MAY 1978 The Analyst Vol. 103 No. 1226 Simultaneous Techniques in Thermal Analysis* A Review F. Paulik and J. Paulik Institute for General and Analytical Chemistry, Technical University, Gellert ter 4, 1502 Budapest XI, Hungary Summary of Contents Introduction Development of simultaneous methods Difficulties of interpretation Derivative methods Standardisation of experimental conditions EGA and EGD Examination of decomposition of organic compounds TD and ETA EC methods Manipulation of experimental conditions Conclusion TG - DTA Keywords : Review ; thermal analysis ; simultaneous methods derivative and Introduction The development of simultaneous thermoanalvtical differential methods ; methods has taken dace during the past two deLades and this development has man; aspects. However, withi; the scope 2 this review we wish first to analyse the causes, motives, aims and trends that led finally to the development of simultaneous methods.In the 1950s thermal analysis entered a new phase of development. The accuracy obtainable with classical methods had not met more stringent requirements and could not be increased by improving the measuring devices. In those days absolute temperature values could be measured by use of thermocouples with an accuracy higher by orders of magnitude than, for example, that with which decomposition temperatures could be determined by means of differential thermal analysis (DTA). For instance, two literature values for the decomposition temperature of calcium carbonate obtained by DTA are 800 and 900 "C. These values were determined with an error of only $1 "C, but in spite of this, and on the basis of these data, we cannot state with a greater certainty than 5 5 0 "C that the decomposition temperature of calcium carbonate is 850 "C.Also, mass change could be determined with an accuracy higher by orders of magnitude than that with which the amounts of two components could be determined if the decompo- sition processes of the components overlapped. Thus, if the sample mass became constant between the two decomposition processes, the amount of each component could easily be determined with an error of only *0.2%. However, if the two processes overlapped, then, as shown in Fig. 1, the error could increase to $20% or more. Accordingly, it became evident that a further development in thermal analysis could only be expected if new principles of examination were introduced and new measuring techniques developed.One of the most significant results of this activity was the birth and rapid development of simultaneous met hods. * Plenary Lecture presented at the First European Symposium on Thermal Analysis, Salford University, 417 As a consequence, intensive research activity began. September, 1976.418 PAULIK AND PAULIK : SIMULTANEOUS TECHNIQUES Analyst, Vd. 103 Development of Simultaneous Methods The aims that were characteristic of this development were, in addition to multiplying the information obtainable, to increase resolution, to standardise the experimental conditions and to increase selectivity. Experimental conditions were manipulaked in order to realise these aims.The idea of coupling single classical thermoanalytical methods is straightforward, as it is well known that extension of an examination by use of an additional method increases both the certainty of interpretation and the accuracy of evaluation of the thermoanalytical curves not in a proportional, but in a multiple, way. This idea is demonstrated by the following example. Knowledge of the mineral composition of bauxites is important from the point of view of alumina production. However, as Fig. 2 shows, for this extreme example (selected for the purposes of demonstration), the composition of a bauxite sample cannot be determined at all if a thermobalance alone is used. The possible components are as follows (with decomposition temperatures in parentheses) : boehmite, a-A1,0,.H20 (520 "C) ; diaspore, y-A120,.H20 (540 "C) ; kaolinite, A120,.2Si0,.2H,0 (570 "C) ; alunite, K,SO,.- 3A1,03.3S0,.6H20 (570 "C) and K2S0,.3A1,0,.3S0, (790 "C) ; and calcite, Ca0.C02 (800 "C).The thermogravimetry (TG) curve alone cannot indicate which components are present and the decomposition of which components was responsible for the two-step mass-loss process. 50 It 10 mg 50 k 10 mg Temperature .-+ Tem pera tu re/"C Fig. 2. TG investigation of a bauxite. Fig. 1. Accuracy of mass measurement by means of thermogravimetry : (a), when the decomposition pro- cesses of the components of a sample do not overlap; and ( b ) , when overlap occurs. As Fig. 3 shows, the composition of the sample cannot be stated even if the derivative thermogravimetry (DTG) curve is also recorded, though the situation undoubtedly becomes clearer.The DTG peaks indicate the single temperature values characteristic of the different mineral components more accurately than does the TG curve. However, the presence of four bauxite minerals or, more precisely, three combinations of these four, is still possible. These combinations are : boehmite and alunite; boehmite, alunite and kaolinite ; and boehmite, kaolinite and calcite. Although it is not sufficient for the complete solution of the problem, it reduces the number of possible combinations to two, as can be seen from Fig. 4. A small exothermic DTA peak can be seen at 960 O C , which is characteristic of kaolinite only and clearly proves the presence of this mineral.The two possible combinations are, therefore : boehmite, alunite and kaolinite ; and boehmite, kaolinite The DTA curve offers further important information.May, 1978 I N THERMAL ANALYSIS. REVIEW 41 9 \Boehmite In v) 5 0 200 400 600 800 1000 Temperature/"C Fig. 3. Simultaneous TG and DTG inves- 40 tigation of the same bauxite as in Fig. 2. 0 200 400 600 800 1000 Temperature/"C Fig. 4. Simultaneous TG, DTG and DTA investigation of the same bauxite as in Figs. 2 and 3. and calcite. Accordingly, in this instance the simultaneous TG, DTG and DTA examina- tions must be completed by use of one of the evolved gas analytical (EGA) methods in order to determine whether sulphur(V1) oxide or carbon dioxide is liberated, i.e., whether the sample contains alunite or calcite, respectively, and if possible to determine the amount of the gaseous decomposition product.As Fig. 5 shows, by using EGA in addition to the other methods, not only can the qualita- tive composition of the sample be identified with certainty, but also the amount of each identified component can be determined with satisfactory accuracy. Uitticulties of lnterpretation Despite the fact that the combination of different methods is connected with the above- mentioned advantages, up to the 1950s, apparently for no reason, thermoanalysts used only single thermoanalytical methods for their investigations. The question of the reason for this choice of method arises. One of the causes undoubtedly lies in the difficulty of the common interpretation and evaluation of curves recorded by means of different thermoanalytical devices.This diffi- culty is shown for a TG and a DTA curve in Fig. 6. The curves demonstrate the thermal decomposition of the same dolomite sample, investigated by using a thermobalance to obtain the TG curve, and a classical steel block type DTA apparatus. On comparing the courses of the two curves, immediately two contradictions can be seen. Firstly, the TG curve represents the thermal decomposition as a one-step process while the DTA curve shows it as a two-step process. The explanation of this phenomenon lies in the different character of the two curves and the poor resolution of the TG curve. As will be shown below, this problem could be solved successfully by the derivation of the TG curve.At least as serious is the other contradiction. According to the two curves, the thermal decomposition of the dolomite did not take place in the same temperature range. According to the TG curve, the sample had totally decomposed below 900 O C , while according to the DTA curve, only the first part of the decomposition had taken place below this temperature and the decomposition was complete only at a temperature 100 "C higher. This phenomenon420 PAULIK AND PAULIK SIMULTANEOUS TECHNIQUES Analyst, Vol. 103 ~~ 9 DTA I - .-. -.-. '*\. W w o \ I * 18 rng of H 2 0 \5? I I 120 rng of boehrnite 1 I I I \ . \ ' I > 10 mg of H20 72 mg of kaolinite 0 200 400 600 800 1000 Ternperatu re/"C Fig. 5. Simultaneous TG, DTG, DTA and EGA investigation of the same bauxite as in Figs.2, 3 and 4. can be explained by the fact that different experimental conditions lead inevitably to different results. In order to eliminate this difficulty, thermoanalysts made efforts to standardise the experimental conditions. According to the thermoanalytical method applied the thermal changes examined are described in the form of either integral or differential functions. The two different kinds of curve illustrate the transformations in different ways, because of the difference in their resolution. We can detect a difference between two thermal effects described by two curves of different character, even if these curves demonstrate thermal processes, such as mass and enthalpy changes, that occur simultaneously. For instance, even the most experienced thermoanalyst could not decide with certainty at first glance whether the TG and DTA curves in Fig.7 illustrate the decomposition of the same bauxite sample or that of two different bauxites of similar but not identical composition. Derivative Methods This statement is supported by Fig. 8, in which, in addition to the simultaneously recorded DTA With the development of derivative methods this problem was fully eliminated.May, 1978 IN THERMAL ANALYSIS. REVIEW 421 I DTA L 0 200 400 600 800 l ( 600 703 800 900 1000 Temperature/" C Fig. 6. dolomite. Parallel TG and DTA examinations of TG 0 0 200 400 600 800 1000 Temperature/"C Fig. 7. Parallel DTA and TG examinations of bauxite. f I I I I I I 4 0 200 400 600 800 1000 Temperature/"C Fig. 8.Simultaneous TG, DTG and DTA investigations of the same bauxite as in Fig. 7.422 PAULIK AND PAULIK: SIMULTANEOUS TECHNIQUES Analyst, vol. 103 and TG curves of the above-mentioned bauxite sample, the DTG curve is shown. It can be seen that the courses of the DTG and DTA curves are similar, which occurs because owing to their mathematical relationship their characters are identical. Thus the DTG curve creates a basis for the comparison of TG and DTA curves. However, the recording of the DTG curve has further, even more significant, advantages than that just mentioned. The DTG curve, owing to its high resolution, is of significant help in selecting the characteristic temperatures as well as in describing the whole course of the transformation, and it therefore makes possible the reliable determination of the qualita- tive composition of multi-component samples.Thus, in Fig. 8, while reactions that follow one another overlap in the TG curve, these reactions can clearly be distinguished in the DTG curve. So, from knowledge of the TG curve alone, we could not decide whether the step on the TG curve in the temperature interval between 500 and 700°C is due to the decomposition of boehmite, diaspore or kaolinite, or to all of them together, but the DTG curve shows without any doubt that the sample contains boehmite and kaolinite. Further, by projecting the 560 "C minimum of the DTG curve on to the TG cu.rve, even the amount of these two minerals can be determined with a limited accuracy, which is an important additional advantage of recording the DTG curve.These advantages were suspected by all those thermoanalysts who were making efforts to develop the various differential and derivative methods in thermal analysis. Table I is a list of the pioneers in the field. Dejean first suggested the derivation of thermoanalytical functions in 1905. His measuring technique was theoretically correct but owing to practical difficulties his method was not used. The method of de Keyser, published in 1953, also has more theoretical than practical significance. The first derivative method that found practical application was worked out in 1954. The technique, based on the principle of induction, is suitable for recording a DTG curve. These initiatives were soon followed up and in the course of a single decade numerous measuring techniques were worked out for the derivation of thermoanalytical curves (DTA, TG), thermal analysis (TA), thermodilatometry (TD), evolved gas detection (EGD), evolved gas analysis (EGA), electrical conductivity analysis (EC) and thermomagnetic analysis (TM).Although the thermoanalytical curves can easily be derived by use of computers, for various reasons conventional methods are used in preference even now. As these computer tech- niques can technically be classed among computer rather than thermoanalytical methods, they are not listed in Table I. On theoretical con- siderations derivative methods are favoured as differential methods, owing to the principle In Table I differential and derivative methods are distinguished.Derivative TA TG TG TG TG DTA } EGD TD TD TD TG TA EGA EGD TG EC TM TABLE I DERIVATIVE AND DIFFERENTIAL METHODS Differential Date Workers 1905 Dejean 1954 Paulik and co-workers 1955 Paulik and co-workers 1956 Waters TG 1953 de Keyser TG 1955 Lambert 1959 Campbell et al. DTA 1959 196 1 1961 1963 DTA 1963 1965 1965 1966 1966 1969 1972 1974 1975 TG Freeman and Edelman Paulik et al. Paulik et al. Wilburn and Hesford Knofel Garn Rupert Paulik and co-workers Pannktier Byrd Price Forster, et al. Moskalewicz Reference 1 2 3, 4 5, 6 7 8 9 10 11, 6 12, 6 13 14 15 16 17, 18 19 20 21 22 23May, 1978 I N THERMAL ANALYSIS. REVIEW 423 of difference formation, are liable to smaller or greater errors, while derivation can be regarded as the ideal limiting case in difference formation.Standardisation of Experimental Conditions The 1950s saw the introduction of another innovation, the basic idea of which is as follows. If examinations are carried out in such a way that different thermal variables are measured in a single sample, then for each variable the experimental conditions will be identical and there will not be any shift in the course of the curves obtained. This idea is the basic principle of simultaneous methods. Fig. 9 demonstrates the thermal decomposition of dolomite. The curves in Fig. 9(a) were obtained by using a DTA apparatus and a deriving thermobalance separately, while the curves in Fig. 9(b) were recorded by applying the simultaneous TG, DTG and DTA technique. It can be seen that in the latter instance the DTA and DTG curves are congruent.Curves recorded by using two different devices differ in phase and in form. By observing the two pairs of curves we can form an idea of the difficulties that thermo- analysts had in the common interpretation and evaluation of thermal curves recorded by means of separate devices. Also, we can understand why the literature contains data for transformation temperatures of the same material that often differ by 50-100 "C. DTA :---To I 1 I 700 800 900 1 LO 700 800 900 1000 Fig. 9. (a), Parallel and ( b ) , simultaneous TG, DTG and DTA examinations of the same dolomite as in Fip. 6. The explanation for this phenomenon is as follows. Most of the decomposition reactions In a closed system a t of inorganic compounds are processes that lead to an equilibrium.decomposition pressure and temperature. For example, the decomposition pressure of calcium carbonate as a function of temperature is shown in curve 1 in Fig. 10. However, in general, in thermoanalytical investigations the sample is not examined in a closed space but in an open sample holder at atmospheric pressure and in the presence of air or of an inert ugc Therefnre the rnrnnncitinn nf tho ~ f a c in r n n f a r t 1xr;t-h the c n l i A nhgco ic rhanuincr Ior a gven pracricai example. correspond to the points given on the TG curve. ine single values 01 me parriai pressure or carDon aioxiae These values were obtained by extra- temperature. However, it should be noted that the curve was constructed by making certain suppositions and neglecting certain factors, so that the picture obtained is only qualitative. In spite of this the figure characterises the exceedingly complex process taking place in the sample.It is well known that under conditions of thermoanalytical investiga-424 loo 80 ae v; 60 % 2 4 0 - 20 0 PAULIK AND PAULIK : SIMULTANEOUS TECHNIQUES Analyst, VoZ. 103 Ti CaCO3 - - 1 I I I - - q!S-- - - _,_ - -1,- -- r 1 Fig. 10. 1, Decomposition pressure of CaCO, veYsus temperature; and 2, TG curve of CaCO, obtained by using a low-walled crucible. tions reactions leading to an equilibrium are composed of many chemical and physical partial processes, some of which impede the transformation while others promote it. The path of the transformation is determined by the infinite succession of alternations taking place continuously in the formation and resolution of micro-equilibrium states.For example, when the dissociation of calcium carbonate begins, carbon dioxide will appear within the sample, in the space between the grains of the substance. If the partial pressure of this carbon dioxide approaches the value of the decomposition pressure that corresponds to the actual temperature, then the decomposition will slow down. It will even stop if the values of the theoretical and actual pressures become equal. The fact that the reaction leads to an equilibrium produces an impeding effect on the progress of the decomposition, i.e., it sets an upper limit to the decomposition rate. However, the momentary quasi-equilibrium will always overbalance and the decomposition will start again because, owing to the increased temperature, only a higher concentration of carbon dioxide will cause a new quasi-equilibrium to be established.In addition, another factor hastens the decomposition. The mixture composed of carbon dioxide and air diffuses continuously outwards from the sample, while air diffuses towards the centre of it. The carbon dioxide released is then replaced by amounts of newly formed carbon dioxide. However, the increase in sample temperature that actually promotes the heating programme and in an atrn&phere of con.May, 1978 I N THERMAL ANALYSIS. REVIEW 425 process occurs without hindrance only until the sample has taken up an amount of heat that corresponds to its heat capacity. In general, the thermal conductivity of the substances investigated is poor.Therefore, the sample is not able instantaneously to absorb from its surroundings the amount of heat necessary for the transformation to take place as this amount is greater by orders of magnitude than its heat capacity. For example, for calcium carbonate the molar heat capacity, Cp, is 26 cal mol-l "C-l at 890 "C and AHOdle8. is 42 600 cal mol-l. This is the reason why transformations generally take place slowly, as is demonstrated by curves 1 and 2 in Fig. 11. Curve 1 illustrates the ideal course of decomposition of calcium carbonate. In constructing this theoretical curve isothermal conditions and the presence of a carbon dioxide atmosphere were assumed. Curve 2 was recorded with a 10 "C min-l dynamic heating programme and in an atmosphere of carbon dioxide.Owing to the atmosphere of carbon dioxide the decomposition process was made independent of the gas transport, as we have supposed in constructing curve 1. Accordingly, the difference between the courses of the two curves can be attributed solely to the effect of the slow heat transfer (see also Fig. 18). Thus the progress of the decomposition is governed by the rate of temperature increase and of gas diffusion, which will be such that the momentary equilibrium of the system always corresponds to the correlation between decomposition pressure and temperature. Fig. 12 demonstrates the different gas-diffusion conditions obtaining in two sample holders of different shape. Comparison of the partial pressure values of carbon dioxide, indicated at the corresponding points of the TG curve, shows that the concentration of carbon dioxide within the sample changed in the two experiments in different ways.For example, when, in the low-walled sample holder, the decomposition of the sample was complete, the tempera- ture was at 850 "C, while the partial pressure of carbon dioxide was at 370 Torr. In contrast, a t the same temperature and pressure values, only 40% of the sample was decomposed in the high-walled sample holder and the decomposition was complete only at 890 "C. However, the partial pressure of carbon dioxide in the sample had reached 680 Torr a t this temperature. 100 80 60 s 2 40 1 v) 20 0 0 200 I- 2 400 & 0 --. v) 0" 600 760 600 700 800 850 900 Temperature/'C Fig.12. 1, Decomposition pressure of CaCO, vemus temperature; 2 , TG curve obtained with a low-walled crucible; and 3, TG curve obtained with a high-walled crucible. It follows from the above that all of the experimental conditions that may influence the rate of gas diffusion and also the rate of heat transfer exert a significant influence upon the course of thermal decomposition. Such conditions are, for example, the amount of sample, its layer thickness and compactness, size of the grains, heating rate, composition and pressure of the gaseous atmosphere and shape and size of the sample holder. However, for the sake of completeness it should be noted that in some instances, owing to the occurrence of even slower partial processes than those mentioned above (nucleus forma-426 PAULIK AND PAULIK: SIMULTANEOUS TECHNIQUES Analyst, vol.103 tion, nucleus growth, gas diffusion through the compact new phase, etc.) the transformation process will be even more complicated. On the other hand, for endothermic reactions not leading to an equilibrium, the situation is simpler, as the course of such transformations is independent of the concentration of the gaseous decomposition products and in most instances it is disadvantageously influenced only by the slow heat transfer. In the 1950s, recognition of these factors led to research activity to find appropriate means for the standardisation of experimental conditions, but a real breakthrough in this field could successfully be achieved only by the introduction of simultaneous techniques.The solution to the problem, which was to use a single sample, was exceedingly simple and the results obtained were ideal. Despite the rapid and wide application of simultaneous techniques, the parallel application of individual methods remained current and they are still employed nowadays. Great efforts were made to standardise the experimental conditions in the course of the develop- ment of these latter methods too. At the present stage of development both types of measurement have advantages and disadvantages. In the construction of equipment for simultaneous techniques, just in the interest of coupling different methods we are often compelled to select less precise techniques of measurement, while methods applied in parallel compensate for the lack of standard results with greater accuracy.Although the principle of examination precludes the possibility of using identical experi- mental conditions when applying complementary methods, these methods do not play a subordinate role among combined techniques. Let us remember, for example, the importance of complementary X-ray spectroscopy, infrared spectroscopy and electron microscopy. In spite of the importance of all the combined methods mentioned, in what follows we shall deal only with simultaneous measuring techniques. TG - DTA Among thermoanalytical methods TG and DTA are the two that yield the greatest and the most valuable information about the substance investigated. Consequently, it is under- standable that the coupling of these two methods was the first to be attempted.Researchers who made efforts in this direction are listed in Table 11. It can be seen that the first equipment of this type was constructed in 1955, and was suitable for the simultaneous recording of TG, DTG and DTA curves. This initiative was soon followed by several others. TABLE I1 SIMULTANEOUS DTA, TG AND DTG Method DTA - TG - DTG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG DTA - TG - DTG DTA - TG DTA - TG - DTG DTA - TG DTA - TG Date 1955 1957 1959 1960 1960 1961 1962 1962 1962 1963 1963 1964 1964 1964 1965 1968 Workers Reference Paulik et al. 24, 25 Powell 26 Papailhau 27 Blazek and Cisar 28 Reismann 29 Piece 30 Torkar et al. 31 Formanek and Dykast 32 Kissinger and Newman 33 McAdie 34 Kriiger and Bryden 35 Wiedemann 36 Khristianov and Korovyatnikov 37 Saito et al.38 Patai et al. 40 Charsley and Redfern 39 EGA and EGD Apart from TG and DTA methods, the examination of the gaseous decomposition products that evolve on heating may perhaps furnish the greatest and the most useful informationMay, 1978 I N THERMAL ANALYSIS. REVIEW 427 regarding both the composition of the substance investigated and the nature of the reactions taking place in the material. For this purpose a number of different methods have been developed, which are usually divided into two groups, EGD and EGA. According to the measurement technique, the character of the information obtained and the field of applica- tion each of these groups can be further divided into two sub-groups. A number of the EGD methods (Table 111) are based on the detection of gas evolution by means of a thermal conductivity cell or a gas density detector (Group I).The method of examination is simple and the equipment is cheap, but the information obtained is modest. The equipment clearly indicates the evolution of gas and conclusions can be reached based upon the amount of gas evolved. However, these methods are not adequate for the examina- tion of the quality of the evolved gas. In spite of this the combination of these methods with othcr thermoanalytical methods, for example with DTA or TD, or with mass spectrometry (hIS), may be very useful, as they can contribute to the better interpretation of the results obtained and enable physical transformations and decomposition reactions to be clearly distinguished.TABLE I11 EGD METHODS Method combined Method with Date Workers Reference Group I- Thermal conductivity and DTA 1960 Lodding and Hammel 41 densitome try DTA 1961 Ayres and Bens 42 DTA l9Gl Garn and Kessler 43 DTA 1962 Wendlandt 44 Group I I- Volumetric Barometric DTA 1953 Teitelbauin and Berg 45 DTA 1957 Gordon and Campbell 46 DTA 1964 Charles 47 TG 1965 Bancroft and Gesser 48 DTA 1967 Guenot et al. 49 TG 1968 Maicock and Pai Verneker 50 DTA 1969 Eousquet et al. 51 An example of the use of these techniques of examination is given in Fig. 13. The DTA curve of barium perchlorate hydrate is shown, and the interpretation of this curve alone would be difficult, but if the EGD curve is recorded simultaneously, as shown, then the common interpretation of the two curves permits us without risk of major error to state the following.The first three maxima on the EGD curve indicate the release of water, while the fourth is due to the escape of oxygen. The first maximum on the DTA curve can probably be explained by the melting of the hydrate, the fifth and sixth maxima by changes in crystal modification, and the seventh may be due to the melting of the anhydride. With the help of gas volumetric and gas barometric examinations (Table 111, Group 11) the amount of gas evolved can also be determined, but obtaining these curves is only an indirect means for the determination of the quality of the evolved gases. In addition to recording the curves of the gaseous decomposition products, it is also necessary to record the TG curve.Owing to the difference in the relative molecular masses of gases, in a f avourable instance we shall be able to select from possible gaseous decomposition products the most probable one. Here the EGD and TG curves of a calcium oxalate sample can be seen. The former indicates that in the three periods of the decomposition the volume of evolved gases was equal, but owing to the difference in the relative molecular masses of water, carbon monoxide and carbon dioxide, the sizes of the three steps in the TG curve are different. In more compli- cated instances this numerical difference forms the basis of calculations. The essential difference between the methods of EGD and EGA lies in the fact that with EGA both the quality and the amount of the gaseous decomposition products can be determined in a selective way.However, in addition to problems of standardisation, the The principle of this examination is demonstrated for a simple example in Fig. 14.PAULIK AND PAULIK: SIMULTANEOUS TECHNIQUES Analyst, VoZ. 103 Melting v Melting I Dehydration Decomposition I 1 1 I I 100 300 500 Temperature/"C 0 100 - E 2 200 - 5 2 300 400 0 200 400 600 800 1000 T e m p era t u r el * C Fig. 14. Simultaneous TG and EGD Fig. 13. Simultaneous DTA (1) and examinations Of CaC204.xH20. EGD (2) examinations of Ba(ClO,),.xH,O. separation of overlapping transformations, or in other words increasing the selectivity of the examination, causes the greatest trouble to thermoanalysts. We have already shown, for the determination of boehmite, alunite and kaolinite, the usefulness of EGA methods in separating overlapping reactions.A similarly convincing example is shown in Fig. 15. During the preparation of aluminium sulphate, if the optimum I I I 200 400 0 8 20 vi B - ln .s 40 60 '/3 H 2 S 04.24 H 2 0 \ -9 21 H2 0 I I I 200 400 Tern pera t u re/' C Fig. 15. Simultaneous DTA, TG, DTG and EGA examinations of (a), A1,(S04),.xH,0 of unknown composition and (b), A1,(SO4),.~H,SO4.24H2O.May, 1978 I N THERMAL ANALYSIS. REVIEW 429 conditions are not observed, then instead of Al,(SO,) ,.18H,O the acid salt is precipitated. When this salt is dehydrated sulphur(V1) oxide is also evolved, In fact, the course of the TG, DTG and DTA curves would not call our attention to this possibility.Further, even an EGD curve would not indicate the water and sulphur(V1) oxide separately. In contrast, the curve denoted SO, in Fig. 15 clearly shows that with the departure of the last two water molecules at 450 "C two thirds of a molecule of sulphur(V1) oxide is also released. Without performing the simultaneous EGA examination, we were not able even to notice the evolution of sulphur(V1) oxide. At the same time, having obtained the curve for the release of sulphur(V1) oxide we not only can determine the amount of sulphur(V1) oxide but may also be prevented from drawing incorrect conclusions regarding the composition of the sample and the kinetics of the process. The sulphur(V1) oxide curve was obtained by applying the technique of thermal gas titrimetry (TGT) (Table IV).With the help of this technique, based on the principle of titrimetric analysis, the amounts of the gaseous decomposition products of inorganic com- pounds can be determined in the presence of one another. TABLE IV EGA METHODS Method combined Method with Group I- TGT .. .. .. .. , . DTGT, DTA, Gas infrared spectroscopy . . TG Absorption liquid conductimetry TG, DTA Absorption liquid thermometry . . DTA DTA TG, DTG Group II- GC . . .. .. .. , . TG DTA TG DTA DTA DTA MS .. .. .. .. .. EGD EGD, DTA EGD, DTA EGD, TG TG EGD, TG, DTA EGD, DTA, TG DTA EGD, DTA, TG, DTG TG DTA TG, GC DTA, TG GC Thermoparticulate analysis . . DTA, TG Thin-layer chromatography .. DTA Date 1966 1967 1964 1963 1963 1963 1966 1968 1969 1975 1975 1965 1965 1966 1968 1969 1969 1969 1969 1969 1969 1969 1971 1973 1974 1960 1967 Workers Reference Paulik et al.52, 17 Keattch 54 Chamberlain and Green 55 Notz and Jaffe 56 Hegediis and Kiss 53 Can0 57 Bandi et al. 58 Chiu 59 Bollin 60 Yamada et al. 61 Mercier 62 Wendlandt and Southern 63 Langer et al. 64 Wendlandt et al. 65 Zitomer 66 Wilson and Hamaker 67 Smith and Johnson 68 Wiedemann 69 Gaulin et al. 70 Redfern et al. 71 Brown et al. 72 Langer and Bradly 73 Chang and Mead 74 Gibson 75 Merritt et al. 76 Doyle 77 Rogers 78 A similar possibility for the selective determination of gaseous decomposition products is offered by infrared spectroscopy and by conductimetric or thermometric measurements on a liquid in which the gases have been absorbed (Table IV, Group I). Examination of Decomposition of Organic Compounds It is known that the decomposition of organic compounds is in most instances a very complex process. Solid, liquid and gaseous decomposition products are formed in the course of the countless reactions taking place in parallel or consecutively.For the selective investigation of these products and processes the methods discussed so far are not suitable. However, these processes can be examined if, for example, gas-chromatographic (GC), mass430 PAULIK AND PAULIK : SIMULTANEOUS TECHNIQUES Analyst, Vd. 103 spectrometric (MS), thin-layer chromatographic or thernioparticulate analyses are combined with DTA and TG measurements. The earliest of these methods are listed in Table IV, Group 11. Particularly good selectivity can be attained when GC and MS are applied together.In this instance the gaseous decomposition products are separated by means of a gas chromatograph into different fractions, which are then further examined by means of a. mass spectrometer. As these examinations can be performed only stepwise, generally the EGD curve is simultaneously recorded in order to determine the gas-evolution process. The equipment required for these analyses is expensive and requires skill in handling; however, it is indispensable when the kinetics and mechanism of the thermal decomposition of organic compounds are to be studied. TD and ETA We think TD has been neglected up to now, yet it provides useful information about changes occurring in the crystal structure of inorganic compounds, which is an area in which other thermoanalytical methods are not very useful. Two examples of the use of TD are shown in Fig.16. It is known that kaolinite loses its water of constitution between 400 and 800 "C. During the course of a solid-state reaction it transforms first into metakaolinite and then into mullite at about 950 "C. The other example, barium chloride dihydrate, first loses its water of crystallisation and in the temperature interval 350-850 "C a significant recrystallisation process takes place in the anhydrous material. In the vicinity of 900 "C, just before the substance melts, a modification froin the a to p crystal form occurs. I IV I C .- I I w .- 0" TD I a 200 600 1000 0 1 2 E- 3 remperatu re/O C I 0 2 8 4 - 6 8 200 600 1000 Fig. 16. Simultaneous DTA, TG and TD examinations of (a), A120,.2Si0,.2H,0 and ( b ) , BaC1,.2H20. As the shape of the TD curves in the figures proves, the course of all the processes, the solid-state reaction and also the thermal decomposition, the recrystallisation and crystal modification, can be followed by this method.However, so far only in the field of coal, ceramic and silicate chemistry has TD found wide application, and even here mostly as a single method. The reason for the neglect of this technique probably lies in the difficult interpretation of the TD curve, as in general it undoubtedly gives a more complicated pictureNay, 1978 IN THERMAL ANALYSIS. REVIEW 431 than the TG or DTA curves. For example, in thermal decomposition the sample decomposes and an amorphous or microcrystalline phase is formed.The recrystallisation that generally follows is in most instances protracted and overlaps the previous process. While the DTA and TG curves indicate only the first process, the TD curve depicts both processes, which causes difficulties in its interpretation. Therefore, in addition to the derivation of this curve, the simultaneous recording of other thermoanalytical curves, e.g., DTA or DTG (Table V), can greatly contribute to the easier and more reliable evaluation of the TD curve. TABLE V TD METHODS Method combined with Date Workers Reference DTA .. .. . . 1956 Lehman and Gatzke 79 DTA .. . . . . 1958 Koehler 80 DTA .. . . . . 1959 Pearce and Mardon 81 DTD, DTA . . 1961 Paulik et al. 11 DTD, DTA, TG, DTG . . 1961 Paulik et al. 82, 11 DTD, DTA, TG, DTG, TGT,DTGT .. .. 1971 Paulik and Paulik 83 Emanation thermal analysis (ETA) is a method cognate with the previous one (Table 1'1). With the help of this method all those chemical and physical processes can be followed which actually cause the transitional migration of the lattice elements in crystals. However, the results of the two kinds of examination differ from and therefore supplement one another. TABLE VI ETA METHODS Reference Method combined with Date Workers TG . . .. . . .. 1961 Bussiere et al. 84 DTA, TD .. .. .. 1965 Zaborenko et al. 85 DTA, TD .. .. . . 1969 Balek 86 DTA, EGD ., .. 1972 Habersberger and Balek 87 DTA, TG .. .. .. 1973 Emmerich and Balek 88 EC Methods An example of the applicability of EC methods coupled with other thermoanalytical methods (Table VII) can be seen in Fig.17. This figure shows the results obtained with a thallium stearate sample. According to the DTA curve the substance melts immediately after its second polymorphous transition, but it forms an isotropic liquid only after the mesomorph - liquid transition. Now, as the course of the EC curve proves, these last two Method combined with DTA .. .. . . DTA .. .. .. DTA . . .. .. DTA .. .. .. X-ray .. .. TG, DTG, DTA . . .. DTA, TD, ETA . . .. DSC .. .. . . DTA .. .. .. TD .. .. .. DTA .. , . .. DTA, EGD .. .. TG .. .. .. TABLE VII EC METHODS Date Workers Reference 1959 1960 1960 1963 1965 1967 1969 1970 1970 1970 1973 1975 1975 Satava 89 Budnikov et al. 91 Pannetier et al. 92 Bessonov and Ustyantsev 93 Chiu 94 Carrol and Mangravite 95 David 96 Judd and Pope 97 Balek 98 Halmos and Wendlandt 99 Berg and Shlyapkina 100 Juranic et al.101 Berg and Burmistrova 90432 PAULIK AND PAULIK : SIMULTANEOUS TECHNIQUES AfiaZyst, vol. 103 processes can also be followed by measurement of electrical conductivity. This example is a simple one, but the information-multiplying effect of the method has its real significance in more complicated instances. Also, remarkable results were obtained recently by thermoanalysts who coupled the methods of thermal X-ray analysis, photometric thermal analysis, hot-stage microscopy, thermomagnetic analysis, dynamic reflectance spectroscopy and thermoacoustic analysis with the methods of TG, DTA, EC, MS, EGD and TD (Table VIII). h a 1 v) E 5 0.5 --. V I.’ u 3 -0 0 0 0 j Mesomorph transitions \ Isotropic liquid Pol ymorph transitions / I I I I Iso- I tropic I I Mesophase , liquid I Solid 0 50 100 150 Temperature/”C Fig.17. Simultaneous DTA and EC examinations of thallium stearate.102 Manipulation of Experimental Conditions The picture we intend to give would not be complete without consideration of the manipulation of experimental conditions. For example, the beneficial effect exerted upon the resolution or the selectivity if a vacuum or high pressure are applied is well known. Further, semi-micro techniques can be used to improve results, and many other similar attempts have been made to achieve the same purpose by the selection of experimental conditions. An example was shown above (Fig. 10) to demonstrate that in the course of decomposition, the concentration of the gaseous decomposition products within the sample is continuously changing in an uncontrollable way.This condition greatly influences the whole course of the decomposition. By altering the experimental conditions the concentration of the gaseous decomposition products, and therefore the course of the decomposition, changes too (Fig. 12). Some thermoanalysts wished to eliminate this effect by ensuring a “self-generated” atmosphere. These researchers wanted to solve this problem by use of sample holders whose shape, as shown in Table IX, makes it possible for the gaseous decomposition products However, we shall mention only two subjects here.May, 1978 IN THERMAL ANALYSIS. REVIEW TABLE VIII 433 THERMAL X-RAY ANALYSIS Method combined Method with Thermal X-ray analysis DTA DTA DTA, EC DTA EGD, MS Photometric thermal analysis DTA TG DTA TD Hot-stage microscopy DTA DTA DTA Thermomagnetic analysis TG Dynamic reflectance spectroscopy EGD Thermoacoustic analysis DTA Date 1965 1967 1967 1968 1973 1972 1972 1972 1974 1966 1967 1968 1966 1970 1975 Workers Wefers Ravich Bessonov et al.Barret et al. Wiedemann David Loehr and Levy Barrall and Johnson Chrony Miller and Sommer Dichtl and J eglitsch Van Tets and Wiedemann Simmons and Wendlandt Wendlandt and Bradley Chatterj ee Reference 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 to be released easily, while the diffusion of air from the opposite direction is possibly hindered. They postulated that a pure self-generated atmosphere would immediately be created in the reaction space at the beginning of the decomposition.Thus, the stability of the partial pressure of the gaseous decomposition products would make the course of the decomposition unambiguous. Owing to the poor thermal conductivity of the sample and the rapid increase in temperature, a temperature drop occurs within the sample. Accordingly, the slow heat transfer is also responsible for the delayed transformation, which takes place in a broad temperature interval (Fig. 11). However, as it turned out later, this solved the problem only to a certain extent. TABLE IX SAMPLE HOLDERS FOR A SELF-GENERATED ATMOSPHERE n Piston Bal I-valve Cap i I lary Labyrinth Type* Date P 1960 BV 1960 C 1962 P 1962 BV 1962 BV 1963 C 1964 L 1971 Workers Garn and Kessler Forkel Amiel and Paulmier Claude1 Pannetier et al.Dormieux Lagier et al. Paulik and Paulik Reference 118 119 120 121 122 123 124 125, 126 * P, piston; BV, ball-valve; C, capillary; L, labyrinth. Researchers wanted to overcome this problem by working out a “quasi-static” heating programme (Table X). The idea was that if the temperature increase in the sample was controlled in such a way that the decomposition reaction took place a t a very low and constant rate, then the error caused by the slow heat transfer may be totally eliminated. One, by limiting the rate of mass change, dm/dT, produced quasi-isothermal heating conditions for TG and simultaneous EGA investigations. The other, thermoanalysis with a constant rate of decomposition Two kinds of heating control systems were developed.434 PAULIK AND PAULIK : SIMULTANEOUS TECHNIQUES Analyst, VoZ.103 TABLE X METHODS WITH QUASI-STATIC HEATING u p1 < ~2 < P3 < 1 atm p = 1 atm = const. PI, P2, P3 = const. Heating Conditions Method Pressure control Date Workers Reference Quasi-static heating TG - dm/dT 1962 Erdey et al. 127 TCRD Vacuum dp/dT 1969 Rouquerol 128 TG PI, pz, p , dm/dT 1971 Paulik and Paulilc 129 Self-generated atmosphere + TG 1 atm dnz/dT 1972 Paulik and Paulik 130 quasi-static heating TG, EGA 1 atm dm/dT 1973 Paulik and Paulik 131 (TCRD), produced a heating control system in vacuum, under which the pressure of the gaseous products became constant, and the correlation between the constant gas flow being evolved and the temperature was recorded.However, the final solution to the problem has been brought about by the simultaneous application of a quasi-static heating programme and a sample holder ensuring a self- generated atmosphere. The gradual progress of this technique and the beneficial effect of its application are shown by the TG curves in Fig. 18. The curves demonstrate the decomposition of calcium carbonate under different experimental conditions. Curves 1 and 2 were recorded a t a heating rate of 10 "C min-l. For curve 1 a conventional open crucible 0 10 \O 0- v; 20 - v) 30 5 40 600 700 800 900 1000 Tern perat u re/ O C Fig. 18. Decomposition of CaCO, investigated under (1 and 2) a dynamic and (3 and 4) a quasi-static heating programme. Sample holder: 1 and 3, conventional; and 2 and 4, labyrinth.was used, while for curve 2 a labyrinth sample holder was used in order to ensure a self- generated atmosphere. Curves 3 and 4 were similarly recorded by using the two different kinds of sample holder, but with a quasi-static heating programme, and the calcium carbonate decomposed under quasi-isothermal conditions, i . ~ . , the temperature of the sample did not change during the decomposition process. It is evident that quasi-isothermal conditions. greatly increase the selectivity and resolution of the examination, as well as the accuracy of the qualitative and quantitative determinations. However, there is a difference even between curves 3 and 4. According to curve 4 the decomposition of calcium carbonate took place at 895 "C, i.e., at its "normal" decoinposition temperature, known from physical chemistry.This situation is of significance from the point of view of standardisation.May, 1978 I N THERMAL ANALYSIS. REVIEW 435 Conclusion In the presentation of the tables our primary intention was to summarise the simultaneous methods that have been reported in a manner that can be easily read. It must be appreciated that a complete survey of the literature is not given here, and the material presented has necessarily been selective. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. ‘33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. References Dejean, F., Revue Me’tall., 1905, 2, 70.de Iceyser, W. L., Nature, L o n d . , 1953, 172, 364. Paulik, F., Paulik, J., and Erdey, L., Hung. Pat., 144 54811954. Erdey, L., Paulik, F., and Paulik, J., Acta Chim. Hung., 1956, 10, 61. Paulik, J., and Paulik, F., Hung. Pat., 143 33211955. Paulik, F., Paulik, J., and Erdey, L., Talanta, 1966, 13, 1405. Lambert, H., Bull. SOC. Fr. Ce’Yavlz., 1955, 28, 23. Waters, P. L., Nature, Lond., 1956, 178, 324. Campbell, C., Gordon, S., and Smith, C. L., Analyt. Chew., 1959, 31, 1188. Freeman, E. S., and Edelman, D., Analyt. Chew., 1959, 31, 624. Paulik, F., Paulik, J., and Erdey, L., Hung. Pat., 150 905/1961. Paulik, F., Paulik, J., and Erdey, L., Hzdng. Pat., 150 906/1961. Wilburn, F. W., and Hesford, J. R., J . Scient. Instrum., 1963, 48, 91. Knofel, G.F., Spreclasaal Keram. Glas Email, 1963, 96, 58. Garn, P. D., “Thermoanalytical Methods of Investigation,” Academic Press, New York, 1965, Rupert, G. N., Rev. Scient. Instvum., 1965, 36, 1629. Paulik, J.. Paulik, F., and Erdey, L., Hung. Pat., 154 372/1966. Paulik, J., and Paulik, F., Talanta, 1970, 17, 1224. PannCtier, G., Revue Chim., Buc., 19G6, 17, 592. Byrd, J . S., United States Atomic Energy Commission, 1969, DP 1211. Price, G. H., J . Phys. E , 1972, 5, 747. Forster, I . B., Lee, J . A., and Tye, F. L., Thevunochim. Acta, 1974, 9, 55. Moskalewicz, R., in BuzAs, f., Editov, “Proceedings of the Fourth International Conference on Paulik, F., Paulik, J., and Erdey, L., Hung. Pat., 145 369/1955. Paulik, F., Paulik, J., and Erdey, L., 2. Analyt. Chem., 1958, 160, 241.Po-cvell, D. A, J , Scient. Instrum., 1957, 34, 225. Papailhau, J., Bull. Soc. Fr. ilfinc’v. CristaLlog./., 1959, 82, 367. Blazelr, A, and Cisar, V., Silikdty, 1960, 4, 52. Reismann, A., Analyt. Chem., 1960, 32, 1566. Piece, R., Schweiz. Miner. Petrogr. Mitt., 1961, 41, 303. Torkar, I<., Lasser, K., and Fritzer, H. P., Sprechsaal Keram. Glas Email, 1962, 95, 212. Formanek, J., and Dykast, J., Silikdty, 1962, 6, 113. Kissinger, EI. E., and Newman, S. B., i n Wine, G. M., Editov, “High Polymers, Analytical Chemistry McAdie, H. G., Analyt. Chem., 1963, 35, 1840. Kriigcr, J. E., and Bryden, J . G., J . Scicnt. Instvu.l.tz., 1963, 40, 178. Wiedemann, €1. G., Chemie-Ingr-Tech., 1964, 36, 1105. Khristianov, A. S., and Korovyatnikov, G. F., Zav. Lab., 1964, 30, 495.Saito, H., Otsuka, R., Iwata, S., and Tsuchiomoto, K., Bull. Sci. Engng Res. Lab. Waseda Univ., Charsley, E. L., and Redfern, J . P., paper presented a t the International Symposium on Thermal Patai, S., Helpern, Y., Estcrman, L., and Weinstein, ill., Israel J . Chew., 1968, 6, 445. Lodding, W., and Hammell, L., Analyt. Chem., 1960, 32, 657. Ayres, W. M., and Bens, E. M., Analyt. Chem., 1961, 33, 568. Garn, P. D., and Kessler, J . E., Analyt. Chein., 1961, 33, 952. Wendlandt, W. W., Agcalytica Chim. Acta, 1962, 27, 309. Teitelbaum, B. V., and Berg, L. G., Zh. Analit. Khim., 1953, 8, 152. Gordon, S., and Campbell, C., Analyt. Chem., 1957, 29, 1706. Charles, R. G., J , Inorg. Nucl. Chem., 1964, 26, 2195. Bancroft, G. M., and Gesser, H. B., J . Inorg. Nucl.Chew., 1965, 27, 1537. Guenot, J., Vallicon, J. L., and Pannetier, G., Bull. Soc. Chim. FY., 1967, 3068. TvIaycock, J. N., and Pai Verneker, V. R., Analyt. Chew., 1968, 40, 1935. Bousquet, J., Blanchard, J. M., Bonnetot, B., and Claudy, P., Bull. Soc. Chim. Fr., 1969, 184. Paulik, J., Paulik, F., and Erdey, L., Mikrochim. Acta, 1966, 886. Hegedus, A. J., and Kiss, B. A,, Magy. Ke‘m. Foly., 1967, 73, 41. Keattch, C. J., “Analysis of Calcareous Materials,” Monograph No. 18, Society of Chemical Industry, Chamberlain, M. M., and Green, A. F., J . Inorg. Nucl. Chew., 1963, 25, 1471. Notz, K. J., and Jaffe, H. H., J. Inorg. Nucl. Chern., 1963, 25, 851. p. 499. Thermal Analysis, Budapest,” Volume 3, AkadCmiai Kiad6, Budapest, 1975, p. 873. of Polymers, ’’ Second Edition, Interscience, New York, 1962, p. 159.1964, 27, 26. Analysis, London, 1965. London, 1964, p. 279.436 PAULIK AND PAULIK : SIMULTANEOUS TECHNIQUES Analyst, V d . 103 Cano, G., Bull. SOC. Chim. Fr., 1963, 2540. Bandi, W. R., Straub, W. A., Buyok, E. G., and Melnick, L. M., Analyt. Chem., 1966, 38, 1336. Chiu, J., Analyt. Chem., 1968, 40, 1516. Bollin, E. M., in Schwenker, R. F., Jr., and Garn, P. D., Editors, “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume I, Academic Press, New York, 1969, p. 255. Yamada, K., Oura, S., and Haruki, T., in BuzAs, f., Editor, “Proceedings of the Fourth International Conference on Thermal Analysis, Budapest ,” Volume 3, Akad6miai Kiad6, Budapest, 1975, p.1029. Mercier, J . G., in BuzAs, f., Editor, “Proceedings of the Fourth International Conference on Thermal Analysis, Budapest,” Volume 3, Akademiai Kiad6, Budapest, 1975, p. 104. Wendlandt, W. W., and Southern, T. M., Anulytica Chim. Acta, 1965, 32, 405. Langer, H. G., Gohlke, R. S., and Smith, D. H., Analyt. Chem., 1965, 37, 433. Wendlandt, W. W., Southern, T. M., and Williams, J. I<., Analytica Chim. Ada, 1966, 35, 251. Zitomer, F., Analyt. Chem., 1968, 40, 1091. Wilson, D. E., and Hamaker, F. AT., in Schwenker, R. F., Jr., and Garn, P. D., “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 1, Academic Press, New York, 1969, p. 517. Smith, J. W., and Johnson, D. R., in Schwenker, R.F., Jr., and Garn, P. D., Editors, “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 2, Academic Press, New York, 1969, p. 1251. Wiedemann, H. G., in Schwenker, R. F., Jr., and Garn, P. D., Editors, “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 1, Academic Press, New York, 1969, p. 229. Gaulin, C. A., Wachi, F. M., and Johnston, T. H., in Schwenker, R. F., Jr., and Garn, P. D., Editors, “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 2, Academic Press, New York, 1969, p. 1453. Redfern, J. P., Treherne, B. L., Aspimal, M. L., and Wolstenholme, W.A., paper presented at the 17th Conference on Mass Spectrometry, Dallas, Tex., USA, 1969. Brown, J . G., Dollimore, J.. Freedman, C. M., and Harrison, B. H., Thermochim. Acta, 1970, 1, 499. Langer, H. G., and Bradly, T. P., in Schwenker, R. I?., Jr., and Garn, P. D., Editors, “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 1, Academic Press, New York, 1969, p. 295. Chang. T. L.. and Mead, T. E., Analyt. Chem., 1971, 43, 534. Gibson, E. K., Thermochim. Acta, 1973, 5, 243. Merritt, C., Sacher, R. E., and Petersen, B. A., J . Chromat., 1974, 99, 301. Doyle, C. D., “Evaluation of Experimental Polymers,” WADD Technical Report 60-283, May, Rogers, R. N., Analyt. Chem., 1967, 39, 730.Lehman, H., and Gatzke, H., Tonind.-Ztg Keram. Rdsch., 1956, 80, 7. Koehler, E., Bull. SOC. Fr. Ckram., 1958, 38, 3. Pearce, J. H., and Mardon, P. G., J . Scient. Instrum., 1959, 36, 457. Paulik, F., Paulik, J., and Erdey, L., Mikrochzm. Acta, 1966, 894. Paulik, F., and Paulik, J., Thermochim. Acta, 1971, 3, 13. Bussiere, P., Claudel, B., Renouf, J. P., Trambouze, Y., and Prettre, M., J . Chim. Phys.. 1961, 58, Zaborenko, K. B., Melichov, L. L., and Portjanoi, V. A., Radiokhimiya, 1965, 7, 315. Balek, V., J . Mater. Sci., 1969, 4, 919. Habersberger, K., and Balek, V., Thermochim. Acta, 1972, 4, 457. Emmerich, W. D., and Balek, V., High Temp. High Pressures, 1973, 5, 67. Satava, V., Colln Czech. Chem. Commun. Engl. Edn, 1959, 24, 3297. Berg, L. G., and Burmistrova, N.P., Zh. Neorg. Khim., 1960, 5, 676. Budnikov, P. P., Gorshkov, V. S., and Titoskaya, V. T., Stroit. Mater., 1960, 6 (12), 30. Pannetier, G., Michel, A., Bergeault, J. M., and Djega-Maridassou, G., Bull. Soc. Chim. Fr., 1963, Bessonov, A. F., and Ustyantsev, V. M., Zav. Lab., 1965, 31, 620. Chiu, J., Analyt. Chem., 1967, 39, 861. Carrol, R. W., and Mangravite, R. V., in Schwenker, R. F., Jr., and Garn, P. D., Editors, “Proceedings of the Second International Conference on Thermal Analysis, Worcester, Mass. , USA, August 18-23, 1968,” Volume 1, Academic Press, New York, 1969, p. 189. David, D. J., Thermochim. Acta, 1970, 1, 277. Judd, M. D., and Pope, M. I., J . App2. Chem., Lond., 1970, 20, 380. Balek, V., Analyt. Chew., 1970, 42, 16A. Halmos, Z., and Wendlandt, W.W., Thermochim. Acta, 1973, 7, 95. Berg, L. G., and Shlyapkina, E. N., J . Thermal Analysis, 1975, 8, 417. Juranic, N., Karaulik, D., and Vucelic, D., J . Thermal Analysis, 1975, 7, 119. Halmos, Z., Seybold, K., and Meisel, T., in BuzAs, f., Editor, “Proceedings of the Fourth Inter- national Conference on Thermal Analysis, Budapest,” Volume 2, Akaddmiai Kiad6, Budapest, 1975, p. 429. Wefers. K.. Ber. Dt. Keram. Ges.. 1965, 42, 35. Ravich, G. B., Kododyazhnyi, V. Z., Brodov, V. G., Kuchumov, A. M., and Zhemarkin, A. I., Zh. Neorg. Khim., 1967, 12, 2256. Bessonov, A. F., Ustyantsev, V. M., and Taksis, G. A., Poroshkovaja Metallurgya, 1967, 7, 92. 1960. 668. 1204. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105.May, 1978 I N THERMAL ANALYSIS. REVIEW 437 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. Barret, P., Gerard, N., and Watelle-Marion, G., Bull. SOC. Chim. Fr., 1968, 3172. Wiedemann, H. G., Thermochim Acta, 1973, 7, 131. David, D. J., Thermochim. Acta, 1972, 3, 277. Loehr, A. A., and Levy, P. F., Am. Lab., 1972, 4, 11. Barrall, E. M., and Johnson, J . F., Thermochim. Acta, 1972, 5, 41. Chromy, S., Silikuty, 1974, 18, 105. Miller, R. P., and Sommer, G., J . Scient. Instrum., 1966, 43, 293. Dichtl, H. J., and Jeglitsch, F., Radex Rdsch., 1967, 716. Van Tets, A., and Wiedemann, H. G., in Schwenker, R. F., Jr., and Garn, P. D., Editors, “Pro- ceedings of The Second International Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 1, Academic Press, New York, 1969, p. 121. Simmons, E. L., and Wendlandt, W. W., Analytica Chim. Acta, 1966, 35, 461. Wendlandt, W. W., and Bradley, W. S., Thermochim. Acta, 1970, 1, 143. Chatterjee, P. K., i n BuzAs, f., Editor, “Proceedings of the Fourth International Conference on Garn, P. D., and Kessler, J. E., Analyt. Chew,., 1960, 32, 1563. Forkel, W., Naturwissenschaften, 1960, 47, 10. Amiel, J., and Paulmier, C., C.R. Hebd. Se‘anc. Acad. Sci., Paris, 1962, 225, 2443. Claudel, B., Thesis, University of Lyon, 1962. Pannetier, G., Bergeault, J. M., and Guenot, J., Bull. Soc. Chim. FY., 1962, 2158. Dormieux, J. L., C.R. Hebd. Se‘anc. Acad. Sci., Paris, 1963, 259, 579. Lagier, J. P., Quahes, R., and Paulmier, C., Bull. SOC. Chim. Fr., 1964, 5, 1082. Paulik, F., and Paulik, J., Hung. Pat., 163 305/1971. Paulik, F., and Paulik, J., Analytica Chim. Acta, 1972, 60, 127. Erdey, L., Paulik. F., and Paulik, J., Hung. Pat., 152 197/1962. Rouquerol, J., in Schwenker, R. F., Jr., and Garn, P. D., Editors, “Proceedings of the Second Inter- national Conference on Thermal Analysis, Worcester, Mass., USA, August 18-23, 1968,” Volume 1, Academic Press, New York, 1969, p. 281. Thermal Analysis, Budapest,” Volume 3, Akaddmiai Kiadb, Budapest, 1975, p. 835. Paulik, J., and Paulik, F., Analytica Chim. Acta, 1971, 56, 328. Paulik, F., and Paulik, J., Analytica Chim. Acta, 1972, 60, 127. Paulik, F., and Paulik, J., Analytica Chim. Acta, 1973, 67, 437. Received October 24th, 1977 Accepted October 31st, 1977