Investigation of Moricizine Hydrochloride Polymorphs

Investigation of Moricizine Hydrochloride Polymorphs LEI-SHUWU", GEORGE TOROSIAN~, KENNETHSIGVARDSON*,

CHRISTINE

GERARD',AND MUNIRA. HUSSAIN'

Received July 26, 1993, from the *The DuPont Merck Pharmaceutical Company, Experimental Station, P.0. Box 80400, Wilmington, DE 19880-0400, and #The DuPont Merck Pharmaceuticai Company, 7000 Steward Avenue, Garden Cify, NY 11530. Accepted for publication May 12, 1994@. Abstract 0 The antiarrhythmic agent moricizine hydrochloride exhibits a single melting-decomposition endotherm peak at temperatures ranging from 209 to 214.5 "C (Form I) when recrystallizedfrom polar solvents, as determined by differentialscanning calorimetric analysis. However, a different polymorphic form (Form II), with a differential scanning calorimeteric melting-decomposition peak temperature of 190 "C,was generated by recrystallizing moricizine hydrochloride from nonpolar solvents. These two polymorphic forms can be reversibly converted to one another by selecting recrystallization solvents. The existence of these polymorphs was confirmed by Fourier transform IR microscopy, X-ray powder diffractometry, and solution calorimetry. Polymorphic Form I exhibited a slightly slower initial dissolution rate than Form 11, which correlated well with heats of solution data (less heat needed to dissolve Form 11). A simulated wet granulation process did not change the polymorphic form, suggesting that wet granulation is feasible for tablet preparation.

Moricizine hydrochloride is an orally active antiarrhythmic agent (phenothiazine derivative) with potent local anesthetic

CtiD

0 II

NH - c- OCH,CH,

0 =c

n

' CH,CH,-~

\

'0 *HCl

/

activity and myocardial membrane stabilizing effect. Results of electrophysiologic indicate that moricizine hydrochloride has properties similar to class 1A and 1B antiarrhythmic agents, with minimal effects seen on the surface electrocardiogram. Polymorphs are the same chemical with different molecular arrangements within the crystal l a t t i ~ e .The ~ pharmaceutical importance of polymorphism was reviewed by Haleblian and M ~ C r o n e .The ~ impact of polymorphism on the physicochemical proper tie^^-^ and bioa~ailabilit?,~of drugs was demonstrated and is of a great concern in the pharmaceutical industry. In this story, polar and nonpolar solvents were used t o recrystallize moricizine hydrochloride in a n attempt t o discover different polymorphs. X-ray powder diffractometry (XRPD) was used t o prove the presence of polymorphs and Fourier transform IR (FTIR) microscopy, differential scanning calorimetry (DSC),and solution calorimetry techniquedo were used to confirm the polymorphs. Differences in the dissolution rate of these polymorphs and the effect of the wet granulation process on altering the crystalline form of the polymorphs were also investigated. @Abstractpublished in Advance ACS Abstracts, July 1, 1994.

1404 / Journal of Pharmaceutical Sciences Vol. 83, No. 10, October 1994

Experimental Section Moricizine hydrochloride [hydrochloride salt of 2-carbethoxyamino1O-(3-morpholylpropionyl)-phenothiazine;lot Y-3946-57 recrystallized from ethanol; the DuPont Merck Pharmaceutical Company], with purity of 98.6%, was used for recrystallization. Polymorphs I and I1 were prepared by dissolving excess moricizine hydrochloride, with heat applied and nitrogen purged, in the following solvents: reagent grade acetone, acetonitrile, ethanol, methylene chloride, and methylene chloride/ethyl acetate for 10 min. The solution was then filtered and cooled down to allow recrystallization to occur a t either room temperature (22 4= 0.4 "C and 25 f.0.5 "C) or under refrigeration (4 "C). Existence of polymorphic forms was confirmed by DSC, FTIR, XRPD, and solution calorimetry. The volatile content was determined by thermogravimetric analysis (TGA). DSC-A TA Instrument DSC (model 910) equipped with thermal analyzer (model 1090) were used. Moricizine hydrochloride was placed in a hermetically sealed pan with its cover reversed and heated under a nitrogen stream from 50 to 250 "C at a heating rate of 10 "Clmin. Thermal transitions were recorded. F'I3R-A nondestructive microtransmission FTIR microscope (model AQS-20 and AQM-515 from Analect Inc.) was employed to obtain an IR spectrum of moricizine hydrochloride. The crystalline powder was mounted on a KC1 plate and examined under the microscope. XRPD-A modified Philips X-ray diffractometer (APD 3500)with generator (XRG 3000)was used to examine moricizine hydrochloride crystals. A selected receiving slit of 0.2 mm and CuK, radiation (40 kV,30 mA) were employed. The sample was scanned from angle (219) 4 to 50" in 0.02" increments. The intensity of the diffracted radiation was automatically detected every 10 s by a scintillation detector. Solution Calorimetry-A Tronac calorimeter (model 450) was used to determine the heat of solution. About 150 to 200 mg of moricizine hydrochloride was packed into a 1-mL glass ampule, which was then sealed and placed in a holder equipped with hammer and stirring function. The device was then inserted into the reaction vessel containing solvent (water or dimethylacetamide). The study was performed in a 29 "C water bath. The glass ampule was broken after the calibration baseline was established, and the heat of solution was recorded. TGA-A TA Instrument TGA (model 951) equipped with thermal analyzer (model 1090) was used. The drug substance was heated under a nitrogen stream from 30 to 170 "C at a heating rate of 10 W m i n . The weight loss representing the volatile content was recorded. Granulations of moricizine hydrochloride were prepared by adding an appropriate amount of water to moricizine hydrochloride powder and blending it with a pestle. The moistened granules were then sieved through a #12 mesh screen. The granules were dried in a 35 "C vacuum oven overnight. The dissolution rate study was conducted with Wood's rotating disk dissolution apparatus at 100 rpm and room temperature (25 & 0.5 "C). The shafts were equipped with dies, which allowed a constant surface of the moricizine hydrochloride pellet to be exposed to the dissolution medium (900 mL of water). The moricizine hydrochloride pellet was prepared by pressing 300 mg of the compound with 5000 pounds force for 5 min. The drug concentrations at various dissolution intervals were analyzed with a spectrophotometer a t a wavelength of 268 nm.

Results and Discussion Moricizine hydrochloride recrystallized from polar solvents (ethanol, acetone, and acetonitrile) exhibited a DSC melting-

0022-3549/94/1200-1404$04.50/0

0 1994, American Chemical Society and American Pharmaceutical Association

Table 1-Thermal Properly (DSC) and Volatile Content (TGA) for Moricizine HydrochloridesRecrystallized from Different Solvents

RecrystallizationSolvent

DSCPeak Temperature, "C

TGA% Residual

Ethanol Acetone Acetonitrile (at RT) Acetonitrile (at 4 "C) Methylene chloride Methylene chloridelethylacetate Methylene chloride, then ethanol Methylene chloride/ethylacetate,then ethanol

213.1 210.5 212.8 213.2 190.0 189.6 213.7 212.4

100.4 100.3 100.2 100.2 99.3 100.1 NDb NDb

a

*I--

u)

z w

IZ

-

In this study, the starting material is Form I. ND = not determined.

11A

I

:I

Y

I

4.0

8.0

12.0

16.0

20.0 24.0 28.0 32.0

20 in degrees Figure 2-X-ray powder diffraction patterns of moricizine hydrochloride recrystallized from (A) absolute denatured ethanol (Form I) and (B) methylene chloride (Form 11). Four major regions at 26 of 5-6", 10-12", 15-17", and 20-22" are different for these two forms. Table 2-Heats of Solution of Moricizine Hydrochloride Polymorphs in Water and Dimethylacetamide

Heat of Solution, kcallmol I

80

~

1

120

'

I

'

200 Temperature(% ) 160

I

'

'

I

Solvent

~

240

wateP

Figure 1-DSC thermograms of moricizine hydrochloride recrystallized from (A) absolute denatured ethanol (Form I) and (B) methylene chloride (Form 11).

decomposition endotherm peak a t a temperature around 213 "C (Table 1, Figure 1A); however, a melting-decomposition endotherm at 190 "C (Table 1, Figure 1B) was obtained for material recrystallized from nonpolar solvents (methylene chloride or methylene chloride/ethyl acetate). No other DSC endotherm and no TGA weight loss (Table 1) for both materials indicates that they are not hydrates or solvates. The DSC and TGA results suggest that the difference in melting points is due t o moricizine hydrochloride with different crystalline forms (designated as Form I from polar solvents, and Form I1 from nonpolar solvents). The XRPD patterns of Forms I and I1 from 28 of 4 to 34" are shown in Figure 2. Four major regions at 28 of 5-6", 1012", 15-17", and 20-22" are different for these two forms. The XRPD is one of the most direct and widely used techniques to identify the morphology differences of crystalline powders because different polymorphs have different molecular arrangements in the crystal lattice. The FTIR spectra of Form I1 (from nonpolar solvents) is different from that of Form I in the oxygen- and nitrogenassociated regions. The most pronounced shift of absorption is in the regions representing the carbonyl groups. Two clearly separated peaks at 1664 and 1736 cm-l for Form I (Figure 3A) were shifted to a broader and closer double

Mean f SD Dimethylacetamideb Mean f SD a

AY

MI

6.7 7.1 6.5 7.1 6.9 f 0.3 1.4 1.3 1.4 1.4 f 0.1

5.6 6.2 5.8 6.1 5.9 f 0.3 0.5 0.4 0.5 0.5 f0.1

A& = AH - AH, = 6.9 - 5.9 = 1.0. A& = AH - AH1 = 1.4 - 0.5

= 0.9.

absorptions at 1704 and 1724 cm-' for Form I1 (Figure 3B). The strong dipole moment of the carbonyl group can be easily affected by environment (i.e., polar or nonpolar solvent) during the crystallization process. Some minor absorption shift at 2350-2700 cm-l representing tertiary amine salt absorption and 1200-1250 cm-l representing C-0 absorption also suggests the difference between the two forms of drug substance. The heats of solution for polymorphs I and I1 when dissolved in water and dimethylacetamide are summarized in Table 2. The individual heats of solution are different for these two polymorphs. However, an insignificant difference in heat of transition using either water (A& is 1.0 kcal/mol) or dimethylacetamide (AHTis 0.9 kcal/mol)was observed. Because polymorphs are identical chemically and different in the crystal structure, their heats of solution (AH1 and AH111 are dependent on the solvents used. However, the heat of Journal of Pharmaceutical Sciences / 1405 Vol. 83, No. 10, October 1994

-E \

m 3

-

200

-

C

.-0

.cI

L

.cI

C

I

Q)

I

0 C 0

I

100-

0

04 0

1

1

.

20

I

.

40

I

60

Time,

-

I

80

-

1

100

.

1

120

.

1

140

minute

Figure 4-Dissolution profiles in water at room temperature (25 f 0.5 "C)for moricizine hydrochloride polymorphs recrystllized from absolute denatured ethanol (Form I: 0) and methylene chloride (Form II; 0).

I

4000

I

3000

I

I

.

2000 1700

I

1300

.

1

900

.

500

Cm-1 Figure 3-FTIR spectra of moricizine hydrochloride crystallized from (A) absolute denatured ethanol (Form I) and (B) methylene chloride (Form 11). The most pronounced shift is at wavelength of 1664-1736 cm-'.

transition (AHT = AHHI- MII) from Form I to Form I1 is independent of the solvent. The microcalorimetry results confirm that these two forms are polymorphs. The heats of solution of these two forms indicate that Form I is more endothermic than Form 11. Therefore, Form I is more thermodynamically stable than Form 11, which is consistent with the higher melting temperatures for Form I. Because moricizine hydrochloride decomposes immediately aRer melting, the heat of transition determined by solution calorimetry is possibly the only direct method to obtain thermodynamic evidence for the morphology difference. Form I1 of moricizine hydrochloride, generated by recrystalling Form I with methylene chloride, can be converted back to Form I by recrystallizing Form I1 once again from ethanol. The DSC thermograms, FTIR spectra, and XRPD patterns of the crystals converted from Form I1 with ethanol are identical to those of Form I. This demonstrates that Forms I and I1 can be reversibly converted from one to another by selection of the recrystallization solvent. The dissolution profiles for these two polymorphs are given in Figure 4. The initial dissolution rates for Forms I and I1 were 3.2 and 4.3 mg/min, respectively, which correlate well

1406 /Journal of Pharmaceutical Sciences Vol. 83, No. 70, October 1994

with heats of solution data (less heat needed to dissolve Form 11, Table 2). Moricizine hydrochloride, obtained by blending Form I with water and drying in a vacuum oven (35"C) overnight, gave the same DSC melting peak temperature as that of Form I. This suggests that the wet granulation process for preparing the tablet dosage form does not alter the polymorphic form. Therefore, wet granulation appears to be feasible for tablet processing. In conclusion, moricizine hydrochloride exhibits two polymorphic forms. The existence of these polymorphs was demonstrated by DSC, XRPD, FTIR, and solution calorimetry. Polymorphic Form I1 exhibits a slightly more rapid initial dissolution rate than Form I. No polymorphic conversion was observed during a simulated granulation process, suggesting that wet granulation is feasible for tablet processing.

References and Notes 1. Moss, A. J.; Rivers, R. J., Jr. Circulation 1978, 57, 103-106. 2. Morganroth, J.; Michelson, E. L.; Kitchen, J. G.; Dreifus, L. S. Circulation Supplement 1981, 64, VI-263. 3. Bym, S. Solid-state Chemistry ofDrugs;Academic: New York, 1982. 4. Haleblian, J.; McCrone, W. J . Pharm. Sci. 1969, 58, 911-929. 5. Doherty, C.; York, P. Znt. J . Pharm. 1988,47,141-155. 6. Macek, T. J. Am. J . Pharm. 1969,137, 217-238. 7. Hamlin, W. E.; Nelson, E.; Ballard, B. E.; Wagner, J. G. J . Pharm. Sci. 1962,51,432-435. 8. Ballard, B. E.; Nelson, E. J . Pharrnacol. Exp. Ther. 1962,135, 120-127. 9. Mullins, J. D.; Macek, T. J. J . Pharm. Sci. 1960,49, 245-248. 10. Ip, D. P.; Brenner, G. S.; Stevenson, J. M.; Lindenbaum, S.; Douglas, A. W.; Klein, S. D.; MeCauley, J. A. Znt. J . Pharrn. 1986,28, 183-191.