37 2.1 Theoretical Considerations Conventional thermal analysis techniques include differential scanning calorimetry, differential thermal analysis (DTA), thermogravimetric analysis (TGA), dynamic mechanical analysis, thermomechanical analysis, microthermal analysis and dielectric thermal analysis. Thermal analysis of a material can be either destructive or non- destructive, but in almost all cases subtle and dramatic changes accompany the introduction of thermal energy. Thermal analysis can offer advantages over other analytical techniques including: variability with respect to application of thermal energy (step-wise, cyclic, continuous, and so on), small sample size, the material can be in any ‘solid’ form – gel, liquid, glass, solid, ease of variability and control of sample preparation, ease and variability of atmosphere, it is relatively rapid, and instrumentation is moderately priced. Most often, thermal analysis data are used in conjunction with results from other techniques such as those based on pyrolysis, infrared spectroscopy and nuclear magnetic resonance (NMR) spectroscopy. Cebe [1, 2] has reviewed current trends in the thermal testing of polymers. Apparatus for TGA is discussed in Appendix 1. TGA involves continuous weighing of a polymer as it is subjected to a temperature programme of temperatures up to 1000 °C. This technique can provide quantitative information about the kinetics of the thermal decomposition of polymeric materials from which the thermal stability can be evaluated. It is used to study the influence of factors such as effect of crystallinity, molecular weight, orientation, tacticity, substitution of hydrogen atoms, grafting, copolymerisation, and addition of stabilisers on polymer degradation. Figure 2.1 shows decomposition profiles for polytetrafluoroethylene (PTFE) and fibre glass-reinforced Nylon. The lifetime or shelf life of a polymer can be estimated from the kinetic data. Ozawa [3] observed that the activation energy of a thermal event could be determined from a series of thermogravimetric runs performed at different heating rates [4–8]. As the heating rate increased, the thermogravimetric changes occurred at higher temperatures. The measurement of lifetime is discussed in more detail in Section 2.2.8. 2 Thermogravimetric Analysis
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
37
2.1 Theoretical Considerations
Conventional thermal analysis techniques include differential scanning calorimetry, differential thermal analysis (DTA), thermogravimetric analysis (TGA), dynamic mechanical analysis, thermomechanical analysis, microthermal analysis and dielectric thermal analysis. Thermal analysis of a material can be either destructive or non-destructive, but in almost all cases subtle and dramatic changes accompany the introduction of thermal energy. Thermal analysis can offer advantages over other analytical techniques including: variability with respect to application of thermal energy (step-wise, cyclic, continuous, and so on), small sample size, the material can be in any ‘solid’ form – gel, liquid, glass, solid, ease of variability and control of sample preparation, ease and variability of atmosphere, it is relatively rapid, and instrumentation is moderately priced. Most often, thermal analysis data are used in conjunction with results from other techniques such as those based on pyrolysis, infrared spectroscopy and nuclear magnetic resonance (NMR) spectroscopy.
Cebe [1, 2] has reviewed current trends in the thermal testing of polymers. Apparatus for TGA is discussed in Appendix 1.
TGA involves continuous weighing of a polymer as it is subjected to a temperature programme of temperatures up to 1000 °C. This technique can provide quantitative information about the kinetics of the thermal decomposition of polymeric materials from which the thermal stability can be evaluated. It is used to study the influence of factors such as effect of crystallinity, molecular weight, orientation, tacticity, substitution of hydrogen atoms, grafting, copolymerisation, and addition of stabilisers on polymer degradation. Figure 2.1 shows decomposition profiles for polytetrafluoroethylene (PTFE) and fibre glass-reinforced Nylon.
The lifetime or shelf life of a polymer can be estimated from the kinetic data. Ozawa [3] observed that the activation energy of a thermal event could be determined from a series of thermogravimetric runs performed at different heating rates [4–8]. As the heating rate increased, the thermogravimetric changes occurred at higher temperatures. The measurement of lifetime is discussed in more detail in Section 2.2.8.
Figure 2.1 TGA decomposition profiles for (a) PTFE and (b) fibre glass reinforced Nylon. Reproduced with permission from T. Ozawa, Bulletin of Chemical Society
• Thermal stability including the effect of various factors such as crystallinity,molecular weight, orientation and tacticity, substitution of hydrogen atoms, grafting, co-polymerisation and effect of additives
• Polymerdegradationordecomposition
• Determinationofactivationenergy(E)
• Polymertransitionstudies
• Effectofantioxidantsonageing
• Polymerisationkinetics
• Combustioninhibitionstudies
• Polymerlifedetermination
• Effectofcatalysts
• Copolymercomposition
• Additivedegradationstudies
SomeexamplesofthesearediscussedinSections 2.2.1 to 2.2.8.
2.2.1 Thermal Stability Studies
TGA has been used to study degradation kinetics and various factors affecting thermal stability of polymers, such as crystallinity, molecular weight, orientation, tacticity, substitution of hydrogen atoms, grafting, co-polymerisation, addition of stabilisers, and so on.
Libiszowski and co-workers [9] used TGA and thermal volatilisation analysis to investigate the thermal stability of α-ester ω-hydroxy terminated poly-L-lactides(PLA) i.e., [C4H9OC(O)-PLA-OH] and its esterified counterpart [C4H9OC(O)-PLA-OC(O)CH(C2H5)C4H9]. Thermogravimetric measurements versus number average
40
Thermal Methods of Polymer Analysis
molecular weight (Mn) reveal, as a result of Mn extrapolation, theupper limitofthermal resistance of PLA. The onset temperature of the thermal degradation and thetemperatureofmaximumdegradationrateofthetwoPLAwereequalto337°Cand 376 °C, respectively.
Similarly,intheirTGAcurvesofpolyhydroxyamideanditsderivativesZhangandco-workers[10]showedthatinthedecompositionofpolyhydroxyamideanditsderivativedecomposition occurs in two distinct stages. For one derivative the weight loss in the first stage (200–500 °C) was about 11%, which agrees very well with the calculated value(11.5%)ifweassumethatpolybenzoxazoleiscyclisedintopolyhydroxyamideby loss of water. The second stage (580–820 °C) is associated with a further 30% weight loss then levels off, producing a high char yield of 56%.
The introduction of bromine and trifluoromethyl groups into these polymers in an attempt to increase their flame retardency does not affect the onset decomposition temperature of the first stage (around 260–290 °C), but it does greatly reduce the thermal stability of the second stage (around 500–510 °C), which suggests that the halogenatedpolybenzoxazoleislessstableinthenon-halogenatedpolybenzoxazole.
In their study of the characterisation of chloromethyl substituted polystyrene divinylbenzene (ST-DVB) networks, Nakagawa and co-workers [11] reported TGA weight loss curves for these copolymers, and it was found that the weight loss occurred stepwise. The first weight loss, occurring around 200 °C, increases as the degree of chloromethylation increases. The main degradation, occurring around 400 °C, shifts to a higher temperature, and the amounts of residue increase as the decrease of chloromethylation increases, even though the degree of divinylbenzene crosslinking, which primarily affects the thermal stability of the network, is essentially the same. These phenomena suggest that dehydrochlorination occurs around 200 °C, and the residue with a methylene crosslinked structure undergoes further degradation around 400 °C.
Thermal stability measurements have been carried out on numerous other polymers including polyethylene ethylene vinyl–alcohol copolymer [12], polyaniline [13], trans-stilbene–N–substituted maleimides [14], cellulose [15-20], polystyrene [14, 16],ethylene–styrenecopolymers[21,22],ST-DVB-basedionexchangers[23],vinylchloride–acrylonitrile copolymers [24], polyethylene terephthalate [25], polyesters suchaspolyisopropylene carboxylate [26], polyglycollate [27-29],Nylon6 [30],polypyromellitimides, poly-N-α-naphthylmaleimides [26, 31], polybenzo-bis(amino-imino pyrolenes) [32], polyvinyl chloride [33-35], acrylamide–acrylate copolymers andpolyacrylicanhydride[36-38],polyamides[39],amine-basedpolybenzo-oxazines[40], polyester hydrazides [41], poly–α–methyl styrene tricarbonyl chromium [42], polytetrahydrofuran[43],polyhexylisocyanate[44],polyurethanes[45],ethylene–
TGA,underbothisothermalanddynamicconditions,isusedextensivelyforstudyingthe kinetics of decomposition reactions [54-56]. Certainly, determination of the kinetic parameters of polymer degradation reactions under isothermal conditions yields more precise and accurate results, however, this is a labour-intensive method, needing much time and many samples. In this connection dynamic thermogravimetry has been widely employed for the last 15–20 years for the analysis of polymers and of polymer composites [57]. Despite several substantial disadvantages (lack of reproducibility, difficulty of control of temperature and rate of heating, sensitivity to thepresenceoflow-molecularmassadmixturesandtothethermalprehistoryofthesample, overlapping of certain stages of the process [54], and so on) the method makes it possible not only to obtain the quantitative characteristics of the decomposition process (initial and final temperatures of the process, the degree of decomposition as a function of temperature and so on) but also to describe this process with reasonable accuracyintheformofkineticequationswhicharecalculatedfromexperimentaldata. A number of publications describe precise methods for the determination of kinetic parameters by mathematically processing the TGA curves [58].
Determination of the values of the kinetic parameters for simple one-stage processes are derived from the rate of mass change with time and are as described by the equation:
High resolution TGA has been applied to decomposition studies on PMMA, ethylene–vinyl acetate copolymer and acrylonitrile–butadiene–styrene terpolymers [59-62]. The results obtained on a supposedly pure sample of PMMA homopolymer indicated that a small quantity of impurity, possibly unreacted methyl monomer or even polyethylene methacrylate, is present. Conventional TGA does not resolve this impurity.
Thermogravimetry of the following rubber blends has been studied in detail by Lockmuller and co-workers [63] from the point of view of controlling decomposition mechanisms and minimisation of variance: chloroprene rubber blends, butadiene–acrylonitrile rubbers, and rubber adhesives.
IntheirTGAstudyofepoxysilicapolymers,Cardinoandco-workers[64]showedthatthepolymerformedbythereactionof3-glycidoxypropyltrimethoxysilanewith3-aminopropyltriethylsilane (2:1 ratio) had different structural properties compared to the properties of polymers obtained at other ratios, these included a higher degradation temperature and a longer decomposition time.
Das and co-workers [65] characterised polymethacrylate, high α-cellulose copolymers and showed that there were three major steps in the TGA curve which indicates a complexdecompositionpatternforthesepolymers.
Thermal decomposition studies have also been conducted on polyimides [66]. Li and co-workers [67] and Grishin and co-workers [68] used TGA to investigate the decomposition kinetics of polymers.
2.2.3 Complementary Pyrolysis Studies
Pyrolysis kinetics have been carried out on poly-L-lactone salts using TGA linked to various methods such as NMR spectroscopy, gas chromatography and pyrolysis–gas chromatography–mass spectrometry to identify volatile decomposition products. The effectoftheendstructuresonpyrolysiskineticswasexaminedandthemechanismsof pyrolytic degradation for both polymers identified.
Pemawonsa and co-workers [69] investigated the pyrolysis of polymethyl vinyl silyl ethylene diamine using TGA, NMR and infrared spectroscopy. The polymer decomposed to yield silicon carbonitride ceramic. Similarly, Ma and co-workers [70] used TGA, Raman spectroscopy, Fourier-Transform infrared spectroscopy, X-ray photoelectron spectroscopy, elemental analysis, scanning electron microscopy and transmission electron microscopy to study the changes in composition and structure ofpolymethylsilsequioxanepyrolysedatvarioustemperaturesunderflowingnitrogen.
43
Thermogravimetric Analysis
2.2.4 Activation Energy
MustataandBicu[71]performedactivationenergymeasurementsonepoxyresinsobtained from aniline o-cresol/formaldehyde and epichlorohydrin resins. Thermal studiesshowedthattheformaldehyderesinsandcuredpolymersexhibitedsimilarthermal behaviour. The activation energy of the thermal degradation was in the range 40 to 85 kJ/mol for the formaldehyde resins and in the range 40 to 45 kJ/mol for crosslinked polymers.
Dyakonov and co-workers [72] determined overall activation energies for the thermal degradationofaromaticaminecuredepoxyresinsystemsbasedonthediglycidylether of bisphenol.
TGA has also been used to determine activation energies of the dehydration of polyaniline [73].
2.2.5 Polymer Transitions
Hu and Xu [74] used TGA to measure the glass transition and onset of decomposition temperature of 3-alkoxy substituted polythiophene.The glass transition and theonset of decomposition temperature of the polymer decrease with the increase of the sidechainlengthandtheglasstransitionisdependentontheflexibilityofthesidechainintroduced.Theextentofconjugationofthesepolymersarehigherthanthatof poly(3-alkylthiophene).
2.2.6 Effect of Antioxidants on Polymer Ageing
The effect on the ageing of ethylene–vinyl acetate copolymer has been studied [46]. Mechanismsaresuggestedfortheformationofthevariousoxidativegroupsfound.Combinations of phenolic antioxidantswith phosphates give greatest protectionagainstdiscolouration,whichisconsistentwithaproposedfreeradicaloxidationprocess of deacetylation.
The kinetics of the process was followed using thermal analysis and gravimetry, and the molecular weights of the resultant polymers were determined by viscometry and gel permeationchromatography.Thesepolymersamplesexhibitmarkedpolydispersity.
44
Thermal Methods of Polymer Analysis
2.2.7 Polymer Lifetime Measurements
The service lifetime of a polymer at various temperatures can be estimated from kinetic data (Grishin and co-workers [68]). Ozawa [3] observed that the activation energy of a thermal event could be determined from a series of thermogravimetric runs performed at different heating rates. As the heating rate increased, the thermogravimetric change occurred at higher temperatures. A linear correlation was obtained by plotting the logarithm of the heating rate or scan speed against the reciprocal of the absolute temperature at the same conversion or weight loss percentage. The slope was directly proportional to the activation energy and known constants. To minimise errors in calculations,approximationswereusedtocalculatetheexponentialintegral[3-6].It was assumed that the initial thermogravimetric decomposition curve (2–20% conversion)obeyedfirstorderkinetics.Rateconstantsandpre-exponentialfactorscouldthenbecalculatedandusedtoexaminerelationshipsbetweentemperature,time and conversion levels. The thermogravimetric decomposition kinetics could be used to calculate:
Figure 2.2a shows thermograms (percentage weight versus temperature) for a 10 mg specimen of PTFE obtained at four different heating rates: 2.5, 5, 10 and 20 °C per minute in a dynamic air atmosphere. From these data can be calculated the rate of decomposition of PTFE, the activation energy and the relationship between the rate constant or half-life and temperature.
The results shown in Figure 2.2a agree with the theoretical prediction that, as the rate of heating increased, the thermograms are displaced to a higher temperature. Activation energies at selected percentage conversion levels were calculated using the results. The activation energy was calculated from the slope of the graph of scan time against the inverse of the absolute temperature (Figure 2.2b). After the activation energy had been determined, the rate constants, half-lives and percentage conversions could be calculated for certain temperatures. In Figure 2.2c the rate constant is plotted against temperature to provide information on the stability of the sample from ambient temperature to 800 °C. The half-life can be calculated from these kinetic data and a graph of half-life versus temperature plotted (Figure 2.2d).
Figure 2.2 (a) Thermogravimetric curves of PTFE; (b) Arrhenius plots for thermogravimetric analysis of PTFE; (c) rate constant versus temperature curves of PTFE; (d) half life temperatures curve of PTFE; and (e) conversion versus time at 500 °C curves of PTFE. Reproduced with permission from T. Ozawa, Bulletin of
Wang and co-workers [75] investigated the kinetics of the thermal degradationofmagnesiumhydroxide-based, halogen-free, retarded low-density polyethylenecomposites containing ethylene propylene (EP) copolymer elastomer. The thermal degradation of the composites was studied using thermogravimetric (TG) analysis and real-time Fourier transform infrared (RT–FT-IR) spectroscopy. The combustion residues from the composites were characterised by Raman spectroscopy and X-ray
47
Thermogravimetric Analysis
photoelectron spectroscopy. The results from TG and RT–FT-IR tests showed that the incorporation of a suitable amount of the elastomer into polyethylene/magnesium hydroxide composites after vulcanisation increased the thermal stability.Ramanspectroscopy showed that a graphite-like char was produced for composites with EP elastomer.
References
1. P. Cebe, M. Jaffe and C.E. Carraher, ACS Polymeric Materials Science and Engineering, 1998, 78, 96.
2. M.L. Cerrada, Revista Plasticos Modernos, 2002, 83, 551, 501.
3. T. Ozawa, Bulletin of the Chemical Society of Japan, 1965, 38, 11, 1881.
4. J.H.FlynnandL.A.Wall,Journal of Polymer Science Part B: Polymer Letters, 1966, 4, 5, 323.