TESTING LONGITUDINAL RELAXATION (T1) ON VARIABLE TEMPERATURE CONDITION FOR DIFFERENTIATE API POLYMORPHS

Over the past five years there has been an increase in reports on the use of Time Domain NMR (TD-NMR) to evaluate polymorphism on active pharmaceutical ingredients (API) (1). Even though TD-NMR was able to differentiate between anhydrous/hydrated forms or crystalline/amorphous forms in API polymorphs, it does not present enough sensitivity to distinguish Mebendazole (MBZ) forms A and C polymorphism, that relies on different crystalline cell packings (2-5). Given the challenges associated with differentiating the spin-lattice relaxation time (T₁) in polymorphic forms, alternative experiments, such as the Torchia pulse sequence, may provide a practical approach to the identification of different components at the structural arrangement level (6). Additionally, since T₁ measurement is inherently tied to the recovery of net magnetization to thermal equilibrium, temperature fluctuations can significantly impact the accuracy of T₁ determinations (7). Therefore, the present work tested a variable temperature condition on TD-NMR T1 measurements to differentiate between Mebendazole A and C polymorphic forms (MBZ A and MBZ C). The raw material MBZ A, manufactured by Supharma Chem, was provided by Farmanguinhos, Fiocruz (Rio de Janeiro State, Brazil). The MBZ form C was prepared by solubilization and recrystallization of form A in dimethyl sulfoxide (DMSO) (5). Characterization was conducted by solid-state NMR (ssNMR) analysis using a ¹³C-CP-MAS pulse sequence and Differential Scanning Calorimetry (DSC) analysis. ssNMR analysis were performed using an Avance III, Bruker, at 9.7 T (400 MHz) with the solid-state probe. Data was acquired using the following parameters: d1 of 5 s, p3 of 3 μs, pl1 of 50.1 W, pl15 of 2 ms, acquisition time of 34ms, number of scans (ns) of 1024, number of points 2048 at 10 kHz spinning rate. DSC measurements were performed using a calorimeter model DSC Q100, TA Instruments working temperature: -90°C to 550°C. The heating ramp was 10°C/min and the observed temperature range was 0°C to 250°C on nitrogen atmosphere. T₁ measurements were performed in a Minispec ND mq-20 Spectrometer (Bruker, Germany) operating at 19.9 MHz (0.47 T) for the ¹H nucleus, using a 10 mm diameter probe with 11.6 μs dead time and 2.4 μs π/2 pulse. The relaxation values ​​were measured with the Torchia pulse sequence (8) and the temperature was controlled using the Bruker BVT 3000 system, with a resolution of +- 1K. Data was processed in OriginPro 2016 from OriginLab Corporation. A monoexponential fitting (ExpDec1) was used for an average value of T₁. Characterization essays provided results similar to the ones reported in the literature for the MBZ A and C forms. T₁ measurements were acquired in nine different temperatures (301.5 K, 319.4 K, 337.3 K, 355.2 K, 373.1 K, 391.0 K, 408.8 K, 426.7 K, 444.6 K). The Torchia pulse sequence, designed for determining T₁ in materials with distinguished proton density systems, was useful to characterize, individually, the behavior of both samples (9). Data demonstrated a temperature dependence on relaxation rates, with a decrease in the T₁ until reaching a minimum. The last measurement demonstrated a similar behavior in both polymorphic forms, which provided evidence for a possible dynamical change associated to thermal transitions (10). A similar approach was conducted with XRD analysis in different temperatures by de Villiers et al. Upon heating between 200 and 225°C, there was observed a conversion of MBZ C to MBZ A (11). To the best of our knowledge, this was the first attempt to obtain API polymorphic forms relaxation profile through a variable temperature condition. This approach may be useful to both qualitative identification of polymorphic forms as well as to track polymorphic conversion in API polymorphic forms due to thermal transition.