CRYSTAL GROWTH & DESIGN
Polymorphism of Norfloxacin: Evidence of the Enantiotropic Relationship between Polymorphs A and B Rafael Barbas, Francesc Martı´, Rafel Prohens,* and Cristina Puigjaner* Unitat de Quı´mica Fina, SerVeis Cientı´fico-te` cnics, UniVersitat de Barcelona, Josep Samitier 1-5, 08028 Barcelona, Spain
2006 VOL. 6, NO. 6 1463-1467
ReceiVed February 24, 2006; ReVised Manuscript ReceiVed April 25, 2006
ABSTRACT: A new interpretation of the polymorphic system of norfloxacin is described. Experimental evidence of the enantiotropic relationship between polymorphic forms A and B is provided by means of a solvent-mediated polymorphic transformation. Form B is the most stable one at room temperature, and it is enantiotropically related to the form A. Full characterization of both forms by X-ray crystallography, differential scanning calorimetry (DSC), IR, Raman, and solid-state NMR spectroscopy is provided. Introduction Polymorphism and crystallization have become an increasingly important and relevant topic in the pharmaceutical/generic drug market. Delivering an active pharmaceutical ingredient with the desired crystal form, size, purity, and acceptable yield is always a challenge. Many drugs exhibit polymorphism, and their physical forms are vital for obtaining the desired therapeutically effective product. The ability of a particular polymorph to crystallize is usually determined by both thermodynamic and kinetic factors. These factors must be well understood to explore and control the polymorphic behavior of a substance. Emblematic of the importance of polymorphs is the cautionary case of ritonavir, the anti-HIV drug made by Abbott Laboratories.1 Introduced in 1996, this drug had been on the market for 18 months when suddenly, during manufacturing, chemical engineers detected a previously unknown polymorph. No one knew what had caused the change, but the scientists discovered that the new polymorph was thermodynamically more stable than the drug in its original form. This new form was only half as soluble as the first, so patients taking prescribed doses would not get enough of the drug into their bloodstreams. The company finally decided to reformulate the drug in the new polymorphic form. Cases such as this are not routine, but they are common enough for drug companies to be concerned about the surprises that polymorphism can bring. As it has been shown in ritonavir case, using a thermodynamically unstable modification in the production of tablets is sometimes the reason unwanted changes take place in such formulations after a time of storage. This is caused by transition into the thermodynamically stable modification at room temperature. It is important to know whether polymorphic modifications can transform reversibly (enantiotropy) or irreversibly (monotropy) at atmospheric pressure, because polymorphic crystal forms of a specific chemical compound have different physical properties caused by different arrangements of the molecules in the crystal lattice. These different characteristics often lead to considerable differences in solubility, hygroscopicity, and bioavailability. When an organic compound exhibits polymorphism of an enantiotropic type, the knowledge of the different domains of thermodynamic stability for every form is essential to obtain the desired form by a robust crystallization process and to define the appropriate storage condition. * To whom correspondence should be addressed. E-mail addresses:
[email protected];
[email protected].
Figure 1. Formula of norfloxacin, 1-ethyl-6-fluoro-1,4-dihydro-4-oxo7-(1-piperazinyl)-3-quinolinecarboxylic acid.
Norfloxacin, 1-ethyl-6-fluoro-1,4-dihydro-4-oxo-7-(1-piperazinyl)-3-quinolinecarboxylic acid (Figure 1), is a synthetic broad antibacterial fluoroquinolone compound used in the treatment of gonorrhea and prostate and urinary tract infections.2 Norfloxacin has been reported to exist in a number of solid forms: two anhydrous polymorphs (forms A and B), an amorphous form,3 and several hydrated forms.2,4,5 The relationship between the two anhydrous polymorphic forms was reported to be monotropic with the higher melting form A being the stable phase. An endothermal solid-solid transition was observed in the differential scanning calorimetry (DSC) thermograms on heating but not on cooling. However, it is wrong to conclude that two polymorphs are monotropically related on the basis that there is not a reversible phase transition in a DSC experiment. There may be a considerable energy barrier involved in moving from a metastable to a stable state. The activation energy for the transition of a solid phase to the other phase is a kinetic parameter, which may be prohibitively high. In the case of enantiotropy, a higher melting form may not revert to a lower melting form on cooling since kinetics is an important factor.6 Moreover, according to the heat of transition rule, if an endothermic phase change is observed at a particular temperature, the transition point lies below that temperature, and the two polymorphs are enantiotropically related.7 So, it seems that monotropy cannot be applied to forms A and B of norfloxacin, but enantiotropy. Since the information about polymorphs A and B reported in the literature3 is doubtful, we decided to study the polymorphic system of norfloxacin. In the present study, we provide experimental evidence of the enantiotropic relationship between the A and B polymorphic forms. All forms have been characterized by X-ray crystallography, DSC, thermogravimetric analysis (TGA), IR, Raman, and solid-state NMR spectroscopy. Experimental Section Materials. Norfloxacin in Form A. This form was purchased from Sigma-Aldrich.
10.1021/cg060101u CCC: $33.50 © 2006 American Chemical Society Published on Web 05/23/2006
1464 Crystal Growth & Design, Vol. 6, No. 6, 2006
Figure 2. DSC curve of form A of norfloxacin carried out at the heating rate of 10 °C/min. Norfloxacin in Form B. This form was produced by heating norfloxacin (50 mg of form A) in 4.5 mL of 2-propanol at reflux temperature. Once the solid was dissolved, it was allowed to slowly cool to room temperature at a rate of 4 °C each day. The solid was isolated by vacuum filtration and washed with 2-propanol. Yield ) 78%. General Methods. Differential scanning calorimetry was carried out by means of a Mettler-Toledo DSC-822e calorimeter. Experimental conditions were as follows: aluminum crucibles of 40 µL volume; atmosphere of dry nitrogen with 50 mL/min flow rate; heating rates of 1-40 °C/min. The calorimeter was calibrated with indium of 99.99% purity. Thermogravimetric analyses were performed on a Mettler-Toledo TGA-851e thermobalance. Experimental conditions were as follows: alumina crucibles of 70 µL volume; atmosphere of dry nitrogen with 50 mL/min flow rate; heating rates of 10 °C/min. XRD patterns were obtained on two pieces of equipment: (A) A Panalytical X’Pert PRO powder diffractometer equipped with a Cu KR source (λ ) 1.540 56 Å) and an X’Celerator detector operating at 45 kV and 40 mA was used. Each sample was scanned between 2° and 50° in 2θ with a step size of 0.017° and a scan rate of 300 s/step. (B) A Debye-Scherrer INEL CPS-120 diffractometer equipped with a Cu KR source (λ ) 1.540 56 Å) and a 120° curved position sensitive detector operating at 40 kV and 30 mA was used. Each sample was scanned between 0° and 115° in 2θ with a step size of 0.029° and a scan rate of 3600 s/step. Raman spectra were collected using a Jobin Yvon T64000 instrument with NIR excitation radiation at 514 nm and a liquid nitrogen cooled bidimensional CCD detector. FT-IR spectra were recorded on a Bomem MB-120 IR spectrophotometer in KBr pellets at 1 cm-1 resolution from 350 to 5000 cm-1. Solid-state 13C NMR spectra were collected on a Varian Unity spectrometer operating at 75.4 MHz. The samples were spun in a 7 mm zirconia rotor and registered at room temperature (22 °C). Highresolution spectra were recorded using the CP/MAS method (crosspolarization/magic angle spinning) at 4000 rpm. A 2.5 s contact time and 3 s relaxation time was used. For norfloxacin A, 6700 scans were collected; for norfloxacin B, 5700 scans were collected. The resulting FIDs were processed with a line broadening of 1 Hz. Samples were referenced to hexamethylbenzene (methyl, 17.3 ppm).
Results and Discussion Preparation and Characterization. The DSC curve of the commercial product of norfloxacin shows an endothermic phenomenon at 219 °C with a heat of fusion of 122.5 J/g, as shown in Figure 2. The onset temperature coincides to the one reported by Sustar and Bukovec3 for form A, the heat of fusion being slightly higher than the reported value of 115.2 J/g. Thermogravimetric analysis of this sample shows no mass loss from room temperature to 230 °C, showing that this product is neither a solvated nor a hydrated form of norfloxacin. The X-ray powder diffraction (XRPD) pattern of this sample is similar to the one reported by Sustar and Bukovec3 for form A.
Barbas et al.
Figure 3. DSC curve of form B of norfloxacin carried out at the heating rate of 10 °C/min.
Figure 4. X-ray powder diffraction patterns of forms A and B of norfloxacin.
A manufacturing procedure for form B of norfloxacin has not been previously reported. After some crystallization trials, polymorph B was obtained by crystallization in 2-propanol and slow cooling to room temperature. The DSC thermogram of this sample shows two endotherms with onset temperatures at 212 and 220 °C (Figure 3). This thermogram is different from the one reported previously in the literature,3 where two endotherms were observed at 196 (solid-solid transition) and 220 °C (melting). The powder XRD pattern is different from the one described for form A (Figure 4). Significant differences are seen on some relevant peaks for both forms. Form A shows peaks at 2θ angles of 9.8°, 16.1°, 20.7°, and 22.7° that are not present in form B. The pattern for form B shows unique peaks at 2θ angles of 13.3°, 16.5°, 17.8°, 18.9°, 19.4°, 21.0°, 23.4°, 25.1°, and 27.4°. As can be seen from Figure 4, several other differences are also observed, thus making the diffraction patterns useful for form identification. No mass loss was observed in a thermogravimetric analysis, confirming that neither a hydrated form nor a solvated one is obtained. Moreover, forms A and B of norfloxacin have been characterized by means of FT-IR and Raman spectroscopy. Raman spectra of the pure samples of forms A and B in the range of 200-1700 cm-1 are presented in Figure 5. Some specific differences between the two forms can clearly be observed. For instance, the peaks at 212, 239, 454, 465, 750, and 1544 cm-1 are characteristic for form A and the peaks at 248, 735, 785, 1318, and 1582 cm-1 are characteristic for form B. Raman spectroscopy can measure inelastic scatterings of light, which have different wavelengths from the incident light. The differences in the wavelengths are related to the vibrational modes of the molecules.8 The vibrational modes of the polymorphs can be different; in this case, the Raman spectra are the fingerprints for the polymorphs.9 The large differences observed
Polymorphism of Norfloxacin
Crystal Growth & Design, Vol. 6, No. 6, 2006 1465 Table 1. NMR Chemical Shifts and Peak Assignments for Solid Forms of Norfloxacin, Together with Solution-State Data 13C
chemical shift δ (ppm)
assignment
form A
form B
solution2
C2, C6, C7 C3 C4 C4a C5 C8 C8a C9 C10 C11 CR,β
149.8,147.4 108.9 176.8 120.4 110.2 106.2 137.1 167.5 51.5 15.0 44.2-49.0
148.8, 151.5,146.4 117.7 174.3 124.1 111.4 105.6 137.7 169.6 49.6 16.1 40.0-45.3
149.5, 154.6, 146.1 108.2 177.5 121.5 113.1 106.9 138.6 169.9 51.1 14.9 44.7-47.7
Figure 5. Raman spectra of forms A and B of norfloxacin.
Figure 6. FT-IR spectra of forms A and B of norfloxacin.
Figure 7.
13
C CP/MAS NMR spectra of forms A and B of norfloxacin.
in Raman spectra of the two forms of norfloxacin can be caused by conformational differences of the molecules in each crystal lattice of A and B forms. FT-IR spectra in the range of 400-1800 cm-1 obtained from the pure samples of norfloxacin forms A and B are shown in Figure 6. These spectra are similar to the ones reported by Sustar and Bukovec3 for both polymorphs. Differences between the two polymorphic forms can be seen in the whole spectral region. For instance, characteristic peaks of form A appear at 1522, 1197, 942, 885, and 801 cm-1, whereas characteristic peaks of form B are observed at 1580, 1330, 1177, 919, and 737 cm-1. The 13C CP/MAS NMR spectra collected for the two polymorphic forms of norfloxacin are shown in Figure 7. Differences between the chemical shifts in the solid-state spectra of both forms are quite significant; for instance, the ketonic and the carboxylic carbons show a difference of 2 ppm,
while one aromatic carbon (C3) is shifted 9 ppm downfield in form B relative to form A. The reason the chemical shift of C3 varies dramatically from one crystalline form to another is due to the in plane versus out of plane conjugation of the carboxylic acid group. In polymorph A, the orientation of the carboxylate group with respect to the phenyl ring is out of plane (as expected in solution), whereas in polymorph B, it is in plane. Solid-state NMR peak assignments, made by comparing the chemical shifts observed in solution2 (CD3COOD), are summarized in Table 1. Norfloxacin has been described by the original manufacturer as a very hygroscopic compound.2 We have observed from DSC analysis that form A picks up moisture under ambient conditions after some days. However, polymorph B does not pick up moisture after months of exposure to ambient conditions. Relative Stability. A method described by Haleblian and McCrone can be used to determine the most stable polymorph at room temperature and to estimate the transition temperature.10 This method utilizes the fact that the most stable polymorph will also be the less soluble one at a given temperature and pressure. If crystals of both polymorphs are present in a saturated solution, the most stable form will grow at the expense of the less stable one. A small difference in solubility between two forms can require long time periods for the transformation. This method is called the solution-phase transformation or solventmediated transformation.11 Solvent-mediated transformation experiments were carried out by stirring a suspension of 1 g of form A of norfloxacin and 0.05 g of form B in 30 mL of DMF at room temperature. After 9 days, the suspension was filtered, and the solid was investigated by DSC and XRPD, concluding that after solvent-mediated polymorphic transformation, pure polymorph B was obtained. This could be a procedure for obtaining form B, only in cases where form A contains small amounts of form B. In the case of monotropy, the higher melting form is always the thermodynamically stable form. In the case of enantiotropy, the lower melting form is the thermodynamically stable form at temperatures below the transition point, and the higher melting form is the thermodynamically stable form at temperatures above the transition point.12 Polymorphs A and B are enantiotropically related because form B is the most stable form at room temperature (since it has been obtained by solvent-mediated transformation) but the less stable form at melting temperature. The DSC thermogram shows the melting endothermic process of form B followed by an exothermic process (crystallization of the higher melting form A, the most stable at melting temperature), which happens simultaneously, the net heat flow being smaller than that expected for the melting of form B (see Figure 3). Finally, melting of crystallized form A occurs. Occasionally,13 samples
1466 Crystal Growth & Design, Vol. 6, No. 6, 2006
Figure 8. DSC curve of polymorph B recorded at the heating rate of 10 °C/min. This thermogram shows rare occasions when the sample underwent partly a solid-phase conversion at 196 °C before melting.
Barbas et al.
Figure 10. DSC curves of polymorph B recorded at the heating rates of 1-40 °C/min. Table 2. Influence of the Heating Rate on the Onset Temperatures of the Solid-Solid Transition and the Meltings of Forms A and B
Figure 9. DSC curves of polymorph A mixed with small amounts of polymorph B, recorded at the heating rates of 1-40 °C/min.
of pure polymorph B show endothermic solid-solid transition of form B to form A, followed by melting of remaining form B and subsequent melting of form A (Figure 8). When polymorph A is mixed with small amounts of polymorph B, an endothermic phenomenon is observed at 196 °C in a DSC experiment at 10 °C/min, followed by melting of form A. Heating rate influences the onset temperature of the first endotherm, whereas the onset temperature of the melting of form A is not affected (Figure 9), thus confirming that the first endotherm is a solid-solid transition. Different heating rates have also been applied to form B of norfloxacin, showing that neither the first endotherm (melting of form B) nor the second one (melting of form A) are affected (Figure 10). The use of faster heating rates in an attempt to observe the pure melting of polymorph B was unsuccessful, thus preventing us from accurately describing its heat of fusion. Table 2 shows the influence of the heating rate on the onset temperature of the solid-solid transition (Figure 9), while onset temperatures of melting forms A and B remain invariable (Figure 10). Additional support for the conclusion that the pair A and B is enantiotropic is given by the “heat of transition rule” of Burger and Ramberger.7 The rule states in part that if an endothermal transition (for a solid-state conVersion) is obserVed at some temperature, it may be assumed that there is a transition point below it; i.e., the two forms are related enantiotropically. It is important to note that the commercial sample of norfloxacin used in this study is form A, which is metastable at room temperature. Usually the most stable polymorphic form is preferred in a marketed formulation, because any other polymorphs are metastable and may therefore transform to the
heating rate (°C/min)
transition
1 10 20 40
192.4 195.8 199.3 202.7
onset temperature (°C) form B 211.0 211.7 211.7 212.4
form A 220.3 220.5 220.6 221.4
more stable form. Overlooking the most stable polymorph may cause failure of a marketed product due to phase transformation during storage. A late-appearing stable polymorph can have a great impact on development timelines.14 Although metastable forms may survive years if a considerable activation energy barrier has to be overcome in moving from the metastable state to the stable state, this activation-energy barrier may be reduced by moisture, catalysts, impurities, excipients, or temperature, and the transformation into the stable form occurs spontaneously. Seeds of the stable form may also accelerate transformations.15 As it has been shown, a slurry of form A of norfloxacin with 5% of form B is fully converted into form B. Therefore, it is most important to fully dry the crystals because traces of residual mother liquor can induce a solvent-mediated polymorphic transformation during storage. Conclusions Norfloxacin has been described as a pair of enantiotropically related polymorphs. We have proven that the previous knowledge of this active pharmaceutical ingredient was erroneous due to a wrong identification of the system as a monotropic type. The results of solvent-mediated transformation experiments are in concordance with the fact that the observed solid-solid transition by DSC is endothermic. This evidence is enough to demonstrate the enantiotropic relationship between the two solid forms. This can be important because many commercial samples of norfloxacin are provided as the metastable form at room temperature, and then, undesirable transformations could occur. Acknowledgment. The authors thank the Unitat de Difraccio´ de Raigs X, Unitat d’Espectrosco`pia Molecular, and Unitat de Ressona`ncia Magne`tica Nuclear of the Serveis Cientı´ficote`cnics(University of Barcelona) for their contributions to this work. References (1) Chemburkar, S. R.; Bauer, J.; Deming, K.; Spiwek, H.; Patel, K.; Morris, J.; Henry, R.; Spanton, S.; Dziki, W.; Porter, W.; Quick, J.; Bauer, P.; Donaubauer, J.; Narayanan, B. A.; Soldani, M.; Riley, D.; McFarland, K. Org. Process Res. DeV. 2000, 4, 413-417.
Polymorphism of Norfloxacin (2) Mazuel, C. Anal. Profiles Drug Subst. 1991, 20, 557-600. (3) Sustar, B.; Bukovec, N.; Bukovec, P. J. Therm. Anal. 1993, 40, 475481. (4) (a) Katdare, A. V.; Ryan, J. A.; Bavitz, J. F.; Erb, D. M.; Guillory, J. K. Mikrochim. Acta Wien III 1986, 1-12. (b) Hu, T.; Wang, S.; Chen, T.; Lin, S. J. Pharm. Sci. 2002, 91, 1351-1357. (5) Florence, A. J.; Kennedy, A. R.; Shankland, N.; Wright, E.; Al-Rubayi, A. Acta Crystallogr., Sect. C 2000, C56, 1372-1373. (6) Giron, D. Thermochim. Acta 1995, 248, 1-59. (7) Burger, A.; Ramberger, R. Mikrochim. Acta I 1979, 259-271. (8) (a) Raman, C. V.; Krishnan, K. S. Nature 1928, 121, 501. (b) Lewis, I. R.; Edwards, H. G. M. Handbook of Raman Spectroscopy; Marcel Dekker: New York, 1999. (9) Ono, T.; ter Horst, J. H.; Jansens, P. J. Cryst. Growth Des. 2004, 4, 465-469. (10) Haleblian, F.; McCrone, W. J. Pharm. Sci. 1969, 58, 911-929.
Crystal Growth & Design, Vol. 6, No. 6, 2006 1467 (11) Carlton, R. A.; Difeo, T. J.; Powner, T. H.; Santos, I.; Thompson, M. D. J. Pharm. Sci. 1996, 85, 461-467. (12) Bernstein, J. Polymorphism in Molecular Crystals; IUCr Monographs on Crystallography, Vol. 14; Oxford Science Publications: Oxford, U.K., 2002. (13) Behme, R. J.; Brooke, D.; Farney, R. F.; Kensler, T. T. J. Pharm. Sci. 1985, 74, 1041-1046. In this reference, there is an example of a drug that occasionally shows different DSC (in some cases melting and crystallization, which occur simultaneously, and in other cases only melting). (14) Desikan, S.; Parsons, R. L.; Davis, W. P.; Ward, J. E.; Marshall, W. J.; Toma, P. H. Org. Process Res. DeV. 2005, 9, 933-942. (15) Giron, D.; Mutz, M.; Garnier, S. J. Therm. Anal. Calorim. 2004, 77, 709-747.
CG060101U
