Thermal Analysis – A Review of Techniques and Applications in the Pharmaceutical Sciences

Introduction

The current field of thermal analysis is both diverse and dynamic. Although not a new field, more advanced instrumentation, techniques and applications are constantly appearing on the market and in the literature. Theoretically, almost any substance whether solid, semi-solid or liquid can be analyzed and characterized with thermal analytical techniques. Common materials include foods, pharmaceuticals, electronic materials, polymers, ceramics, organic and inorganic compounds, even biological organisms. In theory, all thermal analytical techniques simply measure the change of a specific property of a material as a function of temperature. This in turn allows researchers access to information regarding macroscopic theories of matter including, equilibrium and irreversible thermodynamics and kinetics [1]. While numerous techniques are available, the primary differences in the techniques are the properties of the material being studied as listed in Table 1.

Table 1    -    Common Thermal Analysis Methods and the Properties Measured

In the pharmaceutical sciences, only a handful of the techniques are commonly employed but the information gained and phenomena that can be explored are countless. The primary workhorses in the pharmaceutical sciences include, differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), differential thermal analysis (DTA) and dynamic mechanical analysis (DMA). Admittedly, as the needs of the researcher change and new materials are identified in formulation development, less commonly used techniques are being utilized and developed resulting in a very dynamic and exciting field of research. This review will be presented in two major sections. The first will be thermal analytical methods commonly used in the pharmaceutical sciences, primarily DSC (including several specialized techniques) and TGA. The second will focus on applications in the pharmaceutical sciences including solid-state characterization of polymorphism, solid dispersions and polymeric dosage forms.

Common Thermal Analysis Methods In the Pharmaceutical Sciences

Differential Scanning Calorimetry (DSC)

If any laboratory, ranging from the pharmaceutical industry to academic research, were to purchase only one piece of thermal analysis equipment it would most likely be a DSC. These instruments are available from several manufacturers in a wide range of price and applications. Instruments are available from simple robust DSC’s with robotics and high-throughput capabilities for screening and quality control, to high-end, extremely sensitive instruments for research applications.

The concept of DSC was originally derived form earlier DTA instruments. The basic difference between the two techniques is that DTA measures a difference in temperature whereas DSC, in theory allows for the measurement of a change in enthalpy. Classical DTA measures the difference in temperature between the sample and reference (which undergoes no phase changes) as the furnace goes through a computer controlled temperature program. The temperatures of both the sample and reference material increase (or decrease) uniformly until the sample reaches a point that it undergoes a phase change, which is either endothermic or exothermic and the difference in temperature is recorded. Although useful information is gained using DTA, there are several drawbacks to the technique which result in making it difficult to quantify the results and obtain information regarding the enthalpy of the sample transitions [2].

These difficulties were overcome with the development of DSC. The International Confederation for Thermal Analysis and Calorimetry (ICTAC) defines DSC as a technique where the heat flow rate difference into a sample and reference material is measured [3]. Two basic types of DSC instruments are commercially available today. One is heat-flux DSC (hf-DSC), initially referred to as Quantitative DTA, and power compensation DSC (pc-DSC). Although the literature is quick to point out technical differences and advantages/disadvantages of each, the end user will most likely find that the two techniques are very comparable and extremely versatile. While employing different techniques to obtain the measurement, both types of instruments measure heat flow and this seem to qualify as DSC under the ICTAC definition. Another important difference that should be pointed out is that at this time there is no common convention as to how DSC results should be presented. Depending on the instrument used, the default settings may be to report endothermic events in the upward direction or downward with exothermic events in the opposite direction. To the author’s knowledge, the software for all instruments has the capability to change this convention but it is usually left up to the researcher’s preference.

hf-DSC

Initially the term heat-flux DSC was used to describe quantitative DTA instruments [4]. Today it is commonly referred to as a DSC method. This development was an improvement over DTA in that it allowed for a measurement in the changes in heat flow as opposed to temperature only. This was accomplished by the addition of a second series of thermocouples to measure temperatures of the furnace and a heat sensitive plate. During a phase change, the heat emitted or absorbed by the sample would alter the heat flux through the heat sensitive plate. By measuring the heat capacity of the heat sensitive plate as a function of temperature during the manufacturing process, an estimate of the enthalpy of transition can be made from the incremental temperature fluctuations [4].

pc-DSC

Power-compensation DSC differs from hf-DSC in both operating principle and basic instrument design. As the name implies, pc-DSC measures the change in energy or power necessary to maintain the sample and references material at the same temperature throughout the heating or cooling cycle. (Technically the temperature difference is maintained below a threshold of typically <0.01ºK.) This is accomplished through a different instrument design than that commonly found in hf-DSC instruments. With pc-DSC, two individual heaters are used to control the heat flow to the sample and reference holders. Individual resistance sensors are placed within each holder and measure the temperature at the base of each. When a phase change occurs and a temperature difference is detected between the sample and reference, energy is supplied (or removed) until the temperature difference is below the threshold previously mentioned. Energy input as a function of temperature (or time) is then recorded which is proportional to the heat capacity of the sample [4].

DSC Results

Figure 1 - Representative DSC Thermogram

Figure 2 - Typical TGA and DSC Results for Various Transitions. (From Reference 4)

Regardless of the type of instrument used researchers will find that DSC offers a plethora of information regarding phase changes of materials. All DSC instruments operate over a wide temperature range both isothermally and dynamically and can be used for solids, semi-solids and liquids. A typical DSC thermogram is shown in figure 1. Although DSC results reveal temperatures or temperature ranges, at which endothermic or exothermic events occur, interpretation of the results can be challenging. Figure 2 presents several of the events which are typically measured with DSC. As noted in the figure, several events can be endothermic or exothermic but with a basic understanding of thermodynamics the researcher can begin to deduce which event is most likely. For example, both melting and evaporation are endothermic events. However for most crystals (especially when pure) melting peaks are usually much sharper indicating a very small temperature range over which the transition occurs. Likewise, decomposition and oxidation are both broad exothermic peaks, but with the ability to control the atmosphere in which the experiment is conducted the difference can be determine. Figure 2 also shows how thermogravimetric analysis (TGA) is also commonly used in the interpretation of DSC results. Crystallization and decomposition may both produce fairly sharp exothermic peaks on DSC but only the decomposition will be accompanied by decrease in mass on TGA in the same temperature range.

Modulated Temperature – DSC (MT-DSC)

As shown in figure 2, most transitions detected by DSC will appear as peaks, where a change (exothermic or endothermic) is detected and then there is a return of the heat flow to a baseline. These results are typical of first- or second-order thermodynamic phase transitions, which are in an equilibrium state. Glass transitions, on the other hand, are neither first- nor second-order transitions since neither the glassy state nor the viscous state is an equilibrium state [4]. Typical DSC thermograms will reveal glass transitions as step-wise increases in the heat capacity (Cp) of the sample. This is due primarily to the increase in molecular motion of the sample above the Tg. In some cases, the determination of Tg is relatively straightforward but this work can be some of the most challenging done with DSC. More on detecting and determining Tg’s will be presented in the applications section of this review but glass transitions are mentioned here because this application has help drive the development of modulated-temperature DSC instruments and methods.

Frequently a sample in which glass transitions need to be studied will contain material that is only partially amorphous or may be in complex mixtures of materials that are crystalline or amorphous. This can result in glass transitions occurring in a temperature range that is relatively close to other endothermic or exothermic transitions, which due to their larger signal can “cover up” the Cp increase indicative of the glass transition. Additionally the thermal history of an amorphous sample will greatly influence possible glass transitions. To overcome some of these challenges, manufacturers have developed software techniques that can in theory separate kinetically reversible and irreversible events, which allows the researcher to more clearly detect changes in the Cp. This is accomplished by applying a perturbation to the heating program of a conventional DSC followed by a deconvolution of the results by mathematical processes to separate the reversible and irreversible events. In theory, changes in the Cp of the sample are considered reversible and most but not all changes in enthalpy can be identified by a change in heat flow considered irreversible. The separation or deconvolution of Cp and enthalpy signals is described in equation 1 below [5].

Where Q = amount of heat evolved, Cp = heat capacity, T = absolute temperature, t = time, and f(t,T) is a function of time and temperature that governs the response associated with the physical or chemical transformation [5]. While this can be accomplished with either pc- or hf-DSC instruments, recent improvements in the sensitivities and more accurate temperature control for both types of instruments have also contributed to making MT-DSC a valuable and accepted technique.

Thermogravimetric Analysis (TGA or TG)

Thermogravimetric analysis is an experimental method whereby changes in mass are used to detect and measure the chemical or physical processes that occur upon heating a sample [5]. Figure 2 represents common TGA results and demonstrates how they are commonly used in interpreting results from DSC analysis. Although several variations in the overall design exist, the simple design of TGA instruments includes a highly precise analytical balance to which the sample pan is attached. The sample pan is typically suspended within a heater, which is under computer control. The control of the balance and sample atmospheres again varies but in basic instruments, they share a controlled atmosphere usually of nitrogen gas. Instruments are also available that allow the user to provide one atmosphere for the balance and another for the sample/heater area and control the pressure as well. Most instruments also have the ability to use isothermal or dynamic heating and cooling cycles and some can also be programmed to hold a specific temperature once a change in mass is detected then maintain that temperature until there is no further change and resume the heating program. TGA results like DSC can vary greatly depending on sample and experimental conditions which can make it very difficult to compare the work of one researcher to another. An excellent explanation of some of these challenges can be found in references [2,5].

In addition to supporting other thermal analysis techniques, TGA is extremely useful in studying various kinetic processes of solids and liquids as long as the processes involve the loss of mass. This is achieved with the accuracy of the balances used as well as the precise control of heating/cooling rates and atmospheric conditions. Other common applications in the pharmaceutical sciences include the characterization of hydrates including the desolvation process and the determination of decomposition, vaporization or sublimation temperatures.

A few instruments are commercially available that allow simultaneous TGA and DSC analysis from a single sample. This may be a convenient and time saving feature and although quite accurate, the performance of each technique does suffer somewhat due to construction requirements.

Thermal Analysis Applications in the Pharmaceutical Sciences

Although most thermal analysis methods can deal with samples as solids, semi-solids or liquids, a review of the current literature would suggest that the broad term, solid-state characterization, could apply to a majority of the applications in pharmaceutical research. Common applications include the characterization of the physicochemical properties of crystalline solids and the detection and identification of polymorphic forms. With the increased utilization of solid dispersions and other polymeric dosage forms, thermal analytical techniques have been called upon more frequently to aid researchers with their development and characterization. Thermal analytical techniques are also used to study the effects of lyophilization and to develop optimal lyophilization formulations and cycles. Several techniques are also used to study kinetics in the solid-state, including decomposition, accelerated stability and the effects of aging on various formulations.

Polymorphism

A crystalline solid may form polymorphs under common manufacturing conditions. Because the physicochemical properties of various polymorphs can vary greatly, identifying the possibility of polymorph formation is critical in product development. A review of the thermal analysis of kinetically reversible/irreversible polymorphic transitions was recently published by Kawakami [6]. DSC has proven to be very useful in the identification of polymorphic transitions primarily due to the ability to easily study the sample under various heating and cooling conditions needed to induce the polymorph formation [7,8,9,10]. Additionally once identified, DSC can be used to monitor samples for the development of polymorphs under various storage conditions or under various manufacturing conditions such as grinding, heating, and drying [11, 12].

Solid Dispersions and Polymeric Dosage Forms

Efforts to formulate drugs with poor aqueous solubility and bioavailability have led to the increased utilization of solid dispersions. Solid dispersions consist of at least two solid components one being the matrix forming component the other being the drug. The drug may be dispersed in the matrix as either crystals, amorphous clusters or uni-molecularly in what is commonly termed a solid solution. These dispersions may then be formulated into granules, beads, films, microspheres, tablets etc. and used for immediate or controlled-release products by various routes of administration including, oral, transdermal, transmucosal and topical.

Karavas, et. al., used DSC to characterize a solid dispersion system of felodipine and polyvinylpyrrolidone (PVP) [13]. The authors found that the felodipine was present in nano-scale particle sizes that were dependent on the felodipine/PVP ratios. DSC revealed that partial miscibility was present between the two components which led to an optimal glass dispersion of the felodipine in the matrix and resulted in a significant enhancement of the dissolution and release kinetics of the felodipine. Similarly, Mura, et.al., used DSC in combination with other techniques to study the dissolution of solid dispersions of ketoprofen and PEG 15000 with and without various anionic or ionic surfactants [14]. DSC was used to characterize and confirm the formation of the dispersions produced by various techniques and dissolution profiles were studied with standard methods. Solid dispersion formulations are also finding increased use in the development of transdermal formulations. Cho, et.al. offer one example in which DSC was used to characterize the dispersion of the drug quinupramine in an ethylene-vinyl acetate (EVA) matrix [15]. The dispersion, which was also studied by X-ray diffraction, was found to contain quinupramine in an amorphous state which led to enhanced drug release. DSC results showed that the pure drug existed in the crystalline state and remained in the crystalline state after a physical mixture with EVA was produced as evidenced by an endothermic peak corresponding to the melting point of pure drug. However after a 1:2, matrix of drug and EVA was formed by the casting method, no endothermic peak was observed suggesting drug was present in the amorphous state.

Repka, et.al, have made extensive use of DSC and TGA in the development and characterization of solid dispersions for various routes of administration produced by hot-melt extrusion technology [16-22]. DSC methods were used to assess the miscibility of drug-polymer and polymer-polymer blends and to determine formulations most likely to produce solid solutions or dispersions upon extrusion. Additionally, samples could be analyzed post extrusion to confirm the solid solution or dispersion formation and periodically sampled form accelerated storage conditions to determine if drug was recrystallizing in the matrix. Another important application was predicting optimal extrusion conditions while using the small sample sizes utilized in DSC. Thermal analysis (DSC and TGA) was used to determine temperatures at which drug decomposition occurred and temperatures at which the polymers or polymer blends would melt which is critical in predicting extrusion temperatures that will assure polymer melting yet reduce drug degradation. Studies could also be performed in which formulations were held isothermally at determined temperatures to study the effects of dwell time in the extruder.

Other examples of the complexities encountered when dealing with HME products can be found in recent work published by Qi et.al. [23]. In this work solid dispersions of paracetamol were prepared in a copolymer based on dimethylaminoethyl methacrylate and neutral methacrylic esters by hot-melt extrusion. In order to better characterize these systems the authors employed a combination of thermal analytical techniques including, DSC, MT-DSC, and microthermal analysis (μ-TA).

Pharmaceutical Applications of Modulated Temperature DSC (MT-DSC)

Over the years MT-DSC has been used extensively in lyophilization optimization. A review of this topic is available in American Pharmaceutical Review [24].

DinNunzio et.al., utilized MT-DSC to study amorphous engineered particle compounds of itraconazole (ITZ) with the polymers, cellulose acetate phthalate (CAP) and polyvinyl acetate phthalate (PVAP) produces by an ultra-rapid freezing technique [25]. MT-DSC demonstrated that the ITZ:CAP engineered particles had a strong correlation with the Gordon-Taylor relationship while the ITZ:PVAP particles displayed positive deviations from expected values suggesting hydrogen bonding between the drug and polymer. Further in vitro and in vivo testing revealed improved bioavailability and enhanced intestinal targeting for the CAP containing particles.

Several authors have published work using MT-DSC in the characterization of a wide variety of polymer based formulations and solid dispersions including, drug-loaded hydrogels [26], solid dispersions of itraconazole [27], film-coated melt-extruded pellets [28], and PEO/griseofulvin systems [29].

Studies are also available addressing the use of MT-DSC in the analysis of crystal growth from pharmaceutical melts [30] and pharmaceutical liquid crystals [31].

Conclusion

There are several clichés derived from the statement that thermal analysis is a “hot topic”. A review of the recent literature, application notes from manufacturers as well as instrument brochures and websites would certainly confirm this. As formulations become more and more complex and characterizing them becomes more difficult, manufacturers have done an excellent job in keeping pace with more precise and sensitive yet more durable instruments. The applications discussed in this review represent a tiny fraction of what is currently being done and with continual advances in the field, future applications are limited only by the investigators imagination and of course, budget. APR

References

1. Cheng, S.Z.D., Li, C.Y., Calhoun, B.H., Zhu, L., Zhou, W.W., Thermal analysis: the next two decades. Thermochim. Acta 355 (2000): 59-68.
2. Sestak, J. Thermophysical properties of solids. In: Svehla, G. ed. Comprehensive Analytical Chemistry: Volume XII, Thermal Analysis Part D. New York, NY: Elsevier Science Publishing; 1984: ISBN 0-444-99653.
3. Recommendations for Names and Definitions in Thermal Analysis and Calorimetry. ICTAC website. http://www.ictac.org/ICTACNews/articles/dec98/ind98030.pdf. Accessed January, 2010.
4. Hatakeyama, T., Quinn, F.X. Thermal Analysis; Fundamentals and Applications to Polymer Science. New York, NY: John Wiley & Sons; 1994: ISBN 0-471-95103-X.
5. Craig, D.Q.M., Reading, M. Thermal Analysis of Pharmaceuticals. Boca Raton, FL: CRC Press; 2007: ISBN 0-8247-5814-5.
6. Kawakami, K. Reversibility of enantiotropically related polymorphic transformations from a practical viewpoint: thermal analysis of kinetically reversible/irreversible polymorphic transformations. J. Pharm. Sci. (2007), 96(5), 982-989.
7. Aceves-Hernandez, J.M., Nicola-Varquez, I., Aceves, F.J., Hinojosa-Torres, J., Paz, M., Castano, V.M. Indomethacin polymorphs: Experimental and conformational analysis. J.Pharm. Sci. (2009), 98(7), 2448-2463.
8. Schmidt, A.C., Schwarz, I., Mereiter, K. Polymorphism and pseudopolymorphism of salicaine hydrochloride crystal polymorphism of local anaesthetic drugs, part V. J. Pharm. Sci. (2006), 95(5), 1097-1113.
9. Vega, D., Petragalli, A., Fernandez, D., Ellena, J.A. Polymorphism of leflunomide: stability and crystal structures. J. Pharm. Sci. (2006), 95(5), 1075-1083.
10. Vogt, F.G., Cohen, D.E., Bowman, J.D., Spoors, G.P., Zuber, G.E., Trescher, G.A., Dell’orco, P.C., Katrincic, L.M. Debrosse, C.W., Haltwanger, R.C. Structural analysis of polymorphism and salvation in tranilast. J. Pharm. Sci. (2005), 94(3), 651-665.
11. Ke, X., Ping, Q., Shi, H. Interconversion studies of Betamethasone Acetate polymorphs. Drug Dev Ind Pharm (2005), 31(8) 813-818.
12. Niederwanger, V., Gozzo, F., Griesser, U. J. Characterization of four crystal polymorphs and a monohydrate of S-bupivacaine hydrochloride (levobupivacaine hydrochloride). J. Pharm. Sci. (2009), 98(3), 1064-1074.
13. Karavas, E., Ktistis, G., Xenakis, A., Georgarakis, E. Miscibility behavior and formation mechanism of stablilized felodipine-polyvinylpyrrolidone amorphous solid dispersions. Drug Dev Ind Pharm (2005), 31(6), 473-489.
14. Mura, P., Moyano, J.R., Gonzalez-Rodriguez, M.L., Rabasco-Alvarez, A.M., Cirri, M., Maestrelli, F. Characterization and dissolution properties of ketoprofen in binary and ternary solid dispersions with polyethylene glycol and surfactants. Drug Dev Ind Pharm (2005), 31(4-5), 425-134.
15. Cho, CW., Choi, JS., Shin, SC. Physicochemical characteristics of quinupramine in the EVA matrix. Int. J. Pharm. (2006), 320(1-2), 1-3.
16. Repka MA, Gutta K, Prodduturi S, Munjal M, Stodghill SP. Characterization of cellulosic hot-melt extruded films containing lidocaine. Eur J Pharm Biopharm (2005), 59(1), 189-196.
17. Prodduturi S, Manek RV, Kolling WM, Stodghill SP, Repka MA. Solid-state stability and characterization of hot-melt extruded poly(ethylene oxide) films. J Pharm Sci (2005) 94(10), 2232-2245.
18. Munjal M, Stodghill SP, Elsohly MA, Repka MA. Polymeric systems for amorphous Delta-9-tetrahydrocannabinol produced by hot-melt method. Part I: chemical and thermal stability during processing. J Pharm Sci (2006) 95(8), 1841-1853.
19. Prodduturi S, Manek RV, Kolling WM, Stodghill SP, Repka MA. Water vapor sorption of hot-melt extruded hydroxypropyl cellulose films: effect on physic-mechanical properties, release characteristics, and stability. J Pharm Sci (2004), 93(12), 3047-3056.
20. Repka MA, Elsohly MA, Munjal, M. Ross SA. Temperature stability and bioadhesive properties of delta-9-tetrahydrocannabinol incorporated hydroxypropylcellulose polymer matrix systems. Drug Dev Ind Pharm (2006), 32(1), 21-32.
21. Repka MA, Prodduturi S, Stodghill SP. Production and characterization of hot-melt extruded films containing clotrimazole. Drug Dev Ind Pharm (2003), 29(7), 757-765.
22. Repka MA, Mididoddi PK, Stodghill SP. Influence of human nail etching for the assessment of topical onychomycosis therapies. Int J Pharm (2004), 282(1-2) 95-106.
23. Qi S, Gryczke A, Belton P, Craig DQM. Characterization of solid dispersions of paracetamol prepared by hot-melt extrusion using thermal, microthermal and spectroscopic analysis. Int J Pharm (2009) 354, 158-167.
24. Rameau A, Chevalier A, Neves C. Lyophilization optimization using DSC and MT-DSC. American Pharm Rev (2004) 7(1), 50-60.
25. DiNunzio JC, Miller DA, Yang W, McGinity JW, Williams RO. Amorphous Compositions Using Concentration Enhancing Polymers for Improved Bioavailability of Itraconazole. Molecular Pharmaceutics (2008), 5(6), 968-980.
26. Geever LM, Cooney CC, Lyons JG, Kennedy JE, Nugent MJD, Devery S, Higginbotham, CL. Characterization and controlled drug release from novel drug-loaded hydrogels. Eur J Pharm Biopharm (2008), 69(3), 1147-1159.
27. Janssens S, Novoa de Armas H, Remon JP, Van den Mooter, G. Use of Kollicoat IR in the formulation of solid dispersions of itraconazole. Eur J Pharm Biopharm (2007), 30(3-4), 288-294.
28. Young CR, Crowley M, Dietzsch C, McGinity JW. Physicochemical properties of film-coated melt-extruded pellets. Journal of Microencapsulation (2007), 24(1), 57-71.
29. Stachurek I, Pielichowski K. Characterization of PEO 34000/griseofulvin systems using modulated temperature DSC. Modern Polymeric Materials for environmental Applications, International Seminar, 2nd, Krakow, Poland, Mar.23-25 (2006), 177-180.
30. Di Martino, P. Crystal growth from pharmaceutical melts. From; Horizons in World Physics (2007), 257(Bosons, Ferromagnetism and crystal Growth Research), 221-245.
31. Stevenson CL, Bennett DB, Lechuga-Ballesteros D. Pharmaceutical liquid crystals: The relevance of partially ordered systems. J Pharm Sci (2005), 94(9), 1861-1880.

This article was printed in the March 2010 issue of American Pharmaceutical Review - Volume 13, Issue 2. Copyright rests with the publisher. For more information about American Pharmaceutical Review and to read similar articles, visit www.americanpharmaceuticalreview.com and subscribe for free.