STUDY OF A MICROWAVE ASSISTED VACUUM DRYING PROCESS APPLIED TO THE GRANULATED PHARMACEUTICAL DRUG HYDROCHLORTHIAZIDE (HCT)

Michele N. Berteli1, Antonio Marsaioli, Jr.1 and Elisabeth Rodier2

1

Universidade Estadual de Campinas/Faculdade Engenharia de Alimentos/Departamento Engenharia de Alimentos P.O. Box 6121, Campinas, SP, BRAZIL berteli@gmail.com; tonymars@fea.unicamp.br 2 cole de Mines dAlbi Carmaux UMR CNRS 2392, Albi, FRANCE rodier@enstimac.fr

Drying processes have evolved considerably over the years, aiming at the best conditions to shorten the processing time, but maintaining the highest final product quality. The application of vacuums to the microwave drying process might be interesting, especially for thermally sensitive products because the reduced boiling point of the solvent allows a lower temperature processing. The objective of the present work is to study and evaluate the process of drying a granulated product that is the basis of the drug hydrochlorthiazide, with an initial moisture content of 21% in dry basis. This study monitored the drying kinetics, product temperature and power absorbed by the sample using a bench scale vacuum microwave dryer. The equipment consisted of a cylindrical pressure vessel crossed by a wave guide, setting up a system whereby the vacuum pressure, the sample weight and the incident, reflected and residual microwave powers could be measured and evaluated, throughout the entire process. The experimental runs were established with an approximate incident microwave power at 20 W for absolute pressure levels of 50 and 75 mbar, working with samples of about 1.4 g. It was observed that the vacuum microwave process kinetics at both pressure levels showed little difference. The processes were carried out almost entirely in a regime of water evaporation, the products temperature remaining below the solvent boiling temperature. The drying times were similar for both processes, whereas the absorbed power was slightly higher at the pressure of 75 mbar. Submission Date: 20 July 2006 Acceptance Date: 14 April 2007 Publication Date: 15 June 2007 INTRODUCTION The use of vacuums in the microwave drying process could be of interest, especially with thermo-labile products such as food, pharmaceutical powders and granules. Under the vacuum
Keywords: Drying, microwave, vacuum, granule, drugs
International Microwave Power Institute

the solvent boiling temperature lowers, thus permitting evaporation at lower temperatures and directly influencing the final product quality [Pr, 2002]. Drouzas [1999] reported a study of a combined process of drying by microwaves under vacuum, starting from a model fruit gel, aiming to dry from 38% to 3% in dry basis (d.b.).
241

The drying operation took just 4 minutes. In comparison, similar gel samples dried by hot air at 60o C and atmospheric pressure took 8 hours to reduce the product moisture content to 10% d.b. The same product, when compared to that obtained by microwave assisted drying at atmospheric pressure, exhibited a darker color. In another example of microwave drying under a vacuum, Cuis experiment with drying carrots [2004] showed the influence of the applied power levels with respect to the product drying kinetics curves. On the other hand, the vacuum pressures variation showed little influence on the kinetics. The objective of the present work is to study the drying kinetics of a pharmaceutical granule, the basis of the drug hydrochlorthiazide (HCT), from the initial moisture content ranging from 21 to 25% (d.b.) to the final moisture content of 3% (d.b.) as well as determining the power absorbed by the product, using a microwave assisted vacuum dryer operating at two absolute pressure conditions: 50 and 75 mbar. EXPERIMENTAL PROCEDURES The raw material employed is the compositional basis of a pharmaceutical drug called hydrochlorthiazide (HCT) with an initial moisture content of about 21% d.b. For the present study, the formula for the HCT is based on 30% of the effective drug principle hydrochlorthiazide, excipients being 28% mannitol, 42% cornstarch, with de-ionized water as the moistening agent [Calligaris, 2001]. Characterization of the constituent powders Calorimetric measurements were made in order to obtain information about the eventual structural evolution of the granule constituents during the drying process. The measurements were taken using Perkin Elmer DSC-7 equipment. Measurements of the three separated constituents of the granules showed
242

that no modifications were observed in the temperature range from 20 to 80o C, moreover the granules temperature can not exceed 70o C in order to preserve the effective drug principle, hydrochlorthiazide [Calligaris, 2001]. Thus, a similar temperature ramp was adopted for the tests. The sorption isotherm curve was collected using a Dynamic Vapor System (DVS) SMS, UK type [Arlabosse, 2003]. Laboratory scale microwave vacuum dryer The experiments were carried out using a piece of equipment composed of a cylindrical vacuum chamber of approximately 100 liter volumetric capacity (Figure 1), crossed with a horizontal single mode microwave guide [Pr, 2002]. The waveguide operated at atmospheric pressure and the vacuum was realized within a vertical quartz tube, where the pressure was reduced. Quartz was used because it would not absorb microwaves, which were generated by a magnetron (at one end of the waveguide) at a frequency of 2450 MHz. Four stubs and two circulators were located between the magnetron and the waveguide so as to protect it from reflected power and to control the output power reaching the product sample. Thus, since the magnetron provided 800 W of a fixed power, the first stub was used to compel part of the power back to the first circulator, whereas the other three stubs were used to finely match the remaining power (measured in a first milliwattmeter) to the load, the second circulator absorbing any other reflected power (measured in a second milliwattmeter). A data acquisition system directly provided the incident power to the load from the difference between readings of the first and the second milliwattmeters. Such a system was capable of varying the incident power to the load within the range from 10 to 100 W [Pr, 1999]. In this study the incident power was adjusted at a fixed 20 W value. A third milliwattmeter was used to measure the residual
Vol. 40, No. 4, 2007

Journal of Microwave Power & Electromagnetic Energy

Figure 1. Laboratory scale microwave vacuum dryer. power absorbed at the end of the waveguide by a water load, enabling the determination of the power absorbed by the sample by taking the difference between incident and residual powers. A PTFE sample holder was placed in the middle of the quartz tube, standing on the plate of a weighing instrument, whose electronic part was placed outside the vacuum chamber. The study of the microwave power absorbed by the HCT granule and its drying kinetics was developed at two absolute pressure levels, 50 mbar and 75 mbar, and using a sample weight of 1.4 g at an average moisture content of 21% d.b. Five repetitions were made for the two absolute pressure levels. To monitor temperature evolution in the sample, a Testo 925 type thermocouple was inserted in the sample after every microwave vacuum drying cycle by turning off the system. The readings were taken at six different cycle times ( 3, 6, 9, 12, 15 and 20 minutes), in triplicate for each cycle, for every absolute operational pressure. The final product moisture was determined
International Microwave Power Institute

using an infrared Mettler LJ16 drying balance at the end of each experiment. RESULTS Drying kinetics Figure 2 shows the sorption curve of the studied granules. The analyses were carried out at a temperature of 25o C. It can be observed that the material has practically no bound water at equilibrium, under conditions of high relative humidity with low water contents.This could be assigned to the interactions among the constituent particles (starch, mannitol and hydrochlorthiazide), responsible for their agglomeration, and bringing about the granules ability to release water into the environment. Sample weight was recorded every 5 seconds by a data acquisition system. The graphical representations were plotted using the average moisture value of five tests per each pressure, in order to check the influence of the pressure
243

0.030

Moisture content, X (kg water / kg dry matter)

0.025

0.020

0.015

0.010

Adsorption Desorption
0.005

0.000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Water activity, aw

Figure 2. Isotherms of adsorption and desorption of the HCT granule at 25C.

Figure 2. Isotherms of adsorption and desorption of the HCT granules at 25 C. ture as a function of product moisture content during the microwave assisted vacuum drying at pressures of 50 and 75 mbar. The results were obtained using a thermocouple, taking readings at six points in time, and at each temperature reading a new experiment was started. The graph presents the average of all the tests, accomplished in triplicate for every time interval, at the two working pressures. The boiling water temperatures are also shown, at pressures 50 and 75 mbar, respectively 32.9 and 40.2o C. Absorbed power The residual power was monitored under two conditions: with the sample holder full and empty. When operating with an empty sample holder, it was observed that the residual power was stable at a value of about 20 W, equivalent to the incident power. When operating with the sample holder filled with material, the value for residual power varied with time as a function of the power absorbed by the material. Considering
Vol. 40, No. 4, 2007

on the drying kinetics of the granulated HCT (Figure 3). Figure 4 shows the drying curve obtained with a dimensionless moisture content (X - Xeq) / (X0 - Xeq) as function of time (s). The moisture content X is defined as the ratio between the mass of water in the product and the mass of dried solids; Xeq being the equilibrium moisture under the air conditions used and X0, the initial product moisture. The equilibrium moistures applied to these calculations were equivalent to the final moisture values reached in the experiments at the pressures 50 and 75 mbar, being respectively 1.6 and 2.0% d.b. Figure 5 shows the curves obtained with the rate of evaporation density Rdrying (g / m2 s), defined as the mass of water evaporated per unit of product surface area in contact with the support, per unit time, as a function of the dimensionless moisture ratio (X - Xeq) / (X0 - Xeq). Temperature Figure 6 shows the evolution of the tempera244

Journal of Microwave Power & Electromagnetic Energy

0.30

Microwave-Vacuum - 50mbar Moisture content, X (g water / g dry matter)

0.25

Microwave-Vacuum - 75mbar

0.20

0.15

0.10

0.05

0.00
0 200 400 600 800 1000 1200 1400

time (s)

Figure 3. Product moistures as functions of timeas at functions 50 and 75 of mbar. Figure 3. Product moistures time at 50 and 75 mbar.

the difference in the values for residual power between the full and empty sample holder, the power absorbed by the material could be calculated. The errors observed from the power measurements could be estimated as a maximum relative deviation at 20%, meaning, for instance, that when measuring 20 W the deviation would be within the range of 4 W, whereas for a 2 W measurement the deviation could reach 0.4 W. DISCUSSION Drying kinetics It can be observed in Figure 3 that most of the moisture was eliminated at the beginning of the drying process. When drying at 50 mbar, the beginning of the drying was considered capable of extending up to a mean time of 400 s (7 minutes). During this period the moisture content of the granule decreased from an initial value in
International Microwave Power Institute

between 21 to 25% d.b. to an average value of 4% d.b. In the following 800 s (13 minutes) the granule lost a further 3% of moisture. Thus, at the end of the total 1200 s treatment time the product presented a moisture content of 1.6% d.b. Referring to the sorption isotherm, it can be observed that with a moisture content above 3% d.b., the product possessed free or unbound water, thus explaining the greater facility of drying up to this moisture value. Still referring to Figure 3, it can be observed that at 75 mbar the product reached a moisture content of about 4% d.b. during the first 500 s (8 min), and at the end of 1200 s the product reached a moisture content of 2% d.b. Even though the real measured moisture values remained within the range 21 to 25% d.b., a small deviation in the higher range to 27% d.b. in Figure 3 can be observed. This was caused by an accumulation of errors from calculations obtained by using a spread-sheet software. According to Figure 4, which shows the drying curve obtained with a dimensionless
245

1.0

Microwave-Vacuum - 50mbar
0.9

Microwave-Vacuum - 75mbar
0.8 0.7

(X -Xeq ) / (X0 -Xeq )

0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 200 400 600 800 1000 1200 1400

time (s)

Figure 4. Dimensionless moisture contents as functions of time at 50 and 75 mbar.

Figure 4. Dimensionless moisture contents as functions of time at 50 and 75 mbar.

4.0 3.5

Rate, Rdrying (g water / m2 s)

3.0 2.5 2.0 1.5 1.0 Microwave-Vacuum - 75mbar 0.5 Microwave-Vacuum - 50mbar 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

(X-X eq ) / (X o -X eq )

Figure 5. Drying rates/unit area as a function of the dimensionless moisture content at 50 and 75 mbar.

Figure 5. Drying rates/unit area as a function of the dimensionless moisture content at 50 and 75 mbar.
Journal of Microwave Power & Electromagnetic Energy Vol. 40, No. 4, 2007

246

50 45 40 35

Temperature (C)

30 25 20 15 10 5 0
0 1 2 3 4 5 6 7 8 9 10 Microwave-Vacuum - 75mbar Microwave-Vacuum - 50mbar Temperature evap. water - 50mbar Temperature evap. water - 75mbar

Moisture content (g water/ g dry matter)

Figure 6. Product temperature as a function offunction moisture content at 50 and 75 mbar Figure 6. Product temperature as a of moisture content at

50 and 75 mbar.

moisture ratio on the ordinate axis, it can be observed that the evolution of water loss as a function of time was the same at both absolute pressures. It can be observed in Figure 5 that the curves obtained for the rate of evaporation density were decreasingly linear, showing basically the same evolution at both pressures. Temperature It can be observed in Figure 6 that at a pressure of 75 mbar the product temperature remained lower than the boiling water temperature up to when a moisture content of 3% was reached. The same occurred during the experiments carried out at 50 mbar, but when the product moisture content was almost 3% d.b. the product temperature rose above the boiling water temperature. This indicates that other constituents of the granule started absorbing microwaves, thus leading to an increase in the product temperature. As a conclusion, it can be stated that the

drying procedures should be carried out in a regimen of water evaporation, with the product temperature lower than the boiling water temperature. Absorbed power It is evident from the curves shown in Figure 7 that the sample absorbed more power at the higher absolute pressure (75 mbar). The greater pressure difference at 50 mbar probably caused an increased mass diffusion, requiring less microwave energy for vaporization. On the other hand, the smaller pressure difference at 75 mbar accompanied by a higher absorption of microwave energy, resulted in similar drying rates for the two pressures studied. Figure 8 indicates the ratio of the absorbed power to the available water in the product, as a function of the real moisture values. According to the plot, it can be said that the absorbed power per unit of available water in the product was constant for the free water, i.e. for moisture contents above 3%. When little water remained

International Microwave Power Institute

247

9 8 7

Absorbed Power (W)

6 5 4 3

Microwave-Vacuum - 50mbar
2

Microwave-Vacuum - 75mbar
1 0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

(X -X eq ) / (X 0 -X eq )

Figure 7. Power absorbed by the product moisture as a function of the dimensionless moisture content at 50 and 75 mbar.

Figure 7. Power absorbed by the product moisture as a function of the dimensionless moisture content at 50 and 75 mbar.

160

Microwave-Vacuum - 50mbar
140

Microwave-Vacuum - 75mbar

Absorbed Power (W) / g water

120

100

80

60

40

20

0 0.00

0.05

0.10

0.15

0.20

0.25

0.30

Moisture content (g water / g dry matter)

Figure 8. Ratio of power absorbed by the product water as a function of moisture content at 50 and 75 mbar

Figure 8. Ratio of power absorbed by the product water as a function of moisture content at 50 and 75 mbar.
Journal of Microwave Power & Electromagnetic Energy

248

Vol. 40, No. 4, 2007

160

Microwave-Vacuum - 50mbar
140

Microwave-Vacuum - 75mbar

Absorbed Power (W) / g water

120

100

80

60

40

20

0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

(X-X eq ) / (X 0 -X eq )

within the product, thus being more strongly bound, the absorbed power per unit of available water in the product increased drastically, also being absorbed by other constituents of the granule. Figure 9 represents the graph plotted with the dimensionless values for moisture, indicating the same process. CONCLUSIONS

Figure 9. Ratio Ratio power absorbed the product water vs. the dimensionless moisture content Figure 9. ofof power absorbed by theby product water vs. the dimensionless moisture content at 50 and 75 mbar. at 50 and 75 mbar. to a moisture content of about 3% d.b., such that the drying rate was higher up to this moisture content. The amount of water in the material influenced the absorbed power. For moistures close to 5% d.b., the absorbed power per unit water mass in the product was constant, thereafter increasing due to absorption by the granule constituents. ACKNOWLEDGEMENTS The authors are grateful to CAPES Coordenao de Aperfeioamento de Pessoal de Nvel Superior, Brazil; CNPq - Conselho Nacional de Pesquisa, Brazil; FURP - Fundao Para o Remdio Popular, Brazil and to cole des Mines dAlbi Carmaux, France. REFERENCES
Arlabosse, A., Rodier, E., Ferrase, J.H., Lecomte, D. (2003). Comparison between static and dynamic methods for sorption measurements. Drying Tech249

Based on the results obtained, it was concluded that for the microwave assisted vacuum process the drying kinetics were similar for the two absolute pressures. The power absorbed at 75 mbar was slightly higher. The drying process to a moisture content of 3% d.b. was accomplished in the regimen of water evaporation, since the product temperatures were lower than the boiling water temperatures at the respective pressures. Concerning the granule studied, in agreement with the sorption isotherm, it was verified that the product showed water slightly bound up
International Microwave Power Institute

nology. 21(3), pp.479 497. Cui, Z. W., S. Y. Xu, D. W. Sun, (2004). Microwavevacuum drying kinetics of carrots slices. Journal of Food Engineering. 65, pp.157-164. Drouzas, E. T, Saravacos, E.D. (1999). Microwave / vacuum drying of model fruit gels. Journal of Food Engineering. 39, pp.117-122. Calligaris, D. personal communication, Furp Fundao para o Remdio Popular, Guarulhos, So Paulo, 2001. Pr, C. E. Rodier, (2002). Microwave vacuum drying of porous media: experimental study and qualitative considerations of internal transfers. Chemical Engineering and Processing. 41, pp.427 436. Pr, C. (1999). tude du schage sous vide et sous microondes de billes de verre et granuls pharmaceutiques (thse de doctorat), cole de Mines de Paris, Paris, France.

250

Journal of Microwave Power & Electromagnetic Energy

Vol. 40, No. 4, 2007