European Journal of Pharmaceutical Sciences, 7 (1999) 331338

Physicochemical properties of amorphous clarithromycin obtained by

grinding and spray drying
Etsuo Yonemochi a , Shinichi Kitahara a , Shuei Maeda b , Shigeo Yamamura c , Toshio Oguchi a ,
a,
Keiji Yamamoto *
a
Faculty of Pharmaceutical Sciences, Chiba University, 1 33 Yayoi-cho Inage-ku, Chiba 263 -8522, Japan
b
Pharmaceutics and Pharmaceutical Technology Laboratory, Tsumura and Co., 3586 Yoshiwara, Ami-machi, Inashiki-gun, Ibaraki 300 -1155, Japan
c
School of Pharmaceutical Sciences, Toho University, 2 -2 -1 Miyama, Funabashi, Chiba 274 -8510, Japan
Received 31 March 1998; accepted 7 July 1998

Abstract

In order to characterize the amorphous clarithromycin (CAM) obtained by grinding and spray drying, physicochemical properties
(crystallinity, thermal behavior, stability and solubility parameters) were evaluated. From powder X-ray diffraction, it was estimated that
the crystalline state of CAM was changed into an amorphous state by grinding and spray drying. In differential scanning calorimetry
measurements, both broad and sharp peaks for crystallization were observed in ground samples, whereas spray dried samples showed one
broad peak due to crystallization. As to the stability test under high humidity, structural difference was confirmed between ground CAMs
and spray dried CAM. The heat of dissolution of ground CAMs was greater than that of intact CAM. In the solubility parameter
measurement, the increase of the special term, ds , indicated that the energy change was due to the polarity of the surface energy of the
powder particles by grinding. 1999 Elsevier Science B.V. All rights reserved.

Keywords: Clarithromycin; Amorphous; Crystallinity; Heat of dissolution; Solubility parameter

1. Introduction than the crystalline form (Okonogi et al., 1997; Bolhuis et

al., 1996). For detecting the amorphous state of materials,
One of the important objectives of pharmaceutical infrared spectroscopy, solid nuclear magnetic resonance,
technology is to secure the stability and effectiveness of extended X-ray absorption fine structure and differential
pharmaceutical products. Crystallinity is known, as one of scanning calorimetry (DSC) are used (York, 1994). Pow-
the most important physicochemical properties of drug der X-ray diffractometry is also useful to investigate
substances, to affect physicochemical stability, solubility crystalline and amorphous characteristics. The pattern of
and absorption of drug substances as well as compression powder X-ray diffraction in the amorphous state is typical-
properties in the manufacturing process. The amorphous ly a halo pattern due to its disordered structure. Yamaguchi
state is defined as the complement of the crystalline state, et al. reported that amorphous forms of antibiotic com-
i.e., it is characterized by a disordered molecular or atomic pounds of similar powder X-ray diffraction pattern, how-
arrangement. It is known that the amorphous form of a ever, exhibited different thermal properties and dissolution
drug substance is produced by the grinding of a single behavior depending on preparation methods of the amor-
component, grinding it with some excipients, mixing it phous materials (Yamaguchi et al., 1993b).
with porous substances, freeze drying, compression, de- Differences of amorphous states in the microscopic
sorption of water of crystallization or spray drying (Han- structure could be determined by measuring the heat of
cock and Zografi, 1997; Yamaguchi et al., 1993a). The solution with a micro-calorimeter as the heat of solution
dissolution behavior and the bioavailability of water reflects the molecular arrangement within the particles
insoluble drugs could be improved by producing them in (Yamaguchi et al., 1993a). Yonemochi et al. determined the
amorphous form, as the amorphous form is more soluble heat of solution of ursodeoxycholic acid in varied crys-
tallinity samples, and they compared each feature in
*Corresponding author. Tel.: 181 43 2902937; fax: 181 43 2902939; detecting amorphous characteristics among X-ray, DSC
e-mail: yamamotk@p.chiba-u.ac.jp and isothermal analysis (Yonemochi et al., 1997).

0928-0987 / 99 / $ see front matter 1999 Elsevier Science B.V. All rights reserved.
PII: S0928-0987( 98 )00040-2
332 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338

Inverse gas chromatography is one of the techniques 2.2.3. Measurements of powder X-ray diffraction and
used to measure the solubility parameters of bulk drug determination of crystallinity
substances and inactive ingredients (Hildebrand and Scott, Powder X-ray diffraction patterns were measured with
1962). Solubility parameters have been widely applied to Rigaku X-ray powder diffractometer (Model 2027) with
evaluate the compression characteristics of various pow- CuKa radiation at 5358 (2u ). Crystallinity of CAMs was
ders (Nakai et al., 1990), to compare the surface charac- determined by Rulands method using a Rigaku X-ray
teristics of different batches of bulk drug (York, 1994), to powder diffracometer with rotating anode generator (Rint
predict interactions and determine the solubility of solid 2500) with CuKa radiation at 3428 (2u ) (Alexander,
mixtures (Parker and Rowe, 1991) and to investigate 1973).
changes in the physicochemical properties of solid drug
substances in different crystallinities (Egawa et al., 1992). 2.2.4. Differential scanning calorimetry ( DSC)
Maeda et al. evaluated solubility parameters to investigate DSC measurements were performed on a MAC Science
the crystal polymorphs and physicochemical properties of DSC 3100 at a heating rate of 58C / min from 30 to 2508C
lactose commonly used as inactive ingredients (Maeda et using an aluminium sealed pan and a du Pont Instruments
al., 1995). Solubility parameters were used to determine TA9900 at a rate of 58C / min from 30 to 2508C using an
the degree of crystallization based on surface energy aluminium open pan. The sample sizes were 2 to 3 mg for
before and after glass transition of polymers (Orts et al., each measurement.
1992).
In this study, we used a 14-membered ring macrolide 2.3. Stability test
antibiotics (clarithromycin:CAM), which is barely soluble
in water. The physicochemical properties of the CAMs Stability test was carried out for estimating the crys-
prepared by grinding and spray drying were evaluated by tallization behavior of amorphous CAMs using a desic-
powder X-ray diffraction, DSC, heat of dissolution and cator controlled at 408C and 82%RH (relative humidity)
stability test. The solubility parameters were also evaluated for appropriate periods. A temperature of 408C was
by using inverse gas chromatography to investigate maintained using an air bath. A constant RH of 82% was
changes in the surface energies. generated using a saturated aqueous solution of KCl.

2.3.1. Determination of heat of dissolution

Heat of dissolution was determined with a multipurpose
2. Materials and methods
calorimeter (Model MPC-11, Tokyo Riko). Dibasic sodium
phosphatecitrate buffer (pH 6.00 at 258C), listed in the JP
2.1. Materials
XII, was used as a solvent. It may be necessary in some
cases to take into account the quantity of heat generated
Clarithromycin (Taisho Pharmaceutical Co. Ltd. Japan)
when protons dissociate in the buffer. The quantity of heat
was supplied and used. Organic solvents used as molecular
due to dissociation was neglected for the reason that the
probes for the determination of solubility parameters were
dissociation enthalpy of protons was very small compared
n-pentane, n-hexane, n-heptane, n-octane, n-nonane, n-
to the heat of dissolution of the sample dissolving in the
decane, dioxan, toluene, tetrahydrofuran, methyl alcohol,
buffer. Measurement conditions were as follows: tempera-
ethyl alcohol and 1-butyl alcohol. These solvents were of
ture 258C, temperature precision 61310 25 8C, rotational
special reagent grade and the tetrahydrofuran was of HPLC
speed of stirrer 50 rpm, and output range 1 mV. A 3 ml
grade.
glass ampoule was filled with 15 mg of the sample and
then sealed. Calibration was carried out at known quan-
2.2. Methods tities of heat (1000, 2000, and 3000 mJ) for each de-
termination.
2.2.1. Preparation of ground sample of CAM
2.3.2. Determination of solubility parameters
3.0 g of CAM was ground by a vibrational mill (TI-200,
Solubility parameters were determined by using a gas
Heiko Seisakusho, Ltd.) with alumina cell for 1 min to 30
chromatograph (Shimadzu GC-12A with a flame ionization
min.
detector) according to the proposed method by Kesselring
et al. (Phuoc et al., 1986). The adsorption energies of each
2.2.2. Preparation of spray dried sample of CAM CAM sample tested by different molecular probes were
80 g of CAM was dissolved into 20 kg methyl alcohol used to evaluate the solubility parameters. In this study,
and then spray dried with a spray dryer (CL-12, Okawara multiple linear regression analysis was carried out to
Kakoki, Ltd.) under the following conditions: Rotational calculate two-dimensional solubility parameters. Chro-
speed of atomizer, 15 000 rpm; jet velocity, 6 kg / h; tube matographic conditions were as follows; the carrier gas
inlet temperature, 1008C and tube outlet temperature, 438C. was dry nitrogen gas, the gas flow rate was maintained
 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338 333

between 5 and 10 ml / min, and injection port temperature was found that the intensities of X-ray diffraction peaks
was 1808C. Glass columns were 3 mm i.d., and 60 cm in decreased with increased grinding time and the powder
length. No crystallization was observed for the ground X-ray diffraction patterns of CAM came to be almost halo
CAMs even after storage at 708C for 3 days. From this patterns after 30 min grinding. When the degree of
confirmation, the conditioning was carried out at 708C for crystallinity and the degree of disorder parameter were
24 h. The column temperatures were fixed at 48, 55, 63 calculated by Rulands method, intact CAM and 1, 3, 5, 15
and 708C. A sample after measurement showed no crys- and 30 min ground CAM had the following values
tallization estimated by DSC, whereas in a spray dried (designated values in parentheses were disorder parame-
sample, crystallization occurred after storage at 708C for 1 ters): 79.2%(3.1 A 2 ), 66.6%(4.6 A 2 ), 55.0%(4.5 A 2 ),
2 2 2
day. For this reason, solubility parameter measurements 33.7%(3.2 A ), 25.4%(3.9 A ) and 22.9%(3.7 A ), respec-
were carried out for each intact and ground CAM. tively. From the calculated degree of crystallinity, it was
estimated that increased grinding time changed the crys-
talline state of CAM into an amorphous state. However,
3. Results and discussion distinct differences in disorder parameter were not esti-
mated for each ground sample.
3.1. Reduction in crystallinity by grinding and spray In Fig. 2, the powder X-ray diffraction patterns of spray
drying dried CAM are shown with results of intact and 30 min
ground CAM. The powder X-ray diffraction pattern of
The powder X-ray diffraction patterns of samples ob- spray dried CAM was shown to be almost a halo pattern. It
tained by grinding for 1 to 30 min are shown in Fig. 1. It was suggested that spray dried CAM changed to an
amorphous state in the same manner as 30 min ground
CAM. The degree of crystallinity and the degree of
disorder parameter for spray dried CAM were calculated as
12.9% and 2.9 A 2 , respectively. The degree of disorder
parameter of ground CAM was higher than that of the
spray dried CAM. This result suggested that the crystalline
part of ground CAM could have imperfections or defects
in the crystalline lattice.

Fig. 1. Changes in the powder X-ray diffraction pattern of CAM by

grinding. Fig. 2. Powder X-ray diffraction pattern of CAM.
334 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338

3.2. Thermal properties of ground CAM but a sharp exothermic peak was noted only at about
1108C. After this sample was allowed to cool down to
The DSC curves of ground CAM obtained after various room temperature, X-ray diffraction measurement was
grinding times are shown in Fig. 3. A broad exothermic carried out again. Shown as (C) in Fig. 4, the X-ray
peak appeared in the temperature range of 751208C in the diffraction pattern showed higher crystallinity than the
1 min ground sample. In addition to the broad exothermic sample heated up to 1008C. Moreover, the background
peak, a sharp exothermic peak appeared in the range of level of the X-ray diffraction pattern in the 2nd run was
1101208C in the sample ground for 3 min or more. lower than in the 1st run. In the 3rd run, the 2nd run
Further study was conducted to evaluate these two peaks sample was heated up to 2508C. The exothermic peaks
and the results of the evaluation are shown in Fig. 4. A 30 which had appeared in the 1st and 2nd runs were not
min ground sample was heated up to 1008C once and observed, but an endothermic peak due to fusion was
allowed to stand until the temperature of the sample observed at 2268C. From these results, the broad ex-
returned to room temperature (1st run). The powder X-ray othermic peaks occurring in the range from 75 to 1208C
diffraction pattern of this sample showed a crystalline and the sharp exothermic peak occurring in the range from
diffraction pattern noted as (B) in Fig. 4. The 30 min 110 to 1208C were considered to be due to an irreversible
ground sample was cooled to 708C after being heated to crystallization. The heat of crystallization of each CAM
1008C and then heated again to 1508C (2nd run). In the calculated from the DSC peak area for 1, 3, 5, 15 and 30
2nd run, no exothermic peak was observed around 758C, min ground CAM was 5.66, 12.4, 15.6, 15.9, and 16.7
kJ / mol, respectively. It was suggested that the higher
energy states produced by increased grinding time changed
the crystalline structure of CAM into an amorphous state,
since as the grinding time increased the exothermic peak
due to crystallization was observed to become clearer and
the heat of crystallization also became greater.
The DSC curves of spray dried CAM are shown in Fig.
5. Thermal properties of spray dried CAM were different
from 30 min ground CAM and showed only a broad
exothermic peak occurring in the range from 808C to
1208C. Having first been heated at a 58C / min heating rate,
the powder X-ray diffraction pattern of spray dried CAM
after cooling from 1408C to room temperature appeared
similar to the X-ray diffraction pattern of (C) in Fig. 4.
From these results, the broad exothermic peak in the range
from 80 to 1208C was considered to indicate crystalliza-
tion. The heat of crystallization calculated from the DSC
peak area was 17.6 kJ / mol.

3.3. Estimation of physicochemical stability and

crystallization behavior of amorphous samples under
storage in conditions of high humidity

A storage test under conditions of 408C and 82% RH

was carried out to estimate the physicochemical stability
and the amorphous state of ground CAMs and one spray
dried CAM sample. Powder X-ray diffraction patterns of 5,
15, and 30 min ground samples and a spray dried sample
after 7 days storage are shown in Fig. 6. Crystallization
was found to occur in each sample from the results in
which crystallization peaks were revealed in X-ray diffrac-
tion patterns of each ground sample and spray dried
sample.
In order to estimate the thermal properties of these
samples in amorphous states, the relationships between
heat of crystallization calculated from DSC measurements
Fig. 3. Changes in the DSC curves of CAM by grinding (Instrument: and storage time are shown in Fig. 7. While the rate of
TA9900). crystallization decreased in each ground sample during 3
 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338 335

Fig. 4. Change in the powder X-ray diffraction pattern of CAM by heat treatment (Instrument: DSC MAC 3100). (A), before heating; (B), after 1st run;
(C), after 2nd run.

days storage, complete crystallization did not occur even heat of dissolution with a twintype conduction calorimeter.
after 7 days storage. On the other hand, in the spray dried Dibasic sodium phosphatecitrate buffer (pH 6.00 at 258C)
sample, complete crystallization occurred without any was used as a dissolution medium. The heat of dissolution
change in crystallization rate. It was found that there for each of intact CAM, 5 min ground CAM, 30 min
should be a difference of structure between ground CAM ground CAM and spray dried CAM for buffer solution are
and spray dried CAM. shown in Table 1. The dissolution of these samples was
Fig. 8 shows the effect of storage time on DSC patterns found to be an exothermic process and the heat of
of 5, 15 and 30min ground CAM. It was noted that the dissolution for each was 72.8, 88.8, 77.0 and 87.6 kJ / mol,
broad exothermic peak occurring in the range from 75 to respectively. While the results suggest that the energy
1208C for each sample disappeared with increased storage states of ground and spray dried CAM were higher than in
time. On the other hand, there was no change in the sharp intact CAM, microscopic structural differences resulting
exothermic peak in the range from 110 to 1208C. From from different preparation methods were not found be-
these results, it was estimated that the amorphous state tween ground and spray dried CAMs.
which, relatively speaking, tended toward crystallization
formed in the early grinding state, and the other amorphous 3.5. Solubility parameters of CAM samples
state which tended to resist crystallization was formed after
increased grinding time. In contrast, without any other Solubility parameters of ground CAM were estimated to
force being applied to the samples, the amorphous state evaluate changes in the surface energy state. Egawa et al.,
tending toward crystallization formed in spray dried CAM measured the solubility parameters of cephalexin with a
due to its faster rate of drying, but this same amorphous low degree of crystallinity which was prepared for various
state did not form in CAM dried under ordinary con- grinding times and reported that solubility parameters
ditions. varied with changes in the degree of crystallinity (Egawa
et al., 1992).
3.4. Determination of heat of dissolution The solubility parameters of intact CAM and 30 min
ground CAM are shown in Table 2. The dispersion terms
Energy states of CAM samples prepared by grinding and dd were 7.6 and 7.9 (cal / cm 3 )1 / 2 for intact and 30 min
spray drying methods were examined by measuring the ground CAM, respectively, showing less change by grind-
336 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338

Fig. 7. Changes in heat of crystallization of CAM by storage at 408C

under 82% RH Key: h, ground for 5 min; , ground for 15 min; s,
ground for 30 min; m, spray dried sample.

ing. On the other hand, the special term ds was 6.8 and
13.6 (cal / cm 3 )1 / 2 for each of intact and 30 min ground
CAM, respectively. The special term of 30 min ground
CAM was doubled against that of the intact CAM. The
dispersion term is attributed to van der Waals force and
the special term is mainly attributed to the hydrogen bond
Fig. 5. DSC cruves of CAM (Instrument: TA 9900). a), intact; b), ground
and other interactions due to polarity. Accordingly, it was
for 30 min; c), spray dried sample.
suggested that a change in the special term related to an
increase or decrease in polarity on the surface of the

Fig. 6. Changes in powder X-ray diffraction patterns of CAM by storage at 408C under 82% RH.
 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338 337

Fig. 8. Changes in DSC curves of ground CAM by storage at 408C under 82% RH (Instrumetn: TA 9900).

Table 1 intact CAM in the estimation of heat of dissolution and

Heat of dissolution of CAM (at 258C) solubility parameters.
Sample Heat of dissolution (kJ / mol) mean6S.D.
Intact 272.866.5 (n55)
5 min Grinding 288.8612.5 (n56) Acknowledgements
30 min Grinding 277.0612.5 (n57)
Spray Drying 287.668.2 (n56)
The present work is part of a research project
Solvent: disodium hydrogen phosphatecitric acid buffer solution, pH 6.0 (09672184) carried out with a Grant-in-Aid for Scientistic
(JP XII).
Resarch (C) from the Ministry of Education, Science,
Sports and Culture, and Uehara Memorial Foundation. We
are grateful to Taisho Pharmaceutical Co., Ltd, for the kind
powder particles. The increase of the special term indicated gift of CAM.
the energy change due to the polarity of the surface energy
affected by the grinding method.
References

Alexander, L.E., 1973. X-Ray Diffraction Methods in Ploymer Sciences.

4. Conclusion John Wiley, New York.
Bolhuis, G.K., Zuurman, K., te Wierik, G.H.P., 1996. Improvement of
The amorphous states of CAM prepared by grinding and dissolution of poorly soluble drugs by solid deposiotion on a super
spray drying methods were different in physicochemical disintegrant. II. The choice of super disintegrants and effect of
granulation. Eur. J. Pharm. Sci. 5, 6369.
properties. Especially, it was found that in the early stages
Egawa, H., Maeda, S., Yonemochi, E., Oguchi, T., Yamamoto, K., Nakai,
of grinding CAM, the amorphous state tending towards Y., 1992. Solubility parameter and dissolution behavior of cefalexin
crystallization was formed and the other amorphous state powders with different crystallinity. Chem. Pharm. Bull. 40, 819820.
tending to resist crystallization was formed with increased Hancock, B.C., Zografi, G., 1997. Characteristics and significance of the
grinding time. In the spray dried CAM, only the amor- amorphous state in pharmaceutical systems. J. Pharm. Sci. 86, 112.
Hildebrand, J.H., Scott, R.L., 1962. Regular Solution. Prentice-Hall
phous state tending toward crystallization was formed. The
Englewood Cliffs, New Jersey.
energy states of amorphous forms prepared by both Maeda, S., Nakai, Y., Yonemochi, E., Oguchi, T., Yamamoto, K., 1995.
grinding and spray drying methods were different from the Determination of partial solubility parameters of lactoses and its
Table 2
application to solubility evaluation. Yakugaku Zasshi. 115, 234240.
Solubility parameters of CAM
Nakai, Y., Yamamoto, K., Oguchi, T., Yonemochi, E., Maeda, S., Egawa,
Sample dd (cal / cm 3 )1 / 2 ds (cal / cm 3 )1 / 2 dt (cal / cm 3 )1 / 2 H., 1990. Determination of solubility parameters for medicinals and
excipients. Yakugaku Zasshi. 110, 3439.
Intact 7.6 6.8 10.2
Okonogi, S., Yonemochi, E., Oguchi, T., Puttipipatkhachorn, S., Yama-
30 min Grinding 7.9 13.6 15.9
moto, K., 1997. Enhanced dissolution of ursodeoxycholic acid from
d, dispersion term; s, special term; t, total term. the solid dispersion. Drug Dev. Ind. Pharm. 23, 11151121.
338 E. Yonemochi et al. / European Journal of Pharmaceutical Sciences 7 (1999) 331 338

Orts, W.J., Romansky, M., Guillet, J.E., 1992. Measurement of the Yamaguchi, T., Nishimura, M., Okamoto, R., Takeuchi, T., Yamamoto,
crystallinity of poly(b-hydroxybutyrate-co-b-hydroxyvalerat) copoly- K., 1993. Glass formation of 49-O-(4-methoxyphenyl)acetyltylosin and
mers by inverse gas chromatography. Macromolecule 25, 949953. physicochemical stability of the amorphous solid. Int. J. Pharm. 85,
Parker, M.D., Rowe, R.C., 1991. Source variation in the wet massing 8796.
(granulation) of some microcrystalline celluloses. Powder Technol. 65, Yonemochi, E., Ueno, Y., Ohmae, T., Oguchi, T., Nakajima, S., Yama-
273281. moto, K., 1997. Evaluation of amorphous ursodeoxycholic acid by
Phuoc, N.H., Luu, R.P.T., Munafo, A., Ruelle, P., Tran, H., Buchmann, thermal methods. Pharm. Res. 14, 798803.
M., Kesselring, V.W., 1986. Determination of partial solubility parame- York, P., 1994. Raw materials: new solutions to old problems. Pharm.
ters of lactose by gas-solid chromatography. J. Pharm. Sci. 75, 6872. Tech. Eur. 6, 1723.
Yamaguchi, T., Nishimura, M., Okamoto, R., Takeuchi, T., Yamamoto,
K., 1993. Physicochemical stability of glassy 16-membered macrolide
compounds. Chem. Pharm. Bull. 41, 18121816.