Solid‐state Properties of Creatine Monohydrate

Solid‐state Properties of Creatine Monohydrate

Solid-State Properties of Creatine Monohydrate ALEKHA K. DASH,1 YOONSUN MO,1 ABIRA PYNE2 1 Department of Pharmacy Sciences, School of Pharmacy and Al...

296KB Sizes 1 Downloads 33 Views

Solid-State Properties of Creatine Monohydrate ALEKHA K. DASH,1 YOONSUN MO,1 ABIRA PYNE2 1

Department of Pharmacy Sciences, School of Pharmacy and Allied Health Professions, Creighton University, 2500 California Plaza, Omaha, Nebraska 68178 2

Department of Pharmaceutics, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota 55455

Received 7 August 2000; revised 8 November 2001; accepted 9 November 2001

ABSTRACT: Creatine monohydrate (CM) is a nutritional supplement and an ergogenic aid for athletes. It appears to increase lean body mass, high-intensity power output and strength in healthy humans. The crystal structure of creatine monohydrate has previously been reported. However, little information is available on its solid-state properties. In this investigation, creatine monohydrate was subjected to Thermal Analyses, Karl-Fisccher Titrimetry (KFT), Scanning Electron Microscopy (SEM), and Variable Temperature X-ray Powder Diffractometry (VTXRD) to characterize its solidstate properties. The results of this study suggested that commercially available creatine monohydrate dehydrates at about 97±1258C. A phase transition after dehydration was con®rmed by X-ray diffraction studies. This dehydrated phase at a temperature above 2308C undergoes intramolecular cyclization with a loss of an additional mole of water to form creatinine. Creatinine ®nally melts with decomposition at about 2908C. VTXRD, con®rmed that the above solid-state thermal transformation was kinetically driven, and occurred within a narrow temperature range. Mass Spectrometric (MS) studies further indicated a possible dimerization of creatinine formed during the solid-state transformation. ß 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci 91:708±718, 2002

Keywords: MS

creatine monohydrate; cratinine; thermal analyses; solid-state; VTXRD;

INTRODUCTION Creatine monohydrate is a nutritional supplement that is popular among athletes. Oral ingestion of creatine supplement has been shown to increase the creatine and phosphocreatine content in human muscle.1 The high concentration of creatine may increase the muscle's ability to maintain high ATP turnover rates during strenous exercise. Therefore, the creatine monohydrate supplement appears to enhance an individual's ability to maintain power output during highintensity exercise.1,2 Correspondence to: Alekha K. Dash (Telephone: 402-2803188; Fax: 402-280-1883; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 91, 708±718 (2002) ß 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association

708

Creatine is commercially available as creatine monohydrate (C4H11N3O3) with a molecular weight of 149.15. The crystal structure of creatine monohydrate was determined by Mendel and Hodgkin in 1954 using three-dimensional X-ray data that was analyzed utilizing least-square calculations.3 Jensen also reported the crystal structure of creatine monohydrate determined by using a two-dimensional work.4 A re®nement of the crystal structure of creatine monohydrate has also been reported by Kato and coworkers.5 This study determined positions of the hydrogen bonds and con®rmed the existence of a zwitterion structure in the solid state. Single-crystal neutron re®nement of creatine monohydrate has also been investigated.6,7 Despite such ®ndings, the solid-state properties of this compound have not yet been reported. Therefore, this

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

SOLID-STATE PROPERTIES OF CREATINE MONOHYDRATE

investigation deals with the characterization of the solid-state properties of creatine monohydrate using powder X-ray diffraction, thermal analyses, Karl Fischer titrimetry, SEM, HPLC and MS.

709

Karl Fischer Titrimetry

MATERIALS AND METHODS

The total water content of creatine monohydrate was determined using a Karl Fischer titrimeter (Model CA-05 Moisture Meter, Mitsubishi, Japan). Sample sizes for this study ranged from 5±10 mg.

Materials

Scanning Electron Microscopy (SEM)

Creatine (Aldrich, Milwaukee, WI), Creatinine (Acros Organic, Fair Lawn, NJ), Stearic Acid, HPLC Water, acetonitrile, sodium acetate, sodium hydroxide (Fisher Chemicals, St. Louis, MO) were used as received.

The crystal habit and surface morphology of the samples were determined using a Philips XL20 SEM. The voltage was set at 1 kV and the operation current at 200 mA. No coatings were applied to these samples. The scanning electron (SE) mode was selected to examine the overall surface morphology. A through lens detection (TLD) mode was selected to view the detailed structures. A back-scattered (BS) mode was selected to distinguish the phases containing various chemical compositions of creatine.

Thermal Analyses A differential scanning calorimeter (DSC) (model DSC-50, Shimadzu, Kyoto, Japan) and a thermogravimetric analyzer (TGA) (model TGA-50, Shimadzu, Kyoto, Japan) were connected to a thermal analysis operating system (TA-50WS, Shimadzu, Kyoto, Japan). The heat of fusion was calibrated using indium (purity 99.99%; m.p. 156.4; DH 6.8 mcal/mg). The sample to be analyzed (5±10 mg) by DSC was crimped nonhermetically in an aluminum pan and heated from 30 to 3208C at a rate of 108C/min under a stream of nitrogen (¯ow rate of 20 mL/min). For the thermogravimetric analysis (TGA), approximately 10 mg of the sample was weighed into platinum pans and heated from 30 to 3208C at a heating rate of 108C/min under nitrogen purge. Thermomicroscopy The thermomicroscopic studies were performed by placing about 0.1 mg of the sample with or without silicone oil on a glass slide covered by a glass cover slip. Samples were heated from 30 to 3208C at a programmed rate (108C/min). Physical changes (e.g., dehydration, phase transition, and melting) and chemical changes (e.g., degradation) were con®rmed visually. The hot stage used was a Mettler type (model FP900 Thermosystem, Mettler-Toledo AG, Greifensee, Switzerland). A Pt-90 central processor and Pt-100 temperature sensor were used, and a Nikonopitphot microscope (Nippon Kogaku, Garden City, NY) was attached to the hot stage. The hot stage microscope was calibrated with stearic acid. The observed melting temperature of stearic acid was 68 to 708C. This was in excellent agreement with the reported melting temperature of stearic acid (69±708C).

Variable Temperature X-Ray Powder Diffractometry (VTXRD) The powder patterns were obtained in a wideangle powder X-ray diffractometer (Model XDS 2000, Scintag; CuKa radiation; 45 kV  40 mA). The Bragg-Brentano focusing geometry was used, with a 1-mm incident slit, a 0.3-mm detector slit, and a solid-state germanium detector. Using a temperature controller attachment (Model 828D, Micristar, R.G. Hansen & Associates) mounted on the diffractometer, the sample could be subjected to a controlled temperature program ranging from 190 to ‡ 3008C. The samples were heated at 58C  min 1 or 108C  min 1 from room temperature up to 2708C. This permitted XRD patterns to be obtained at the desired temperatures. However, during the XRD runs, the samples (150 mg) were maintained under isothermal conditions for 10 min at the selected temperatures. The angular range was 5 to 408 2y, and data was collected in a continuous mode at chopper increments of 0.038 2y. When the scanning rate was 58 2y min 1, the entire scan took about 11 min. High-Pressure Liquid Chromatography (HPLC) Analysis An HPLC method was used to determine the solubility and stability of CM, CM samples heated to 1408C (after dehydration), and CM samples heated to 2608C. The chromatographic system JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

710

DASH, MO, AND PYNE

consisted of a GPM gradient pump and pulsed electrochemical detector interfaced to a Zenith PC through an AI-450 chromatography automation system from Dionex (Dionex Inc., Sunnyvale, CA). The mobile phase consisted of a mixture of water, acetonitrile, 0.01 M sodium acetate, and 1.0 M sodium hydroxide (2.5:2.5:90:5, v/v/v/v). The chromatographic separation was achieved at 458C on a (250  4.6 mm) polyhydroxylated glucose and sulfonated column (Jordi Glucose-DVB Column, Alltech, Deer®eld, IL) using a ¯ow rate of 1 mL/min. Solubility The solubility of the samples in water, at 378C, was determined using an equilibrium solubility method. The saturated samples of creatine monohydrate, and CM samples heated to 140 and 2608C were transferred into 100 mL of water. These solutions were shaken in a water bath (378C) over a period of 48 h. The supernatants were collected and the creatine content in the solution was determined by HPLC.

Figure 1. (a) DSC curve of creatine monohydrate (CM) heated from 30±3208C, and (b) TGA thermogram of CM heated from 30±3208C.

Mass Spectrometry Mass spectra were obtained on a Waters/Micromass ZMD, Quadropole instrument (Millford, MA) using electrospray ionization (ESI) mode. The evaporation temperature was 3508C, with 3± 4 V applied voltage and the sheath gas (N2) ¯ow was set at 400 L/h. All data were collected in the full scan mode (100±1000 m/z).

RESULTS AND DISCUSSION The DSC thermogram of creatine monohydrate resulted in four endothermic events with peaks at 102, 255, 270, and 2938C as shown in Figure 1a. The peak temperature is de®ned as the point on the temperature scale where maximum deviation from the baseline existed. The TGA curve of creatine monohydrate (Figure 1b) showed a twostep weight loss within the same temperature range. We were also interested in evaluating the reversibility of three initial endothermic events of the DSC curve. The results of this study are depicted in Figure 2. In this study, the sample was heated up to 2708C in a DSC, cooled to room temperature, and reheated from 30±3208C. The ®rst, second, and third endothermic peaks were JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

Figure 2. DSC curves of: (a) CM heated from 30± 3208C, (b) CM heated from 30±2708C, and (c) sample from (b) cooled to room temperature and reheated from 30±3208C.

SOLID-STATE PROPERTIES OF CREATINE MONOHYDRATE

absent in the DSC thermogram indicating that none of these events is reversible. When samples were heated in a TGA a weight loss of 11.9% occurred between 55 and 1258C. Stoichiometric calculation from the molecular formula of creatine monohydrate indicates that it should contain 12.07% (w/w) water. Therefore, the ®rst endothermic event was attributed to the dehydration of creatine monohydrate. The total water content of the creatine monohydrate was then determined by KFT and found to be 11.53  0.19% (w/w) (mean  SD; n ˆ 3). The result of the KFT study was in good agreement with the TGA results. This dehydration event was further con®rmed by thermomicroscopic studies. In this study, sample was placed on a glass slide with a drop of silicone oil and heated from 30±3208C at a rate of 108C/min. Figure 3a and b depict the thermomicrographs of creatine monohydrate sample, in silicon oil immersion, heated to 97 and 2158C, respectively. The liberation of bubbles

711

through the silicone oil provided direct visual evidence of water loss from the sample within this temperature range. The VTXRD patterns of creatine monohydrate sample heated to various temperatures are shown in Figure 4. The powder pattern of creatine monohydrate as depicted in Figure 4 was different from that of the sample heated to 1408C, i.e., after dehydration. The difference in the diffraction patterns suggests the appearance of a phase change after dehydration. Three cases of lattice structures can be distinguished after dehydration:8 (a) the dehydrated sample can be poorly crystalline, (b) the sample after dehydration can have a different crystal lattice, and (c) the crystal lattice of the dehydrated sample could be identical to that of the hydrate. Therefore, the VTXRD data indicated that dehydration of creatine monohyrate belongs to the second category. The SEM studies were also used to identify the difference in the crystal habits, if any, between these two solid

Figure 3. Thermomicroscopic photographs of creatine monohydrate subjected to: (a) 978C in a silicone oil immersion, (b) 2158C in a silicone oil immersion, (c) room temperature (before heating) without silicone oil, (d) 2308C without silicone oil, (e) 2558C without silicone oil, and (f) 3208C without silicone oil. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

712

DASH, MO, AND PYNE

phases and shown in Figure 5a and b. Figure 5a represents the crystal habit of the monohydrate at 50  magni®cation. Figure 5b represent the SEM pictures of the creatine monohydrate after heating to 1408C (after dehydration) at similar magni®cation. Both micrographs showed a clear

Figure 3. (Continued)

Figure 4. VTXRD patterns of creatine monohydrate at: (a) room temperature, (b) at 1408C, (c) at 2258C, (d) at 2458C, (e) at 2608C, and (f) at 2708C. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

Figure 5. Scanning electron micrographs of: (a) CM at 15  magni®cation, (b), CM heated to 1408C in a DSC at 15  magni®cation, (c) CM heated to 2458C in a DSC at 15  magni®cation.

SOLID-STATE PROPERTIES OF CREATINE MONOHYDRATE

difference in the crystal habit between the two phases. Figure 5c depicts the crystal habit of CM heated to 2458C. The second endothermic peak, which is represented as a shoulder in the DSC thermogram occurred between 230 and 2608C and shown in Figure 1a. During the thermomicroscopic studies without silicone oil immersion, it was observed that ®ne needle-like crystals appeared exactly at the same temperature range corresponding to the second endothermic peak. Figure 3c represents the thermomicroscopic picture of creatine monohydrate prior to heating, and Figure 3d indicates the formation of the needle like crystal at a temperature of 230±2368C. This second endothermic peak was thought to be due to a possible solidsolid polymorphic transition of anhydrous creatine. As mentioned earlier, thermomicroscopic photograph in Figure 3d and VTXRD patterns, between 230±2608C as shown in Figure 4 provided supporting evidence for this claim. The Xray diffraction patterns of creatine monohydrate samples heated to 1408C, i.e., after dehydration and prior to the second endothermic peak at 2258C in the DSC curve were identical as shown in Figure 4. However, when the samples were heated to 245 and 2608C and subjected to powder X-ray analysis, appearance of some new peaks and decrease in the intensities of some existing peaks were clearly evidenced. These results, in conjunction with the thermomicroscopic and DSC studies, suggested possible existence of a polymorphic transition of anhydrous creatine. However, solid-solid polymorphic transitions are

Scheme A.

713

usually weak, i.e., the enthalpy of transition is small. In Figure 1a the second endotherm is not really small. Second, TGA weight loss begins to occur at around 2308C. If indeed it is a polymorphic transition around 2308C, one should not expect a weight loss during this transition. Therefore, an alternative pathway for this transformation was hypothesized and tested. Both pathways: (I) solid-solid phase transition, or the (II) solid-state thermal transformation are presented in Scheme A. Creatine has been reported to be converted to creatinine in acidic solution.9 However, solidstate thermal stability of creatine monohydrate has not yet been reported. Solid-state reactions can be classi®ed into two major categories: (1) physical transformations including polymorphism and desolvation, and (2) chemical transformation including chemical and photochemical reactions.10 Solid-state thermal stability of model dipeptides has been reported by Leung and Grant.11 According to Pathway II, creatine monohydrate dehydrates around 1008C to form the anhydrous creatine. This dehydrated phase at temperatures above 2308C loses another molecule of water and undergoes intramolecular cyclization to form creatinine. In the DSC curve (Figure 1a), the ®rst endothermic peak was due to dehydration of creatine monohydrate to form the anhydrous creatine. The overlapping second and third endotherms can be explained due to the intramolecular cyclization of creatine with a loss of 1 mole of water (formation of creatinine). This cyclization event is expected to be highly

Two possible pathways of solid-state transformation of creatine monohydrate. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

714

DASH, MO, AND PYNE

Scheme B. Possible pathways of dimerization of creatinine during the thermal transformation of creatine to creatinine.

energetic and endothermic. The last and ®nal endotherm can be attributed to melt-decomposition of creatinine. According to the TGA (Figure 1b) curve, the ®rst weight loss is due dehydration of creatine monohydrate. The second weight loss begins at 2308C, exactly at the temperature where loss of one molecule of water occurs during the intramolecular cyclization of creatine. Because these two events are occuring simultaneously, one should expect overlaps of these two endothermic events. The third endotherm can be attributed to the melt-decomposition of creatinine. Comparison of the two pathways presented in Scheme A, one can conclude that Pathway II explains our experimental data better than Pathway I. To con®rm this solid-state transformation of creatine to creatinine, the following experiments JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

were then carried out. Creatinine was subjected to VTXRD and the powder patterns are shown in Figure 6. No differences in the powder patterns were noticed at different temperatures. The powder patterns of creatine monohydrate heated to 2608C and creatinine samples heated to the same temperature were then compared and presented in Figure 7. Both the powder patterns were found to be identical which, con®rms the solid-state transformation of creatine to creatinine around 230±2608C. The above two samples (CM and creatinine heated to 2608C) were subjected to elemental analysis and mass spectrometry. The elemental analysis results for both the samples were found to be almost identical and close to the theoretical values as shown in Table 1. The mass spectra of creatine monohydrate heated to 2608C and creatinine heated to the

SOLID-STATE PROPERTIES OF CREATINE MONOHYDRATE

Figure 6. VTXRD patterns of creatinine at: (a) room temperature, (b) at 608C, (c) at 1208C, and (d) at 2608C.

same temperature are shown in Figure 8. Both mass spectra were identical. Interestingly, both the samples showed a peak around m/z of 227. The appearance of this peak in the mass spectra can be explained due to the formation of dimers and presented in Scheme B.

Figure 7. Comparison of the X-ray diffraction patterns of creatine monohydrate heated to 2608C versus creatinine heated to 2608C.

715

To evaluate the kinetic of the solid-state transformation of creatine into creatinine, the samples were subjected to VTXRD at two different heating rates (108C/min and 58C/min). The extent of transition of creatine to creatinine was found to be slower when the heating rate was faster. This could be explained by two possible contributing factors (1) thermal lag and/or (2) the nature of the kinetics of the transition. At about 2458C, the intensities of some new peaks (due to the formation of creatinine) were increased while intensities of previous peaks formed after dehydration (due to creatine) showed a substantial decrease (Figure 4). We also obtained continuous scans of creatine monohydrate samples in a VTXRD holding it isothermally at 2458C. The results of this study (data not shown) indicate that the conversion of creatine to creatinine was not instantaneous but was complete at around 2608C. Repeated scans at 2458C shown in Figure 9 indicated that as the holding time was increased, the solid-state thermal transformation was more pronounced and complete. For increased clarity, zoomed in overlaid patterns of creatine monohydrate samples were obtained isothermally at 2458C (data not shown). This overlaid plot clearly indicated that the peak intensity at 17.48 2y decreased and peak intensity at 16.98 2; increased as the holding time at 2458C was increased. The conversion of creatine to creatinine at about 2458C was also visually con®rmed by thermomicroscopy (Figure 3c and d). In this study, the tiny needle-like crystals began to appear at approximately 2308C. This was thought to be due to internal cyclization with a removal of one mole of water to form creatinine. The solubility of creatine monohydrate, CM samples heated to 140 and 2608C, respectively, were carried out in triplicate at 378C. The solubility of creatine monohydrate at room temperature, CM samples heated to 1408C (anhydrous creatine), and CM samples heated to 2608C (after the solid-state thermal transformation) were 27.97  5.12 mg/mL, 69.67  5.10 mg/mL and, 79.46  4.60 mg/mL (mean  SD; n ˆ 3), respectively. Because the anhydrous form has a higher solubility than the monohydrate, one should expect a higher solubility for creatine monohydrate heated to 1408C (anhydrous form) compared to the parent CM.12,13 Because we have already con®rmed that creatine is converted to creatinine around 230±2608C, one should also expect a higher solubility for CM sample heated to 2608C (which is virtually creatinine) compared JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

716

DASH, MO, AND PYNE

Table 1. Elemental Analysis of Creatine Monohydrate and Creatinine Heated to 2608C Creatine Monohydrate Heated to 2608C Elemental Analysis for C H N

Creatinine Heated to 2608C

Theoretical (%)

Experimental (%)

Theoretical (%)

Experimental (%)

42.47 6.24 37.15

41.97 6.11 35.91

42.47 6.24 37.15

42.84 6.04 36.06

to CM. Creatinine is known to be more water soluble than creatine monohydrate.14 The creatinine formed undergoes melt decomposition at around 280±3208C. Therefore, the last endothermic peak corresponds to the melting with decomposition of creatinine. Thermomicroscopic photographs of creatine monohydrate heated to 255 and 3208C are shown in Figures 3e and f respectively. Based on the direct visual evidence provided by thermomicroscopy, the ®nal endotherm was further con®rmed to be due to the

decomposition of creatinine around 280±3208C as depicted in Figure 3f. The weight loss shown in the TGA curve (starting from 2308C until the ®nal melt decomposition) corresponding to the second, third, and ®nal endothermic peaks of the DSC curve can also be explained as a two-step process. The ®rst part of the weight loss corresponds to the loss of an extra mole of water from anhydrous creatine during intramolecular cyclization to form creatinine. The second part of the weight loss was due to

Figure 8. Mass spectra of: (a) creatine monohydrate heated to 2608C and (b) creatinine heated to 2608C. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

SOLID-STATE PROPERTIES OF CREATINE MONOHYDRATE

717

CONCLUSIONS Creatine monohydrate, dehydrates to anhydrous creatine at approximately 97±1258C with a subsequent phase change. Anhydrous creatine undergoes intramolecular cyclization along with a loss of 1 mole of water to form creatinine. Creatinine formed from this solid-state thermal transformation undergoes melting with decomposition around 280±3108C. This solid-state thermal transformation was con®rmed with VTXRD, HPLC, and Mass spectrometry. Mass spectrometric data further reveals a possible dimerization of creatinine during this thermal event.

ACKNOWLEDGMENTS Figure 9. Repeated scans at 2458C of creatine monohydrate from VTXRD study with a holding time of: (a) 0 min, (b) 12 min, (c) 24 min, (d) 36 min, (e) 48 min, and (f) 60 min.

the decomposition of creatinine at a higher temperature. The solid-state thermal conversion of creatine to creatinine was ®nally con®rmed by HPLC. The chromatograms of creatine monohydrate, creatine monohydrate heated to 2608C and physical mixture of cratine monohydrate and creatine monohydrate heated to 2608C are shown in Figure 10. Creatine monohydrate heated to 2608C had a retention time of 4.57 min compared to 3.40 min for creatine monohydrate. Moreover, creatinine sample heated to 2608C and analyzed by HPLC, had a retention time of 4.6 min, which is identical to the retention time of creatine monohydrate heated to 2608C. Therefore, HPLC studies further con®rmed the thermal conversion of creatine into creatinine.

Figure 10. Representative chromatogram of: (a) Creatine monohydrate, (b) creatine monohydrate heated to 2608C, and (c) physical mixture of CM and CM heated to 2608C.

The authors would like to thank Fortress Systems Int. for ®nancial support. We also wish to thank Dr. David Dobberphul, Department of Chemistry, Creighton University, for his help in HPLC studies. This work was presented, in part, at the Annual Meeting of the American Association of Pharmaceutical Scientists, November 1999, at New Orleans, LA.15

REFERENCES 1. Volek JS, Kraemer WJ. 1996. Creatine supplementation: Its effect on human muscular performance and body composition. J Streng Cond Res 10:200± 210. 2. Guerrero-Ontiveros ML, Wallimann T. 1998. Creatine supplementation in health and disease. Effects of chronic creatine ingestion in vivo: Down regulation of the expression of creatine transporter isoforms in skeletal muscle. Mol Cell Biochem 184:427±437. 3. Mendel H, Hodgkin DC. 1954. The crystal structure of creatine monohydrate. Acta Crystallogr 7:443± 446. 4. Jensen LH. 1954. The crystal structure of creatine monohydrate. Acta Crystallogr 8:237±240. 5. Kato Y, Haimoto Y, Sakurai K. 1979. A re®nement of the crystal structure of creatine monohydrate. Bull Chem Soc Jpn 52:233±234. 6. Wilson CC, Shankland N, Florence AJ, Frampton CS. 1997. Single-crystal neutron diffraction of bioactive small molecules. Phys B Condensed Matter 234:84±86. 7. Frampton CS, Wilson CC, Shankland N, Florence AJ. 1997. Single-crystal neutron re®nement of creatine monohydrate at 20 K and 123 K. J Chem Soc Faraday Trans 93:1875±1879. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

718

DASH, MO, AND PYNE

8. Garner WE. 1955. The kinetics of endothermic solid reactions. In: Garner WE, editor. Chemistry of the solid state. New York: Academic Press. p 214. 9. Edgar G, Shiver HE. 1925. The equilibrium between creatine and creatinine in aqueous solution. The effect of hydrogen ion. J Am Chem Soc 47:1170±1188. 10. Byrn S, Pfeiffer RR, Stowell JG. 1999. A solid state chemistry of drugs. New York: Academic Press. pp. 29±38, 307±313. 11. Leung SS, Grant DJW. 1996. Solid state stability studies of model dipeptides: Aspartame and aspartylphenylalanine. J Pharm Sci 86:64±71.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 3, MARCH 2002

12. Grant DJW, Brittain HG. 1995. Solubility of pharmaceutical solids. In: Brittain HG, editor. Physical characterization of pharmaceutical solids, 1st ed. New York: Marcel Dekker. pp. 321± 386. 13. Shefter E, Higuchi T. 1963. Dissolution behavior of crystalline solvated and nonsolvated forms of some pharmaceuticals. J Pharm Sci 52:781±790. 14. Budavari S. 1989. The Merck index, 11th ed. Rahway, NJ: Merck and Co. Inc. p 2577. 15. Mo Y, Carnazzo J, Dash AK. 1999. Solid-state properties of creatine monohydrate. AAPS PharmSci 1:S±182.