Collected Applications Thermal Analysis PHARMACEUTICALS C

Transcription

1 Collected Applications Thermal Analysis PHARMACEUTICALS C

2 Preface Thermal Analysis (TA) is the term used to describe all the analytical techniques that measure the physical and chemical properties of a sample as a function of temperature. The potential applications of thermal analysis in the pharmaceutical industry are numerous on account of the different chemical-physical aspects of investigations. Amongst others these include method development, characterization and specification of active and inactive ingredients, safety analysis or routine analysis in quality control and stability studies. This booklet describes applications of thermal analysis in the pharmaceutical industry with the help of selected examples i.e. the possible uses of TA in the development of new pharmaceutical substances through to the quality control of commercial products. In particular we would like to thank Dr. Danièle Giron and Dr. Sabine Pfeffer (Novartis Pharma AG, previously TRD, Sandoz Pharma, Basle, Switzerland) for their expert assistance and support which contributed greatly to the success of this booklet, and also to thank Dr. Thomas Gübeli for making facilities available in the Chemical Development Department, Analytical Research, Sandoz Pharma, Basle, Switzerland. We thank Professor Dr. P. C. Schmidt of the Department of Pharmaceutical Technology, Eberhard-Karls- University Tübingen, Germany for preparing numerous application examples. In addition we would like to thank all the other people who were involved in the preparation of this booklet including Helga Judex who was responsible for the layout. Elisabeth Schwarz, Basle Dr. Jürgen de Buhr, Schwerzenbach This application booklet presents selected application examples. These have been tested with the utmost care using the analytical instruments mentioned in the booklet. The experiments were conducted and the resulting data evaluated according to the current state of our knowledge. The application booklet does not however absolve you from personally testing the suitability of the examples for your own methods, instruments and purposes. As the use and transfer of an application example are beyond our control, we cannot accept any responsibility. When chemicals and solvents are used, the safety rules and instructions of the manufacturer must be observed. All names of commercial products can be registered trade marks even if they are not denoted as such. METTLER TLED Collected Applications TA PHARMACEUTICALS Page 1

3 Content Preface... 1 Introduction to Thermal Analysis... 4 Application verview Pharmaceuticals... 8 Some Comments on the Pharmaceutical Industry... 9 Applications of Thermal Analysis in the Pharmaceutical Industry Applications and Techniques Sample Preparation and Method Choice 1 DSC Calibration, Temperature and Heat flow TGA Calibration, Temperature DSC Calibration, Heating Rate Independence The Influence of Heating Rate on the Detection of Polymorphism, Butylated Hydroxyanisole Influences on Crystallization Behavior, Saccharose Solutions Influence of the Heating Rate on Moisture Content Determination, an /W Cream Influence of the Heating Rate on Decomposition, Metolazone Influence of the Pan on Dehydration, Glucose Monohydrate Sample Preparation, Butylated Hydroxyanisole Influence of the Sample Weight, Butylated Hydroxytoluene Influence of the Pan on the Determination of Moisture Content, Cellulose Sample Storage and Hygroscopic Effects xidation Stability of ils Influence of Thermal History and the Evaluation of the Glass Transition, Polystyrene Identification and Characterization 15 Melting Point Determination, Vanillin Characterization of the Melting Behavior, Vanillin Phase Changes, Cholesteryl Myristate Identification Based on Melting Behavior, Polyethylene Glycol Melting Point Depression of Water, Sugar Solutions DSC 'Fingerprint', /W Cream Glass Transition, Poly (D,L-lactide)-Co-Glycolide (DLPLGGLU) Glass Transition and Moisture Content, Hydroxypropoxymethylcellulose Phthalate (HPMC-PH) Quality Control, PE Films Page 2 PHARMACEUTICALS METTLER TLED Collected Applications TA

4 3. Stability 24 Decomposition, Hydrocortisone Decomposition at the Melting Point, Dihydroergotamine Mesylate Melting Behavior and Decomposition, Aspartame Total Decomposition, Malonic Acid Kinetic Analysis of Decomposition, Hydrate Stability, Theophylline Moisture, Starch/NaCMC (Primojel) Polymorphism 31 Polymorphism, Tripalmitin Polymorphism, Tolbutamide Polymorphic Modifications by Annealing, Butylated Hydroxyanisole DSC 'Fingerprint', Magnesium Stearate Polymorphism, L-Polylactide Polymorphism, Sulfapyridine Pseudopolymorphism 37 Pseudopolymorphism, Glucose Monohydrate Enantiomers 38 ptical Purity, Ibuprofen Purity 39 Purity using DSC and HPLC, 4-Hydroxybenzoic Acid and its Esters Purity Determination, Phenacetin + 4-Aminobenzoic Acid Purity and Recrystallization, Cholesterol Phase Diagrams 42 Phase Diagram, Tolbutamide and PEG Eutectic Composition, Methyl-4-Hydroxybenzoate and 4-Hydroxybenzoic Acid Quantification/Detection 44 Solvent Detection by means of TG-MS, Pharmaceutically Active Substance Quantification, /W Creams with Different Water Content Quantification,Theophylline Monohydrate Determination of an Active Substance, Alcacyl Literature Index Notes METTLER TLED Collected Applications TA PHARMACEUTICALS Page 3

5 1 DSC Calibration, Temperature and Heat flow Sample Indium (calibration standard, purity > %) Application Standard for calibration Conditions Measuring cell: DSC821 e Pan: Aluminum 40 µl, with pierced lid Sample preparation: Indium pellet, pressed flat, premelted DSC measurement: Heating from 120 C to 180 C at 10 K/min Atmosphere: Nitrogen, 50 cm 3 /min Interpretation The DSC curve shows the melting of indium. A pure substance melts at an exactly defined temperature, its melting point. The melting point is taken to be the start or onset of the melting process which is defined as the temperature given by the intercept of the extrapolated slope of the melting curve and the continuation of the base line. Evaluation The onset temperature and the heat of fusion of indium are determined. The fully automated evaluation performs a validation which compares the measured values with literature values. If, as in this case, the values lie within the allowed limits then the message The DSC module is within specifications is displayed. Measured Ref. value Tolerance Melting point (onset) ± 0.3 C Heat of fusion ± 0.6 J/g Page 16 PHARMACEUTICALS METTLER TLED Collected Applications TA

6 Conclusion The so-called indium-check is a quick and easy method to check the temperature and heat flow calibration of an instrument. The results are automatically compared with reference values. The instrument displays the appropriate message if an adjustment of the instrument is required. If the instrument is frequently used in other temperature ranges, then further checks with additional standards suitable for those temperature ranges are recommended. The tolerances given in this example are standard values and can be individually adapted. METTLER TLED Collected Applications TA PHARMACEUTICALS Page 17

7 2 TGA Calibration, Temperature Samples Indium and aluminum (calibration standards, purity > %) Application Standards for temperature calibration Conditions Measuring cell: TGA850 Pan: Alumina 70 µl Sample preparation: Two indium pellets, one piece of aluminum wire. The metals are pressed flat. Weight of each sample approx. 12 mg. The pure metals are put in the pan well separated from each other. TGA measurement: Heating from 100 C to 200 C at 10 K/min, from 200 C to 600 C at 50 K/min, from 600 C to 700 C at 10 K/min (indiumaluminium check). Atmosphere: Nitrogen, 20 cm 3 /min Interpretation The SDTA signal shows the melting of both metals. The automatic evaluation determines the melting points (onsets) of both metals and compares them with the reference values. If the deviations are too large then the appropriate message is displayed. In this example the results lie within the specifications. The two peaks at 230 C and 600 C are caused by the change of heating rate. The weight curve is not shown because no effects would be observed. Page 18 PHARMACEUTICALS METTLER TLED Collected Applications TA

8 Evaluation Determination of the melting points (onsets) of indium and aluminum using the SDTA curve. Indium Aluminum measured Ref. value measured Ref. value Melting point 1 ) ± ± 1.5 C Melting point 2 ) ± ± 3 C 1 ) based on the thermocouple of the sample holder (sample temperature) 2 ) based on the thermocouple of the furnace (abscissa unit) Conclusion The so called indium-aluminium check is a quick and easy method to check the temperature calibration of the thermobalance. If the measured values lie outside the given tolerances, then the settings can be adjusted. If the thermobalance (as in the case of the TGA850 or TGA/SDTA851 e ) is equipped with a suitable sample temperature sensor, then the check is performed using the melting of known standards. therwise the method using the Curie point transition temperatures of different metals is used. A special weight calibration is not usually be performed, because many balances already have an automatic internal test procedure available. A certified weight can be weighed at defined intervals. For GMP investigations the use of reference materials to check the weight calibration is recommended e.g. calcium oxalate monohydrate (Pharma Eur. 1997). METTLER TLED Collected Applications TA PHARMACEUTICALS Page 19

9 3 DSC Calibration, Heating Rate Independence Sample Zinc (calibration standard, purity > %) Application Standard for calibration Conditions Measuring cell: DSC821 e Pan: Alumina 40 µl, with pierced lid Sample preparation: Zinc pellet, pressed flat and premelted DSC measurement: Heating from 350 C to 475 C at 5, 10 and 20 K/min. All measurements are performed with the same sample. Cooling rate is 5 K/min. Atmosphere: Nitrogen, 50 cm 3 /min Interpretation The DSC curves show the melting of zinc at different heating rates. If displayed with respect to temperature, the peak area increases with increasing heating rates. The heat of fusion is however the integral of the heat flow with respect to time. This is just as independent of the heating rate as the melting point. Evaluation The onset temperature and the heat of fusion of zinc are determined. Measured Measured Measured Ref. value Heating rate K/min Melting point (onset) C Heat of fusion J/g Page 20 PHARMACEUTICALS METTLER TLED Collected Applications TA

10 Conclusion The calibration data of an instrument are affected by many factors such as the heating rate, the purge gas used, the sample pan material and the temperature range used. nly when these effects can be taken into account in the calibration data, are results obtained that are independent of measurement parameters, as is shown in this example of the melting point and the heat of fusion. METTLER TLED Collected Applications TA PHARMACEUTICALS Page 21

11 5 Influences on Crystallization Behavior, Saccharose Solutions Sample Application D(+) Saccharose solution, 20 weight % in water (= 1.05 mole %) Inactive ingredient (solution stabilizer) CH 2 CH 2 CH 2 Conditions Measuring cell: DSC820 Pan: Aluminum 40 µl, hermetically sealed Sample preparation: ne drop of solution is weighed into the pan using a fine pipette, sample weight mg. DSC measurement: Cooling from 25 C to -50 C at -1, -2, -5, -10, -20 K/min. The same sample is used for all the measurements. Heating from -50 C to 25 C at 5 K/min. Atmosphere: Nitrogen, 80 cm 3 /min Interpretation The curves show that crystallization and melting processes can be measured with the DSC. At low cooling rates, the onset temperatures are almost constant, but are displaced to lower values (supercooling) at higher cooling rates. At very high cooling rates it is even possible that the solution does not crystallize but vitrifies i.e. is transformed to glassy state. The melting point depression and the purity of the water can be calculated from the melting peak. Page 24 PHARMACEUTICALS METTLER TLED Collected Applications TA

12 Evaluation Cooling rate (K/min) nset, C H, J/g Effect crystallization melting Purity calculated using the Melting peak: mole % (theoretical value: 98.95) Melting point depression: C Conclusion The onset temperatures of the melting and crystallization processes are different. Crystallization processes are controlled kinetically by nucleation and are dependent on the cooling rate and the amount of sample (number of nuclei present). The onset temperatures of melting peaks are not normally subject to disturbing influences. METTLER TLED Collected Applications TA PHARMACEUTICALS Page 25

13 6 Influence of the Heating Rate on Moisture Content Determination, an /W Cream Sample /W Cream sample 647-A Application Basic material for the manufacture of creams Conditions Measuring cell: TGA850 Pan: Aluminum 100 µl, with pierced lid. The lid was automatically pierced shortly before the measurement (sample changer with needle, 1 mm diameter) Sample preparation: As received, no preparation TGA measurement: Heating from 20 C to 200 C at 2 and 5 K/min. Both measurements are blank curve corrected. Atmosphere: Nitrogen, 50 cm 3 /min Interpretation The TGA curves show the evaporation of the volatile components (mainly water) in the region between 40 C and 140 C. At higher heating rates the evaporation is diplaced to higher temperatures. The first derivative of the TGA curve is helpful for the determination of the final step of the TGA signal. Evaluation Heating rate Step, % Peak temperature DTG, C 2 K/min K/min Conclusion The influence of the mass and form of the sample, the heating rate and the type of pan have to be considered when developing methods. The heating rate is of special importance when investigating time and temperature dependent effects such as evaporation. Page 26 PHARMACEUTICALS METTLER TLED Collected Applications TA

14 8 Influence of the Pan on Dehydration, Glucose Monohydrate CH 2 Sample α-d-glucose monohydrate H 2 Application Inactive ingredient, filler for tablets and capsules Conditions Measuring cells: DSC820 or TGA850 Pan DSC: Aluminum 40 µl, hermetically sealed or with pierced lid Pan TGA: Aluminum 100 µl, with pierced lid Sample preparation: As received, no preparation DSC measurement: Heating from 30 C to 250 C at 20 K/min TGA measurement: Heating from 30 C to 300 C at 20 K/min Atmosphere: Nitrogen; DSC: 50 cm 3 /min, TGA: 80 cm 3 /min Interpretation A comparison of the two DSC curves of α-d-glucose monohydrate shows the changes that arise when the sample is measured in a sealed pan or in a pan with a pierced lid. In a hermetically sealed pan the sharp melting peak of the monohydrate can be observed. If a pan with a pierced lid is used, the water of crystallization can escape. This is noticeable as a shift of the DSC curve at the beginning of the measurement and as a broad evaporation peak. At the same time, a transition to β-d-glucose anhydrate occurs, the melting point of which is at about 158 C. Above 200 C the glucose starts to caramelize. Evaluation DSC Measuring conditions nset, C Effect Sealed pan 81.4 melting Pan with pierced lid melting METTLER TLED Collected Applications TA PHARMACEUTICALS Page 29

15 Interpretation Thermogravimetric measurements using a pan with a pierced lid confirm the interpretation of the results obtained from the DSC curves, in particular the weight loss caused by the evaporation of the water of crystallization as well as the melting of the β-d-glucose anhydrate afterwards. The weight loss step of 7% between 53 C and 134 C is somewhat less than that expected stoichiometrically. It can be explained however by a loss of water of crystallization during storage of the sample. Evaluation TGA Temperature, C Effect TGA, step % weight loss (water of crystallization) SDTA, onset 59 endothermic peaks SDTA, onset melting peak Conclusion A substance that contains water of crystallization and its anhydrous form normally have different melting points (pseudopolymorphism). The melting point of the form containing the water of crystallization can be determined in a hermetically sealed pan, provided that no decomposition occurs. In an open pan the water of crystallization can escape so that the melting point of the anhydrous form is measured. The presence of a form with water of crystallization should always be confirmed by measuring the weight loss. Page 30 PHARMACEUTICALS METTLER TLED Collected Applications TA

16 9 Sample Preparation, Butylated Hydroxyanisole Sample Butylated hydroxyanisole Application Inactive ingredient (antioxidant) H 3 C C(CH 3 ) 3 Conditions Measuring cell: DSC820 Pan: Aluminum 40 µl, hermetically sealed Sample preparation: As received (1) or crystals ground in a mortar (2) DSC measurement: Heating from 30 C to 70 C at 2.5 K/min Atmosphere: Nitrogen, 50 cm 3 /min Interpretation The two curves show the effects that sample preparation can have on the results. In both cases, two melting peaks can be observed that differ noticeably in temperature range and in the heats of fusion. The explanation lies in the polymorphic behavior of butylated hydroxyanisole. The two peaks correspond to the possible crystal modifications. Evaluation Sample preparation nset 1, C H, J/g nset 2, C H, J/g as received ground in a mortar Conclusion A difference in sample preparation (especially mechanical treatment) can lead to different results. This is particularly the case with substances that exhibit polymorphism. METTLER TLED Collected Applications TA PHARMACEUTICALS Page 31

17 10 Influence of the Sample Weight, Butylated Hydroxytoluene C(CH 3 ) 3 Sample Butylated hydroxytoluene H 3 C Application Inactive ingredient (antioxidant) Conditions Measuring cell: DSC821 e with IntraCooler Pan: Aluminum 40 µl, hermetically sealed Sample preparation: As received, no preparation DSC measurement: Heating from 50 C to 80 C at 2.5 K/min Atmosphere: Air, stationary environment, no flow C(CH 3 ) 3 Interpretation Evaluation The curves show the melting peaks as a function of sample weight. As expected, the peaks in the original presentation (ordinate in mw) increase in height but also in width with increasing weight. Because of this the resolution decreases. In contrast, the normalized presentation shows that the lowest sample weight gives the highest peaks. The onset temperature and heat of fusion of the peaks are determined. The mean values of a number of measurements are presented in the table. Sample weight nset, C Heat of fusion H, J/g 18 ± 0.3 mg 69.4 ± , 84.7, ± 0.3 mg 69.6 ± , ± 0.4 mg 69.5 ± , 84.1, 83.6 Conclusion The sample weight influences the shape of the melting peak. The time required for melting is longer for larger samples because a greater amount of heat has to be transferred. As a result of this, the peaks are shifted to higher temperature. For comparison purposes, the measurement of samples of similar weight is recommended. Samples that are too large are disadvantageous: the peaks become broad (lower resolution) and non-uniform melting leads to irregularly shaped peaks. Page 32 PHARMACEUTICALS METTLER TLED Collected Applications TA

18 11 Influence of the Pan on the Determination of Moisture Content, Cellulose H H H 2 C H 2 C H H H 2 C n H 2 C H Sample Microcrystalline Cellulose (Avicel) Application Inactive ingredient (gel binder, adsorption agent, flow improver) Conditions Measuring cell: TGA850 with sample robot Pan: Aluminum 100 µl, without a lid or with a pierced lid. The lid was pierced automatically immediately before the measurement (needle diameter 1 mm). Sample preparation: As received, no preparation TGA measurement: Heating from 30 C to 300 C at 20 K/min, all measurements are blank curve corrected. Atmosphere: Nitrogen, 80 cm 3 /min Interpretation Cellulose and its derivatives easily take up water from the surroundings or release water depending on the humidity in the laboratory. METTLER TLED Collected Applications TA PHARMACEUTICALS Page 33

19 At the same heating rate, the release of moisture during the measurement occurs more rapidly if no lid is used than with a pierced lid. This has to be taken into account when comparing methods using open pans to methods in which pierced lids are used (sample changer operation). Samples which are in open pans on the sample changer turntable can lose moisture because of the relatively low humidity of the surroundings. The true moisture content of samples can be determined using sealed pans whose lids are pierced immediately before the measurement. Evaluation Pan Step, % Initial value of the curve Corrected sample weight *, % step, % Sealed, lid pierced immediately before the measurement Without lid * riginal sample weight = 100% Conclusion The possible effect of the delay between the weighing out and the actual measurement of the sample must be investigated when determining the moisture content with TGA. When using pans with pierced lids, the smaller the hole in the lid the more the weight loss step is shifted to higher temperature. Page 34 PHARMACEUTICALS METTLER TLED Collected Applications TA