J Infrared Milli Terahz Waves (2012) 33:953–962 DOI 10.1007/s10762-012-9910-1

Characterization of Poly-Amorphous Indomethacin by Terahertz Spectroscopy Makoto Otsuka & Jun-ichi Nishizawa & Naomi Fukura & Tetsuo Sasaki

Received: 12 November 2011 / Accepted: 2 May 2012 / Published online: 15 May 2012 # Springer Science+Business Media, LLC 2012

Abstract Since the stability of amorphous solids of pharmaceuticals differs depending on the method of preparation, there are several solid-state chemical structures in amorphous solids, which like poly-amorphous solids might have different characteristics the same as in crystalline solids. However, it is not easy to identify the differences in solid-state characteristics between amorphous solids using conventional analytical methods, such as powder Xray diffraction analysis, since all of the poly-amorphous solids had similar halo X-ray diffraction patterns. However, terahertz spectroscopy can distinguish the amorphous solids of indomethacin with different physicochemical properties, and is expected to provide a rapid and non-destructive qualitative analysis for the solid materials, it would be useful for the qualitative evaluation of amorphous solids in the pharmaceutical industry. Keywords Poly-amorphous form . Polymorphs . Terahertz spectroscopy . Physicochemical properties . Indomethacin . Material history

1 Introduction Validation of production processes is required to ensure the manufacture of safe and efficacious pharmaceutical products in order to meet regulatory requirements [1]. Therefore, an accurate assessment of the physicochemical properties of bulk powder materials is required for the reproducible preparation of pharmaceutical products. Polymorphs and solvates of crystalline bulk drug powders of pharmaceuticals differ in physicochemical stability, processing characteristics, dissolution rates, etc. [2]. Analytical methods for the screening of polymorphic forms include powder X-ray diffraction (XRD) analysis, thermal gravimetric analysis, differential scanning calorimetry (DSC) and infrared spectroscopy [3– M. Otsuka (*) : N. Fukura Research Institute of Pharmaceutical Sciences, Faculty of Pharmacy, Musashino University, 1-1-20 Shinmachi, Nishi-Tokyo 202-8585, Japan e-mail: [email protected] J.-i. Nishizawa : T. Sasaki Semiconductor Research Institute, Sophia University, 7-1 Kioi-cho, Chiyoda, Tokyo 102-8554, Japan

954

J Infrared Milli Terahz Waves (2012) 33:953–962

5]. However, these methods are too time-consuming to be used in the preparation of samples and/or their measurements. On-line analyses to measure pharmaceutical properties of crystalline forms as process analytical technology (PAT) by the Food and Drug Administration [6], hold the promise of reducing or eliminating reworked batches, increasing manufacturing efficiency, decreasing the burden of finished product testing, and ensuring product quality throughout the pharmaceutical manufacturing process. Since spectrophotometric methods, such as Raman spectroscopy [7] and near-infrared spectroscopy [8], are simple, rapid, and nondestructive, they are applied to pharmaceutical analyses in the industry as tools of PAT. However, Raman spectra measure the surface of samples, but not the inside. Absorption peaks of NIR spectra are also broad and overlap, and it is not easy to assign chemical functions of the molecule. In manufacturing, the formation of amorphous solids can affect the pharmaceutical properties of the products, such as hygroscopicity, dissolution rate, tablet hardness and chemical stability [9]. Since the stability of amorphous solids differs depending on the method of preparation, there are several solid-state chemical structures in amorphous solids [10], which like poly-amorphous solids might have different characteristics the same as polymorphs in crystalline solids. However, it is not easy to identify the differences in solidstate characteristics between amorphous solids obtained by different methods using conventional analytical methods, such as XRD. The detection of terahertz radiation waves (THz), which have a frequency of between 0.1 and 10 THz, is potentially very useful in probing intermolecular-level long range without sample contact and destructive treatment when characterizing solid-state materials, since it can induce low frequency bond vibrations, crystalline phonon vibrations, hydrogen-bonding stretches, and torsion vibrations [11]. The probing of intermolecular interactions makes THz spectroscopy an ideal method of measuring polymorphic content in pharmaceuticals, where a molecule exhibits the ability to pack in two or more crystal forms [12]. Changes in molecular packing between polymorphs result in differences in intermolecular interactions, which have been shown to lead to dramatic differences in THz spectra [13]; THz spectroscopy is expected to enable the rapid identification of polymorphic content in crystalline pharmaceuticals [14–16]. Nishizawa et al., [11, 17] developed a THz Spectrometer using a frequency-tunable THz signal generator based on difference-frequency generation in GaP crystals. Since the generator generates THz radiation with a wide and stable wave range and high power, it is possible to make accurate measurements with low noise [18–24]. They applied the THz system to the qualitative evaluation of polymorphic solids, and reported a sufficiently accurate calibration model to measure polymorphs of mefenamic acid [25]. However, there is no report to measure amorphous solids by THz spectroscopy. The purpose of the present study is to establish a discrimination method between different amorphous states of pharmaceuticals, such as poly-amorphous solids, using indomethacin (IMC) as a model drug, by THz spectroscopy.

2 Materials and methods 2.1 Materials A bulk powder of the γ− form (CG, stable form) of indomethacin (IMC) was obtained from Sumitomo Pharmaceutical Co., Ltd., (Japan). The α-form (CA, meta-stable form) of IMC

J Infrared Milli Terahz Waves (2012) 33:953–962

955

was prepared by recrystallization from ethanol, and the precipitated crystals were dried [26]. Amorphous IMC (AQ) was prepared as follows [9]: The bulk powder (5 g) of CG put on the 100 mL-beaker, and heated at 170°C for 15 min, and then, the melt product was poured into 300 mL-glass beaker containing about 100 mL of liquid nitrogen. After all liquid nitrogen distilled out, the remaining glassy product was ground in an agate mortal by an agate pestle. Amorphous IMC (AS) was prepared as follows [9]: The bulk powder (5 g) of CG put on the 100 mL-glass beaker, and heated at 170°C for 15 min, and then, the beaker containing the melt product was stored and cooled at −20°C in a freezer for 4 h, and the obtained glassy product was ground in an agate mortal as the same as above section. Amorphous IMC-G (AG) was prepared as follows: The bulk powder (5 g) of CG was ground in an agate centrifugal ball mill (Fritsch Co. Ltd., volume, 350 mL), containing agate balls (the diameter and number of balls were: 10 mmΦX10, 15 mmΦX5, 20 mmΦX2) at 4°C at 200 rpm for 2.5 h [27]. Amorphous IMC-A (AA) was prepared as follows: The meta-stable form powder (5 g) of CA was ground in an agate centrifugal ball mill at 4°C at 200 rpm for 2 h [27] 2.2 Powder X-ray diffraction analysis Powder X-ray diffraction (XRD) profiles were recorded with an X-ray diffractometer (RINT 2100 Ultima, Rigaku Co., Japan). The measurement conditions were as follows: target, Cu; filter, Ni; voltage, 40 kV; current, 20 mA; receiving slit, 0.3 mm; scan range, 5◦-45◦ (2θ); step size, 0.02◦; scanning speed, 1◦/min. About 50 mg of the sample powder was carefully loaded into a glass holder, the sample was flattened softly to avoid particle orientation using a spatula and glass plate, and sample weight was accurately measured. 2.3 Thermal analysis Differential scanning calorimetry (DSC) was performed with a DSC 8230 (Rigaku Co., Tokyo). The operating conditions in the open-pan system were as follows: sample weight, 5 mg; heating rate, 10°C/min; N2 gas flow rate, 30 ml/min. All thermodynamic values represent averages of 3 measurements. 2.4 THz spectroscopy After 100 mg of the sample powder was mixed with 900 mg of polyethylene powder in an agate mortar with a pestle, 300 mg of the mixed powder was compressed into a pellet 20 mm in diameter and about 1 mm thick under a pressure of 318 kg/cm2. Attention was paid not to grind the sample at the procedure. The pellet was wedged at an angle of 2 ° to avoid interference fringes in the spectra. THz spectra were recorded using a THz spectrometer [19, 22] with a step of 15 GHz, scanning speed of 5 min/THz, any frequency range of 0.5 to 6.0 THz with two room-temperature operated Deuterated TriGlycine Sulfate (DTGS) pyrodetectors. Each spectrum was normalized using the spectrum of a reference pellet made of 100 wt% pure polyethylene. The THz light source for the spectrometer was a GaP THz signal generator [20–23] based on difference frequency generation (DFG) in a GaP crystal. The generator can generates 1.5 W peak power of THz wave at maximum. Two pump beams for DFG were from Cr: Forsterite lasers pumped by a two channel Q-switched Nd:YAG laser having a repetition rate of 10 Hz. Other details of the GaP THz spectrometer can be found elsewhere [18, 19, 24]. All THz spectra were area normalized by a computer program (Pirouette software Ver. 3.11 Infometrix Co., Woodenville WA, USA).

956

J Infrared Milli Terahz Waves (2012) 33:953–962

3 Results and discussion 3.1 Characterization of poly-amorphous solids by conventional XRD and DSC XRD is the most useful way to obtain qualitative and quantitative information on polymorphic forms for pharmaceuticals. Figure 1 shows the XRD profiles of polymorphic crystalline forms and amorphous solids of IMC. As shown in XRD profiles of polymorphic forms of the α-form (CA) and γ−form (CG) of IMC, the main peaks of pure CA were at 2θ0 7.2, 8.5 and 14.0 °,while those of pure CG were at 2θ010.2, 11.7, 12.7 and 17.0 °, as reported previously [26]. The XRD profiles of CA and CG indicated that XRD could simultaneously distinguish polymorphic crystalline forms. Since the XRD phenomena for crystalline solids follow Bragg’s law [27, 28], crystalline structures can be analyzed based on diffracted X-ray diffractometry, which is affected by sample particle size. Scherrer [28] introduced an equation (Eq. 1) and clarified the relationship between crystallite size and the broadening of diffraction peaks. The equation indicated that a reduction of crystallite size induced a broadening of XRD peaks. Dhkl ¼

Kl b cos θ

ð1Þ

Κ is the Scherrer constant, β is the full width of maximum half intensity, λ is the wavelength of X-rays, Dhkl is crystallite size vertical against to the plate related hkl index in crystalline lattice, and θ is the diffraction angle. X-ray diffraction peaks broaden with a decrease in particle size, and a halo pattern develops for a power of sufficiently small size. Therefore, XRD can not measure nanostructural differences between amorphous solids. Quench and slow cooled amorphous forms (AQ and AS) were prepared by rapid cooling in nitrogen liquid and slow cooling at room temperature [27, 29] of the melt product. Ground amorphous α (AA) and ground amorphous γ (AG) were prepared to grind CA and CG at 4° C [27], respectively. The XRD profiles of their amorphous forms (AS, AQ, AA and AG) had Fig. 1 The X-ray powder diffraction profiles of crystalline polymorphic forms and polyamorphous solids of indomethacin.

J Infrared Milli Terahz Waves (2012) 33:953–962

957

almost the same halo patterns (Fig. 1), and could not be distinguished by the XRD method. Since AQ obtained from the melt product appeared very quickly, a randomized intermolecular structure was formed in the solid, but as AS was cooled slowly, there were some crystalline nuclei in the solid. In contrast, since AG and AA were obtained to grind crystalline solids of CG and CA [27], their particle size decrease during grinding, and eventually all of the diffraction peaks disappeared. If AG and AA have the crystalline structure of starting materials in smaller particles, respectively, those crystalline structures in smaller particles are not able to be detected by an X-ray beam (Fig. 1). The physicochemical properties of the solid-state materials were significantly dependent on the material used to prepare them, meaning that there were differences in impurity content, particle size distribution, absorbed gases on the surfaces etc. The thermal analytical method could detect differences between those materials with a different history. Figure 2 shows the DSC curves of polymorphic crystalline forms and amorphous solids of IMC. The DSC profiles of CA and CG had endothermic peaks due to melting points at 153.8 and 160.8°C, respectively. In contrast, the amorphous forms had peaks attributable to glass transition, crystallization and melt, respectively. The amorphous solid transformed into a super coolant solution, and crystallized into a crystalline solid, and then into a melt product. AA and AG were converted into CG, but AS was converted into a small amount of CA and mostly CG. AQ was converted into almost half of each. All of the thermodynamic parameters of the crystalline and amorphous forms obtained by DSC are summarized in Table 1. The Chemical potential (μ0) of the material solid was linked with the total heat of thermal events

Fig. 2 The differential scanning calorimetry of crystalline polymorphic forms and poly-amorphous solids of indomethacin.

958

J Infrared Milli Terahz Waves (2012) 33:953–962

of the solid, and the heat of total fusion (ΔHft) was evaluated based on glass transition, crystallization and melting as follows: ΔHft ¼ ΔHtg þ ΔHc þ ΔHf ΔHft, heat of total fusion for a solid sample; ΔHtg, heat of glass transition; ΔHc, heat of crystallization; ΔHf, heat of fusion. The ΔHft results of the amorphous solids clearly showed their own physicochemical properties reflecting their chemical potential. CG had the highest ΔHft, since it was the stable crystalline form, and AQ had the lowest value. The order of heat of total fusion for the IMC solids was CG>CA>AG>AA>AS>AQ. The ΔHft of AG was 2.57 times higher than that of AQ, though both forms had the same halo pattern on XRD profiles. The result strongly suggested that amorphous solids were not always the same solid-sate, and there was polyamorphous states reflecting their material history and chemical potential. The ΔHft result was consistent with the crystallization behavior of amorphous solids reported previously [9]. AQ of IMC was stable for more than 9 days at 20°C, and after then transformed into CG, but AG was crystallized after 40 h. The result indicated that the physicochemical stability of amorphous IMC during storage differed significantly between the samples depending on how they were prepared [9]. Sakka [29] proposed that there were two kinds of nanostructures at 5–50 Å for amorphous solids as shown in Fig. 3. One consisted of irregular molecular networks, and the other consisted of nano-level fine crystalline particles. In the present study, AA and AG might be similar to the fine crystalline type, and AQ, to the irregular network type. Thermal analysis can discriminate amorphous solids with different physicochemical properties reflecting their chemical potential, but the method took a long time and, so can not be used as a tool for PAT. 3.2 Characterization of poly-amorphous solids by THz spectroscopy Infrared (IR) spectroscopy is useful for evaluating polymorphic forms, since it can measure intra- and inter-molecular actions (short-range) based on the chemical functional groups of Table 1 Thermodynamic parameters of crystalline polymorphic forms and amorphous solids of indomethacin obtained by DSC. Tg °C ΔHg kJ/ Tc °C mol CG -

-

-

ΔHc kJ/ mol

Tαγ °C ΔHαγ kJ/ Tα °C mol

ΔHα kJ/ Tγ °C mol

-

-

-

-

-

sd CA -

-

-

-

-

-

AG 36.57

0.35

66.67(49.21) 6.78(0.86) -

-

sd

0.14

0.85(0.33)

sd 1.23

1.50

70.23

12.19

127.17

1.29

sd

0.48

0.51

0.30

1.15

0.53

-

-

AS

48.43

1.56

76.83

16.27

sd

2.57

0.39

1.96

0.61

AQ 43.27

1.08

90.30

17.98

sd

0.19

2.29

5.11

1.53

0.15

0.25

-

-

1.87(0.37)

AA 40.93 4.99

153.13 29.08

-

-

-

-

ΔHγ kJ/ ΔHft kJ/ mol mol

160.87 30.86

30.86

0.06

0.70

0.70

-

-

29.08 0.25

159.13 30.39

22.76

0.45

2.57

0.61

159.50 31.08

19.11

0.36

1.45

1.16

151.00 0.96

159.03 26.23

12.49

0.79

0.87

0.44

1.19

1.10

151.90 18.40

158.77 7.36

8.85

0.44

0.78

2.16

1.55

3.10

Tg is glass transition temperature, Tc is crystallization temperature, Tαγ is phase transition temperature from α to form γ, Tα and Tγ are melting points form α and form γ, ΔHg is heat of glass transition, ΔHc is heat of crystallization, ΔHαγ is heat of tansition for from form α to form γ, ΔHα and ΔHγ are heat of fusion for form α and γ, ΔHft is heat of total fusion.

J Infrared Milli Terahz Waves (2012) 33:953–962

959

Fig. 3 Amorphous solid structural models obtained by Grinding and Quench of Melt Product.

pharmaceuticals. There have been several reports [14–16] on the qualification of polymorphic forms by THz spectroscopy, but measurement mechanisms and physicochemical definitions have not been clarified. Therefore, to clarify their scientific background the physicochemical properties of IMC polymorphous solids mentioned above were investigated by THz spectroscopy. Figure 4 shows the THz spectra of the polymorphic crystalline forms and the amorphous solids of IMC obtained by the transmittance method. The spectrum of GC had specific peaks at 2.01, 2.91, 3.93 and 4.33 THz. That of CA had specific peaks at 3.44 and 4.11 THz. CA differed significantly from CG, since they were polymorphic forms as shown by the XRD results. Among, in the ground amorphous solids, AG had peaks at 2.94, 3.94 and 4.34 THz, and AA had a peak at 3.44 THz. Among the quench amorphous solids, AS had peaks at 3.18 and 4.37 THz, and AQ had a broad peak at 3.22 THz. The amorphous solids, AA, AG, AS and AQ, had almost the same halo XRD patterns, but differed significantly in their THz spectra. The THz spectrum of AQ had the flat simplest pattern.

Fig. 4 The terahertz spectra of the powder samples related with α-form and γ-form of indomethacin. Right side is the powder samples related with γ-form; left side is the powder samples related with α-form.

960

J Infrared Milli Terahz Waves (2012) 33:953–962

The amorphous forms of IMC could be separated into two groups, those with peaks at 3.18 and 4.37 THz related with CG (Fig. 4, right side), and those with a peak at 3.44 THz related to CA (Fig. 4, left side). THz spectra of the solids related with CG showed isosbestic points at 3.69 and 4.54, and the spectra had peaks at 3.93 and 4.33 THz the same as CG. The results suggested that AG retained the nano-level-structure of CG after grinding. In recrystallization of AS, a crystalline nucleus with a nano-structure like CG would appear in a melt product of IMC during the slow cooling process. Both results suggested that AG and AS had similar nano-levelcrystalline structure related to CG in very fine particles as shown in Fig. 3. In contrast, the THz spectra of the solids related with CA showed isosbestic points at 3.28 and 4.59 THz. The spectrum of AA had peaks at 3.44 THz with a shoulder at 4.11 THz attributable to CA. The results indicated that AA had transformed into an amorphous solid during grinding, but retained the nano-level-CA-structure after 2 h of grinding. In general, when chemical reaction (one mole of reactant gives one mole of product) involves a pair of substances with an isosbestic point in the spectra, the absorbance of the reaction mixture at this wavelength remains invariant, regardless of the extent of reaction [30], the presence of isosbestic points is an evidence for pure chemical kinetic process of binary components. In the present result, there are 2 way reaction processes from AQ to CG and CA, respectively, since the crystallization pass ways are AQ 0>AG0>CG and AQ0> AA0>CA, meaning that there might be three types of amorphous forms, such as AQ, AA and AG. AG and AA might be nano-crystalline amorphous model (Fig. 3), and AG is packed randomly dimer molecules units in the solid as similar as the crystalline structure of CG [31] and AA is packed randomly molecular clusters related to CA in the solid. In contrast, AQ might be irregular network amorphous model, so AQ is packed randomly monomer molecule in the solid (Fig. 3). All of the THz results for amorphous solids were consistent with thermal behavior in the section above. Since THz has a longer wavelength than IR and IR wave is absorbed on chemical functional groups of a molecule, THz wave absorbed on larger chemical units, such as the vibration of a whole molecule and/or cluster consisting of a few molecules [11] like crystalline nuclei. THz spectroscopy could distinguish amorphous forms of IMC obtained by different methods, but not only crystalline polymorphic forms. THz could measure significant differences in long-rang intermolecular actions among polyamorphous forms.

4 Conclusions The physicochemical stability of the amorphous solids of IMC differed significantly, reflecting their original material. The amorphous solids with nano-order-level structure might be characterized as “poly-amorphous solids”. However, conventional XRD could not detect differences between “poly-amorphous solids” of pharmaceuticals. In contrast, THz could discriminate solids based on measurements of long-range solid-state structures, such as cluster levels. THz is expected to enable a rapid qualitative and quantitative analysis of polymorphous crystalline and “poly-amorphous solids” in the pharmaceutical industry with non-contact and non-destructive measurements.

Declaration of interest This work was supported in part by a Grant from Musashino-Joshi Gakuin and Casio Sciences Promotion Foundation.

J Infrared Milli Terahz Waves (2012) 33:953–962

961

References 1. FDA papers Guidelines: “Manufacturing and Controls for INDs and NDAs”, Pharm. Tech. Japan, 1 835– 850 (1985). 2. J. K. Haleblian, “Characterization of Habits and Crystalline Modification of Solids and Their Pharmaceutical Applications”, J. Pharm. Sci., 64, 1269–1288 (1975). 3. H. Ashizawa, “Thermal Analysis, X-ray Diffraction”, Ed. By H. Ashizawa, Sciences of Polymorphism and Crystallization of Pharmaceuticals, Maruzen Planet Co. Ltd. Tokyo, Japan, pp 58–101 (2002). 4. M. Otsuka, Y. Matsuda, “Polymorphism, Pharmaceutical Aspects”. Swarbrick, J.; Boylan, J. C. eds. Encyclopedia of Pharmaceutical Technology. Vol. 12. New York, Marcel Dekker, 305–326 (1995). 5. J. Bernstein, “Polymorphism in Molecular Crystals”; Clarendon Press: Oxford, Vol. 14 2002. 6. Process Analytical Technology (PAT) Initiative, U.S. Food and Drug Administration Center for Drug Evaluation and Research Home Page (http://www.fda.gov/cder/OPS/PAT.htm) 7. T. Okumura, M. Otsuka, “Application Study of Raman Spectroscopy on Process Analytical Technology in Pharmaceutical Manufacturing process”, PharmTech, Japan, 25, 2435–2439 (2009). 8. M. Otsuka, H. Tanabe, “Quality Control of Pharmaceuticals by Using Chemometrics in Pharmaceutical Manufacturing Process”, Bunseki, 399 (2006). 9. M. Otsuka, N. Kaneniwa, “A Kinetic Study of the Crystallization Process of Noncrystalline Indomethacin under Isothermal Conditions”, Chem. Pharm. Bull., 36, 4026–4032 (1988). 10. G. Buckton, P. Darcy, “The use of gravimetric studies to assess the degree of crystallinity of predominantly crystalline powders”, Inter. J. Pharmaceutics, 123, 265–271 (1995). 11. J. Nishizawa, “Pioneering Work of THz Wave and Its Application for Molecular Sciences, Advance of molecular sciences”, p1-11, Edited by S. Murayama, Tohoku University Press Co. Ltd., Sendai, Japan. 12. P. F. Taday, “Applications of Terahertz Spectroscopy to Pharmaceutical Sciences”, Philos Trans R Soc Lond A;362:351–364 (2004). 13. C. J. Strachan, T. Rades, D. A. Newnham, K. C. Gordon, M. Pepper, P. F. Taday, “Using Terahertz Pulsed Spectroscopy to Study Crystallinity of Pharmaceutical Materials”, Chem. Phys. Lett., 390, 20–24 (2004). 14. P. F. Taday, I. V. Bradley, D. D. Arnone, M. Pepper, “Using Terahertz Pulse Spectroscopy to Study the Crystalline Structure of a Drug: A case Study of the Polymorphs of Ranitidine Hydrochloride”, J Pharmaceut Sci, 92:831–838 (2003). 15. R. M. Woodward, V. P. Wallace, R. J. Pye, B. E. Cole, D. D. Arnone, E. H. Linfield, M. Pepper, “Terahertz Pulse Imaging of ex vivo Basal Cell Carcinoma”, J. Invest. Dermatol., 120, 72–78 (2003). 16. A. Markelz, S. Whitmire, J. Hillebrecht, R. Birge, “THz Time Domain Spectroscopy of Biomolecular Conformational Modes”, Phys. Med. Biol., 47, 3797 (2002). 17. J. Nishizawa, Review, “Extension of Frequencies for Maser to Laser, How the Laser Evolved and was Extended to Terahertz During My Research Life”, A personal review, Proceedings of the Japan Academy Series B, Physical and Biological Sciences, 85, 454–465 (2009). 18. K. Suto, T. Sasaki, T. Tanabe, K. Saito, J. Nishizawa, M. Ito, “GaP THz-Wave Generator and THz Spectrometer using Cr:Forsterite Lasers”, M. Ito, Rev. Sci. Instruments 76, 123109 (2005). 19. J. Nishizawa, K. Suto, T. Sasaki, T. Tanabe, T. Kimura, “Spectral Measurement of Terahertz Vibrations of Biomolecules using a GaP Terahertz-Wave Generator with Automatic Scanning Control”, J. Phys. D: Appl. Phys. 36, 2958–2961 (2003). 20. J. Nishizawa, “Present and Future of the Semiconductor Laser”, Denshi Kagaku, 14, 17–30 (1963). 21. J. Nishizawa, K. Suto, “Semiconductor Raman Laser”, J. Appl. Phys. 51, 2429–2431 (1980). 22. K. Suto, J. Nishizawa, “Low-threshold semiconductor Raman Laser”, IEEE J. Quant. Electron. 19, 1251– 1254 (1983). 23. T. Tanabe, K. Suto, J. Nishizawa, T. Kimura, K. Saito, “Frequency-tunable high-power terahertz wave generation from GaP.”, J. Appl. Phys. 93, 4610 (2003). 24. J. Nishizawa, T. Sasaki, K. Suto, M. Ito, T. Yoshida, T. Tanabe, “High-resolution GaP terahertz spectrometer and its application to detect defects in gamma-irradiated glucose crystal”, Int. J. Infrared Milli. Waves, 2008, 29, 291. 25. M. Otsuka, J. Nishizawa, J. Shibata, M. Ito, “Quantitative Evaluation of Mefenamic Acid Polymorphs by Terahertz-Chemometrics”, J. Pharm. Sci., 2010, 99, 4048. 26. N. Kaneniwa, M. Otsuka, T. Hayashi, ”Physicochemical Characterization of Indomethacin Polymorphs and the Transformation Kinetics in Ethanol”. Chem Pharm Bull 1985, 33, 3447. 27. M. Otsuka, T. Matsumoto, N. Kaneniwa, “Effect of Environmental Temperature on Polymorphic SolidState Transformation of Indomethacin during Grinding”, Chem. Pharm. Bull., 1986, 34, 1784. 28. Hand book of X-ray Diffraction Analysis, Rigaku Co. Ltd., 1998, Tokyo, Japan. 29. S. Sakka, “Noncrystalline Solid, Nature and Properties of Glassy Solid”, Introduction, Amorphous Inorganic Materials, Edt. by R. Makishima, Chemical Reviews, 1983, 41, 2–5.

962

J Infrared Milli Terahz Waves (2012) 33:953–962

30. T. Yamana, “Kinetics of Medical Drugs”, 1979, Nankoudo Press Co. Ltd, Tokyo. 31. T. J. Kistenmacher, R. E. Marsh, “Crystal and Molecular Structure of an Antiinflammatory Agent, Indometacin, 1-(p-Chlorobenzoyl)-5-metoxy-2-methylindole-3-acetic Acid”, J. Am. Chem. Soc., 94(4), 1340–1345 (1972).