). The RSD of peak area for S-Epichlorohydrin was calculated. The intermediate precision of the method was also evaluated using different analysts, different instruments and different columns and performing the analysis on three different days.

For determination of accuracy of method recovery study was carried out by analyzing the spiked samples. Known amount of S-Epichlorohydrin was spiked in triplicate at three different concentration levels of 0.01, 0.02 and 0.03 μL/mL (50%, 100% and 150% of the analyte concentration i.e. 20 μL/mL) to the drug product. The % recoveries for S- Epichlorohydrin was calculated based on mentioned in Equation (1).

To establish linearity of the method was prepared by diluting stock solution to the required concentration. The solutions were prepared at seven concentration levels from LOQ to 150% of the specification level (LOQ, 0.025, 0.050, 0.075, 0.10, 0.125and 0.150%) with respect to the normal sample concentration (20 μL/mL). The correlation coefficients, slopes and Y-intercepts of the calibration curve were determined.

The LOD and LOQ for S-Epichlorohydrin were determined at a signal-to-noise ratio of 3:1 and 10:1, respectively, by injecting a series of dilute solutions with known concentrations. Precision study was also carried out at the LOQ level by injecting six (n = 6) individual preparations and calculating the RSD of the area of S-Epichlorohydrin.

The robustness of an analytical procedure is a measure of its capability to remain unaltered by small, but deliberate variations in method parameters and provides an indication of its reliability during normal usage.

To determine the robustness of the method, the chromatographic conditions were deliberately changed and relative standard deviation of the S-Epichlorohydrin peak was evaluated. As the flow rate was 25 psi to study the effect of flow rate on relative standard deviation of the S-Epichlorohydrin peak, the flow rate was changed to 20 psi and 30 psi. The effects of the column oven temperature were studied at 45˚C and 55˚C instead of 50˚C.

The stability of S-Epichlorohydrin in R-Epichlorohydrin solution was determined by leaving test solutions of the sample and spiked solution in tightly capped volumetric flasks at room temperature for 24 hrs during which they were analysed at 12 hrs intervals.

The main goal of method development was to achieve separation of S-Epichlorohydrin in R-Epichlorohydrin without derivatization. An understanding of the nature of the racemic Epichlorohydrin is the foremost prerequisite for successful method development in GC. Following were the stepwise strategies for the method development in our case.

The primary goal of column selection was to separate S-Epichlorohydrin and R-Epi- chlorohydrin from each other, which were used during the synthesis of Rivaroxaban. As part of method development screened various columns, namely Chiral GTA (30 meters × 0.25 mm I.D, 0.12 µm) and Chiralsil (30 meters × 0.25 mm I.D, 0.25 µm) were employed but no adequate separation was found with above columns. After careful screening of columns, it was observed that Gamaa-Dex-225 (30 meters × 0.25 mm I.D, 0.25 µm) column provides better resolution between S-Epichlorohydrin and R-Epich- lorohydrin and it showed good system suitability parameters.

As the flow rate increase, the viscosity of carrier gas decreased and velocity increased. Check the flow rate from 6 psi to 40 psi. At 6 psi, the retention time was very high and runtime is long, poor separation was observed at 40 psi. 25 psi was selected as finalized flow rate.

Diluent selection study was conducted for the analysis of R-Epichlorohydrin and S- Epichlorohydrin. Four diluents had been tried N-methyl-2-pyrrolidone, Dimethyl formamide, Dimethyl imidazolidine and Dichloromethane. Dichloromethane was finalized as diluent because of no interference at the S-Epichlorohydrin and R-Epichlorohy- drin peaks.

The effect of injection volume on the quantification of the S-Epichlorohydrin and R- Epichlorohydrin were investigated by injecting volume between 0.5 μL to 2 μL of the standard solution. The results show that the peak widths of S-Epichlorohydrin and R- Epichlorohydrin were independent of injection volume within the tested range.

System suitability results are shown in

The % RSD for the content of S-Epichlorohydrin in the method precision was found to be less than 1.1. The % RSD for the content of S-Epichlorohydrin in the intermediate precision (Ruggedness) was found to be less than 2.2 (

The obtained limit of detection and limit of quantification, precision and accuracy at limit of quantification values are given in

Compound | RT (min) | RRT^{a}^{ }(n = 6) | %RSD^{b} (n = 6)^{c} |
---|---|---|---|

S-Epichlorohydrin | 9.169 | 0.85 ± 0.01 | 0.54 |

R-Epichlorohydrin | 10.721 | 1.00 ± 0.00 |

^{a}Relative retention times (RRT) were calculated against the retention time (RT) of R-Epichlorohydrin. ^{b}Relative standard deviation. ^{c}Mean ± SD.

Individual and average recoveries of three preparations and at three concentrations for S-Epichlorohydrin were within 100% ± 5% results shown in

Parameter | S-Epichlorohydrin |
---|---|

Linearity | |

Correlation coefficient | 0.9998 |

Slope | 137034.3 |

Y-Intercept | 0.0 |

%Y-Intercept | 0.22 |

Accuracy (% Recovery) | |

LOQ (n = 3) | 98.2 |

50% (n = 3) | 102.8 |

100% (n = 3) | 101.5 |

150% (n = 3) | 99.8 |

Precision (% RSD) | |

LOQ (n = 6) | 3.29 |

50% (n = 6) | 1.98 |

100% (n = 6) | 1.08 |

150% (n = 6) | 0.98 |

Ruggedness; Different day and analyst (%RSD) | |

100% (n = 6) | 2.16 |

Robustness (%RSD) | |

Actual flow 25 psi | 0.80 |

Different flow 20 psi | 1.34 |

Different flow 30 psi | 1.09 |

Different Column Temperature 45˚C | 2.12 |

Different Column Temperature 55˚C | 1.98 |

Limit of Detection (Concentration in µg/mL) | 0.001 |

Limit of Quantification (Concentration in µg/mL) | 0.004 |

The calibration curve was drawn by plotting the peak area against the concentration. The correlation coefficient (r^{2}) obtained was ˃0.9998. The % Y-intercept with respect to response at 100% level was ˃± 5%. The results for the Correlation coefficient (r^{2}) and % Y-intercept with respect to response at 100% level were shown in

In all the deliberately varied chromatographic conditions, no effect on the Relative standard deviation of the S-Epichlorohydrin peak (

Spike | S.No. | % of S-Epichlorohydrin in spiked sample | Avg S-Epichlorohydrin in sample | % of S-Epichlorohydrin found | % of S-Epichlorohydrin added | % Recovery | Avg. % Recovery |
---|---|---|---|---|---|---|---|

50% | Prep-1 | 0.1003 | 0.0500 | 0.0503 | 0.050 | 100.6 | 102.8 |

Prep-2 | 0.1021 | 0.0500 | 0.0521 | 0.050 | 104.2 | ||

Prep-3 | 0.1018 | 0.0500 | 0.0518 | 0.050 | 103.6 | ||

AVG | 0.10 | ||||||

SD | 0.00 | ||||||

RSD | 0.95 | ||||||

100% | Prep-1 | 0.1525 | 0.0500 | 0.1025 | 0.100 | 102.5 | 101.5 |

Prep-2 | 0.1514 | 0.0500 | 0.1014 | 0.100 | 101.4 | ||

Prep-3 | 0.1506 | 0.0500 | 0.1006 | 0.100 | 100.6 | ||

AVG | 0.15 | ||||||

SD | 0.00 | ||||||

RSD | 0.63 | ||||||

150% | Prep-1 | 0.1960 | 0.0500 | 0.1460 | 0.150 | 97.3 | 99.8 |

Prep-2 | 0.2001 | 0.0500 | 0.1501 | 0.150 | 100.1 | ||

Prep-3 | 0.2032 | 0.0500 | 0.1532 | 0.150 | 102.1 | ||

AVG | 0.20 | ||||||

SD | 0.00 | ||||||

RSD | 1.81 |

S.No | Conc | Actual concentration in mg/mL | S-Epichlorohydrin Area | ` |
---|---|---|---|---|

1 | LOQ | 0.00000403 | 0.520 | |

2 | 25% | 0.00000586 | 0.870 | |

3 | 50% | 0.00001173 | 1.592 | |

4 | 75% | 0.00001759 | 2.426 | |

5 | 100% | 0.00002346 | 3.200 | |

6 | 125% | 0.00002932 | 4.050 | |

7 | 150% | 0.00003519 | 4.821 | |

Correlation | 0.9998 | |||

Slope | 137034.3 | |||

y-intercept | 0.0 | |||

% y-intercept | 0.22 |

SUMMARY OUTPUT | |||||
---|---|---|---|---|---|

Regression Statistics | |||||

Multiple R | 0.999782629 | ||||

R Square | 0.999565306 | ||||

Adjusted R Square | 0.999478367 | ||||

Standard Error | 0.036929457 | ||||

Standard Error × 3 | 0.11078837 | ||||

Observations | 7 | ||||

ANOVA | |||||

df | SS | MS | F | Significance F | |

Regression | 1 | 15.67989908 | 15.67989908 | 11497.34 | 0.00000000 |

Residual | 5 | 0.006818924 | 0.001363785 | ||

Total | 6 | 15.686718 | |||

Coefficients | Standard Error | t Stat | P-value | Lower 95% | |

Intercept | 0.007058657 | 0.02709363 | 0.260528295 | 0.804845 | −0.062587736 |

X Variable 1 | 137034.2905 | 1277.999082 | 107.2256565 | 1.34E-09 | 133749.0893 |

Upper 95% | Lower 95.0% | Upper 95.0% | |||

0.07670505 | −0.062587736 | 0.07670505 | |||

140319.4918 | 133749.0893 | 140319.4918 | |||

RESIDUAL OUTPUT | |||||

Observation | Predicted Y | Residuals | Standard Residuals | ||

1 | 0.559970752 | −0.039970752 | −1.1856596 | ||

2 | 0.810709957 | 0.059290043 | 1.758731197 | ||

3 | 1.614361258 | −0.022361258 | −0.663306009 | ||

4 | 2.418012558 | 0.007987442 | 0.236932931 | ||

5 | 3.221663858 | −0.021663858 | −0.642618921 | ||

6 | 4.025315158 | 0.024684842 | 0.732230897 | ||

7 | 4.828966459 | −0.007966459 | −0.236310495 |

No significant change in the amounts of S-Epichlorohydrin was observed during solution stability experiments. The results from solution stability experiments confirmed that sample and spiked solutions were stable for up to 24 hrs results shown in

Based on the results, the successful separation of S-Epichlorohydrin and R-Epichloro- hydrin from each other. All the validated parameters were found to be within limits. System suitability for 6 injections % RSD was found to be NMT 0.54%. Precision at LOQ, 100% and 150% were found to be NMT 3.29%, Accuracy at LOQ, 50% 100% and 150% were found to be 98.2% to 102.8%. Linearity was performed from LOQ to 150% and graph obtained was linear showing correlation coefficient >0.9998.

Sample “0” Hrs | Sample after 12 Hrs | Sample after 24 Hrs | Spiked “0” Hrs | Spiked after 12 Hrs | Spiked after 24 Hrs |
---|---|---|---|---|---|

% of Area | % of Area | % of Area | % of Area | % of Area | % of Area |

0.0500 | 0.0502 | 0.0503 | 0.1525 | 0.1528 | 0.1530 |

Difference | 0.0 | 0.0 | Difference | 0.0 | 0.0 |

% Difference | 0.4 | 0.6 | % Difference | 0.2 | 0.3 |

A simple gas chromatographic method was developed and validated for the quantitative S-Epichlorohydrin in R-Epichlorohydrin for Rivaroxaban. S-Epichlorohydrin and R- Epichlorohydrin were well separated from each other, indicating that the developed GC method was specific. The method validation data showed satisfactory results for all tested method parameters. This simple GC method is precise, accurate, linear and rugged. Hence, it is proved that developed method can be used for routine testing in quality control laboratories for estimation S-Epichlorohydrin in R-Epichlorohydrin for Rivaroxaban. The method is user-friendly and robust to operate.

The authors wish to thank the management of Dr. Reddy’s Laboratories Ltd. for supporting this work. Co-operation from colleagues of Research & Development and Analytical Research & Development of Dr. Reddy’s Laboratories Ltd. is appreciated.

Kumar, C.V., Vasa, P.K., Kumar, Y.R., Aparna, P. and Pratyusha, P. (2016) Enantiomeric Separation of S-Epichlorohydrin and R-Epichlo- rohydrin by Capillary Gas Chromatography with FID Detector. American Journal of Analytical Chemistry, 7, 772-784. http://dx.doi.org/10.4236/ajac.2016.711069