LARGE VOLUME INJECTION OF HEXANE SOLUTIONS IN RPLC/UV TO ENHANCE ON SENSITIVITY OF THE ASSAY OF GINKGOLIC ACIDS IN GINGKO BILOBA STANDARDIZED EXTRACTS

Abstract

The sensitivity of the compendial European Pharmacopoeia RPLC/UV method for assaying residual ginkgolic acids in Ginkgo biloba standardized extracts was essentially improved by operating modifications on the sample preparation procedure, including sample injection. The acidified methanol/water solution of the Ginkgo biloba standardized extract was extracted “in situ” with n-hexane (concentration factor of 20). From the organic layer, a relatively large volume (50 µL) of the hexane solution was directly loaded to the chromatographic column. A modified gradient elution profile was used in order to force the sample solvent to elute faster than analytes, without affecting peak symmetry. The separation is obtained in 22 minutes. As the amount of the target compounds loaded to column is higher, selective UV detection at 310 nm was possible. The method was validated according to specific operating guidelines for selectivity, linearity range, quantitation limits, precision, accuracy, and robustness. A limit of quantitation of 1 ppm (five times less than the accepted maximal content threshold for ginkgolic acids in standardized G. biloba extracts) was obtained. Confirmation of the target compounds through MS² detection and evaluation of the sensitivity afforded by such a detection system are also discussed.

Keywords:

• direct injection of n-hexane solutions
• ginkgolic acids
• improved sensitivity
• liquid-liquid extraction
• RPLC/UV
• validation

Notes

(*) LOQ calculated according to the relationship [2 x t x (s_A + s_B x C_m)]/(B + 2 x t x s_B), where B is the slope of the linear regression equation, s_A and s_B are the standard deviations of the intercept and of the slope, C_m is the mean value of the concentrations set, and t is the Student coefficient chosen for n-2 degrees of freedom (n = number of concentration levels) and for a level of confidence P% = 95% (bilateral) [25].

(**) LOQ calculated according to the relationship (10 x s_A–A)/B, where A is the intercept, B the slope and s_A the standard deviation of the intercept.

(*) bolded parameters correspond to nominal values considered in the Experimental section, paragraph 2.3.

(*) operational parameters used in the method according to data presented in the Experimental section.