AKTS AG
TechnoArk 3
3960 Siders
Switzerland


Customer service:
Phone:+41-848-800221
Fax:+41-848-800222
 

AKTS-Chemiluminescence fields of application

The problems of stability, quality, safety and guarantee purposes related to the oxidation reactions are important issues in the field of materials engineering, food, pharmaceuticals, plastics, coatings, paints, adhesives, high energetic materials, petrochemistry, conservation science, fundamental research etc. Here, the application of the advanced Chemiluminescence technique with an advanced kinetic description of the investigated processes leads to a great benefit in the characterization of long-term oxidation behavior even at relatively low temperatures.

Application of chemiluminescence and isoconversional kinetics for prediction of the lifetime of organic materials

(Collano AG, Berne University of the Arts and AKTS AG)

Abstract

Chemiluminescence (CL) includes all luminescence phenomena resulting from chemical reactions therefore due to the fact that the oxidative degradation process of the organics is combined with the emission of light this method can be applied for investigating the course of the oxidation processes.

Commonly applied methods of the investigation of the oxidation such as determination of oxidation induction time (OIT) or oxidation onset temperature (OOT) using Differential Scanning Calorimetry (DSC) or other conventional thermoanalytical methods are sometimes unsuitable for long-term prediction of oxidative behaviour because of the use of elevated temperatures during these experiments: the high temperatures may invoke reaction pathways which are different from those encountered under the conditions of use.

Compared to DSC and other conventional thermo-analytical methods, CL offers many advantages: due to its much higher sensitivity, experiments can be performed at much lower temperatures i.e. closer to the application related conditions. This fact is of the importance for the characterisation of substances with low temperature melting points, glass transitions, etc. The outstanding baseline stability of CL is of great benefit when performing long-term experiments; moreover, the CL-signal is related exclusively to the oxidation processes and therefore is not superposed by the signals resulting from the other reactions, including phase transitions.

In the present study we report on a new approach of kinetic analysis of the oxidation reactions of natural rubbers with and without stabiliser in an oxygen atmosphere at moderate temperatures using CL measurements carried out on a newly-developed instrumentation. The kinetic parameters of the oxidation process, calculated from the chemiluminescence’s signals by means of the differential isoconversional method of Friedman, were subsequently applied for the simulation of the rubber aging under different temperature profiles.

1. Introduction

It is known that most of the organic materials readily react with the oxygen even at ambient temperature; therefore the oxidative degradation is a severe material-engineering problem. Monitoring, or better, predicting the stability of organic and especially polymeric materials against oxidation is therefore of paramount importance. Commonly applied analytical methods such as measuring oxidation induction time (OIT) or oxidation onset temperature (OOT) using Differential Scanning Calorimetry (DSC) or other conventional thermoanalytical methods are unsuitable for long-term prediction of oxidative behaviour because of the use of elevated temperatures during these experiments: the high temperatures may invoke reaction pathways which are different from those encountered under the conditions of use.

Oxidation processes, leading to the cleavage of the polymer chains, occur randomly and stealthily. They may not be detected for years undergoing the rapid, autocatalytic acceleration once the reaction has reached a critical stage. Once the reduction of the molecular weight has reached a critical threshold both adhesion and cohesion properties are critically affected what, in turn, leads to the loss of required bond properties.

The oxidation of materials in the solid state generally starts on the surface and the oxidation progress is mainly diffusion controlled. The general overview of commonly applied examination methods and mechanism of the oxidation is given by Feller [1] Zweifel [2] and Scheirs [3]. The accelerated ageing is normally initialised by extreme environmental conditions. During the induction period, stabilisers are consumed while the organic matter remains stable, maintaining its original properties. At the end of the induction period, when the concentration of stabilisers reaches a sub-critical level, oxidative decay starts and the substance’s properties change. Frequently, oxidation reactions are self-accelerating (auto-oxidation) and the reaction progress may rapidly increase after the induction period.

The most common methods to test the kinetics of thermooxidation of organic substances are thermal analysis methods like Differential Thermal Analysis (DTA), DSC or thermogravimetry (TG). Substances are tested using isothermal or nonisothermal temperature profiles under an oxidative atmosphere, and the OIT or OOT are determined when the heat flow (DSC, DTA) starts to be exothermal or the sample mass loss starts (TG). In some special applications, these procedures are standardised (e.g. for automotive oils, cable insulations or polyolefines).

Especially for industrial and commercial applications, OIT- and OOT determination procedures are widely applied. Their advantages include an easy sample preparation, short measurement periods and established methods of data evaluation. A significant disadvantage of these short-time experiments lays in the application of the high experimental temperatures, generally above 180°C. The reasons of the use of such temperatures are following: (i) firstly to make sure that the oxidation starts within around 2 hours and (ii) to provide a distinct signal being larger than the baseline noise: the sensitivity of conventional TA instruments can be too low to record the beginning of the oxidation reaction. The evaluation of the bad correlation of OIT- and OOT data with the observed long-term stabilities under normal environmental conditions is reported in [4-6].

Depending on the properties of the substance, one or several phase transitions can occur between the temperature ranges at which the oxidative characteristic has been measured and those one for which the oxidative properties has been predicted (life-time determination). It seems to be obvious that the degradation kinetics may change for different low- and high temperature phases and the extrapolation of high temperature experimental results to ambient temperature can be of little value.

In such situation the alternative methods, based on the experiments carried out at low temperatures, should be applied for the characterisation of the long-term stability of organic substances. One of very suitable method is the chemiluminescence (CL).

1.1 Chemiluminescence

Luminescence is a term used for various phenomena, originating from electronically excited states.

Luminescence is a ‘cold light’, not an incandescent light. The emission of photons results from the relaxation of excited electrons (triplet-state) into their ground-state. This may be a quite quick process: the delay between the excitation and light emission is at least 10-10 seconds. Chemiluminescence (CL) includes all luminescence phenomena resulting from chemical reactions [7]. The fact that organic substances undergoing oxidation emit light has been recognized already at the second half of XIX century [8]. In the past few years, chemiluminescence has gained wide acceptance as a sensitive method to study the oxidative degradation of organic solid substances [9 -11].

1.2 Principles of chemiluminescence in organics

The emission of light during the oxidative degradation process of the organics is a part of the reaction course. The first step is the formation of unstable alkyl radicals, which immediately scavenge the oxygen from the atmosphere to form peroxy radicals. These react further and transform into different species in an accelerating degradation cycle (auto-oxidation, left part of figure 1). It is normally attributed to a transition of excited triplet-carbonyl-functions (3R=O*) into their ground state. The spectral range of the light emitted varies according to the type of substances involved. In most cases the chemiluminescence is observed in the short wave region of the visible spectrum from 380 to 450 nm. However, there are well-known exceptions: the relaxation of 1O2 can be detected in the infrared region at ca. 12000nm.

Figure 1. Simplified scheme of auto-oxidation of organics including a Russel mechanism (from Lacey et al. [12]).

The required energy (290-340 kJ mol-1) may be supplied by basically three different chemical mechanisms:

1. The combination of two peroxy radicals with concomitant fragmentation in a Russel mechanism [13] is strongly exothermal (460 kJ mol-1) [14]. The CL-emitter is an excited “triplet” carbonyl function (right-hand part of figure 1).

2. The direct homolysis of hydroperoxides followed by a cage reaction leads to an excited carbonyl-function and is combined with the evolution of 315 kJ mol-1 [15].

3. The metathesis of alcoxy or peroxy radicals provides 374 kJ/mol and 323 kJ/mol, respectively [16]. It has been shown, that the CL signal intensity reveals the existence of two kinetic stages during oxidative degradation of organic materials: The first one is correlated with the concentration of peroxide groups [17], the second stage corresponds to the oxidation propagation by the hydrogen abstraction responsible for the carbonyl formation [18].

1.3 Advantages of chemiluminescence in monitoring oxidation of organics

Compared to DSC and other conventional thermo-analytical methods, CL offers many advantages: due to its much higher sensitivity, experiments can be performed at much lower temperatures i.e. closer to the application related conditions. This fact is of the importance for the characterisation of substances with low temperature melting points, glass transitions, etc. The outstanding baseline stability of CL is of great benefit when performing long-term experiments [19]; moreover the CL-signal is related exclusively to the oxidation processes and therefore is not superposed by the signals resulting from the other reactions, including phase transitions. The instrumentation setup may be designed individually for special fields of applications and goals of research. The experiments can be performed with the sample masses as low as approx. 0.1 mg. A basic instrumentation would not be more expensive than a commercial DSC apparatus.

1.4 Determination of the kinetic parameters – isoconversional analysis

The noticeable weakness of the ‘single curve’ methods (determination of the kinetic parameters from single run recorded with one heating rate or isothermal condition only) has led to the introduction of ‘multi curve’ methods over the past few years, as discussed in the International ICTAC kinetics project [20- 23].

Degradation reactions are often too complex to be described in terms of a single pair of Arrhenius parameters and the commonly applied set of reaction models. As a general rule, these reactions demonstrate profoundly multi-step characteristics. They can involve several processes with different activation energies and mechanisms. In such situation the reaction rate can be described only by complex equations, where the activation energy term is no more constant but is dependent on the reaction progress α (E ≠ const but E = E(α)).

The isoconversional methods were introduced by Friedman [24] and Ozawa-Flynn-Wall [25, 26]. A detailed analysis of the various isoconversional methods (i.e. the isoconversional differential and integral methods) for the determination of the activation energy has been presented by Budrugeac [27]. The convergence of the activation energy values obtained by means of a differential method (Friedman) with those resulted from using integral methods (Ozawa-Flynn-Wall) comes from the fundamentals of the differential and integral calculus.

The differential isoconversional method of Friedman is based on the Arrhenius equation:

with

f (a): the model function
A: the pre-exponential factor
E: the activation energy
T: the temperature
t: the time
Friedman has applied the logarithm of the conversion rate d/dt as a function of the reciprocal temperature at any conversion a:

As f(a) is a constant in the last term at any fixed a, the logarithm of the conversion rate da/dt over 1/T shows a straight-line dependence with the slope of m = -E/R.

By the extension of the expression

with



one can predict the reaction rate or reaction progress having determined and using the following expression:

at any temperature profile such as isothermal, non-isothermal, stepwise, modulated temperature or periodic temperature variations, etc.

2. Experimental

The CL emission rate during oxidation of organic substances at ambient temperatures is too low to be detected. However, only moderate temperatures are required for providing detectable signals. The requirements for the oven applied are similar to those used in conventional TA measurements such as DSC or TGA: the necessary is the exact control of the required temperature profile even in long-term experiments and a gas exchange facility. Additionally the sample compartment must be absolutely light-tight.

The detection of the CL-emission may be achieved using a photomultiplier tube (PMT) with photon counting mode or slow scan charged coupled device (CCD) camera. PMTs are highly sensitive devices and allow short gating times, but their dynamic ranges are low and their use must be carried out with caution in order to avoid the saturation of the photocathode. The advantages of solid state detectors (CCDs) are their simplicity in use, their high dynamic range and the feature of imaging the sample to exhibit the inhomogeneous character of oxidation reactions. A third possibility to detect the CL-emission is to use the micro channel plates (MCP) or high-sensitivity CCDs. These types of sensors offer the best sensitivity in combination with the imaging facility, but are characterised by an exorbitant price, high operation complexity and a low dynamic range.

The instrumentations provided by AKTS - Chemiluminescence are fully automated and consist of thermoelectrically cooled PMT with photon counting mode and an oven chamber in combination with an optical path including a shutter system (to protect the highly sensitive detection unit against extensive light during sample handling and to provide background measurements). The single-cell instrumentation is designed especially for sensitive measurements at moderate temperature conditions (isothermal and non-isothermal mode) additionally allowing carrying out the experiments under controlled relative humidity in the temperatures up to 95°C. The multi-cell instrumentation enables the characterisation/comparison of four independent samples at the same temperature profile without cross contamination.

3. Results

In the present study we report on a new approach of kinetic analysis of the oxidation of organic solids at moderate temperatures using CL measurements carried out on newly-developed instrumentation. The kinetic characteristics of the oxidation process calculated from the chemiluminescence signals are subsequently applied for the prediction of the reaction progress under different temperature profiles.

The presented results depict the comparison of the oxidation reactions of natural rubbers with and without stabiliser in an oxygen atmosphere: This system is representative for many hot-melt formulations, especially HMPSA. The results shown in Fig.2 illustrate the influence of the stabiliser (5% Irganox™ 565) on the oxidation behaviour of the rubber (cis-1,4-polyisoprene) during non-isothermal heating in the range 30-120°C with a rate of 0.0132 Kmin-1 in the oxygen atmosphere.

Figure 2: CL signals recorded for unstabilised) and stabilised cis-1,4-polyisoprene during non-isothermal conditioning in the oxygen atmosphere.

CL data can be evaluated using the standard procedures already used in conventional high-temperature thermal analysis methods like OIT, OOT, for defining the points of interest such as initial and extrapolated onset temperatures, temperatures of the peak and extrapolated endpoint. The CL signals can be integrated, the first and second derivation can be determined.

The set of experimental data under isothermal conditions in the range 80-120°C (Fig.3) can be used for the evaluation of an Arrhenius relationship (inset).

Figure 3: Normalised CL-emission signals of unstabilised natural rubber (cis-1,4-polyisoprene) recorded during isothermal oxidation at 120, 110, 100, 90, 80°C. The inset presents the Arrhenius relationship for the extrapolated onset and peak temperatures.

At higher temperatures (> 100°C) the CL data corresponds well with the DSC data. When testing oxidative stabilities below 100°C, the limitations of DSC become obvious: Useful evaluation of such data about oxidation onset becomes tricky and is not reliable (Fig. 4).

Figure 4: Comparison between CL emission (black) and DSC heat flow (grey) signals at different isothermal conditions (120, 110, 100, 90°C) recorded during oxidation of unstabilized cis-1,4-polyisoprene.

3.1 Application of an advanced kinetic analysis of CL signals for the lifetime prediction

Generally kinetic analysis of decomposition processes or any gas-solid reaction can be applied to any type of thermoanalytical data (DSC, DTA, TG etc.) for the study of raw materials and products within the scope of research, development and quality assurance. The experimental procedure starts with the collection of the data directly related to the degradation kinetics such as mass loss or heat flow signals. These data are then used to determine the kinetic parameters such as activation energy (E) or pre-exponential factor (A) which allow predicting the reaction progress under various temperature modes. Generally, direct investigation of the reaction is very difficult at low temperatures (requiring very long measurement times), as well as under complex temperature profiles. Applying Thermokinetics software (www.akts.com), the rate and the progress of the reactions can be predicted on the basis of various well-defined temperature profiles [28].

The CL signals collected during the oxidation of unstabilized and stabilised natural rubber under non-isothermal conditions at different heating rates were used for the determination of the kinetic parameters used later for the prediction of the reaction progress. The normalized reaction rates determined by AKTS-Thermokinetics software after correction and optimization of the base line are depicted in Figure 5.

Figure 5: Normalised CL signals recorded during oxidation of unstabilized (left) and stabilized (right) natural rubber with different heating rates β (0.0054 - 0.073 K min-1).

The results of the evaluation of the kinetic parameters by the isoconversional method of Friedman are presented in Fig.6.

Figure 6: Top: Friedman differential isoconversional analysis of the long term oxidation process of unstabilized (left) and stabilized (right) natural rubber; Bottom: Dependence of the activation energy and the pre-exponential factor of natural rubber oxidation on the reaction progress α calculated by Friedman’s differential isoconversional method: unstabilized (left) and stabilized (right) natural rubber.

Once the kinetic parameters are determined, they can be applied to predict the course of the oxidation under different temperature profiles. The presented results clearly indicate the oxidative induction period after which the rate of the oxidation accelerates rapidly. The prediction of the oxidation of the natural rubber under isothermal conditions at low temperatures (4-40°C) is shown in Fig. 7.

Figure 7: Prediction of the oxidation progress of unstabilized (left) and stabilized (right) natural rubber at isothermal temperatures between 4 and 40°C.

The important goal of the investigation of the kinetics of the thermal decompositions is the need to determine the thermal stability of substances, i.e. the temperature range over which a substance does not decompose with an appreciable rate. The correct prediction of the reaction progress of materials which are unstable under ambient conditions requires accurate application in the calculations of both:

• the kinetic parameters

• the exact experimental temperature profile

The example showing the prediction of the properties of the rubber under more complicated temperature profile is depicted in Figure 8 presenting the oxidation progress of natural rubber at 20°C when the temperature changes with the modulations of 0, 5, 10 and 20 K each 24 h. The dependences shown in Fig. 8 indicate that even small temperature fluctuations can significantly change the stability of the substance: e.g. the amplitude of 10K at 20°C lowers the oxidation stability of natural rubber by half of lifetime.

Figure 8: Prediction of the oxidation progress of unstabilized (left) and stabilized (right) natural rubber at modulated isothermal temperature of  20°C, amplitudes 0, 5, 10 and 20 K each 24 h. Note that  the average temperature is the same for all temperature profiles. Depending on the type of reactions and stabilization the effect of the temperature fluctuations can have significant influence on the reaction rate.

More generally, calculations can be achieved for any fluctuation of the temperature which makes possible the predictions of thermal stability properties for varying climates.

Figure 9: Average daily minimal and maximal temperatures recorded for each day of the year between 1961 and 1990 (New York and Hong Kong).

Figure 10: Prediction of the oxidation progress of unstabilized (left) and stabilized (right) natural rubber at different climatic locations (New York and Hong Kong).

Exact consideration in the calculations of daily minimal and maximal temperature variations of worldwide climates provides therefore very valuable insight when interpreting and quantifying the reaction progress of materials subjected to atmospheric conditions.

4. Conclusions

Oxidative degradation of polymers can be monitored by the chemiluminescence signals resulting from the oxidation reaction. This method is orders of magnitude more sensitive as compared to the conventional methods of thermal analysis such as Differential Scanning Calorimetry, Differential Thermal Analysis or Thermogravimetry. The data acquired during the chemiluminescence experiments carried out iso-or nonisothermally can be evaluated by a model-free kinetic analysis to obtain meaningful and accurate predictions of the life time of organic materials in temperature domains which are representative for the life-cycle of the investigated materials. The method is illustrated by the results of the investigation and prediction of the efficiency of the stabilizers on the significant increase of the life-time of the natural rubber. The results of the investigation of the prediction of the properties of adhesives on the basis of the elaboration of the chemiluminescence signals by advanced kinetic method (in preparation) will additionally illustrate the potential of the application of chemiluminescence for the prediction of the life-time of organic solids.

5. References

[1] R. Feller, Accelerated Ageing. The Getty Conservation Institute, Marina del Rey, 1994.
[2] H. Zweifel., Plastic Additives Handbook, Verlag Carl Hanser, München, 2001.
[3] J. Scheirs, Compositional and Failure Analysis of Polymers, John Wiley & Sons, Chichester, 2000.
[4] J. Pauquet, R.Todesco, W. Drake, 42nd Int. Wire & Cable Symp, St. Louis, 1993, p. 77.
[5] E. Kramer, J. Koppelmann, ’Thermo-oxidative degradation of polyolefins observed by isothermal long-term DTA’, Polym. Eng. Sci., 27, 1987, p. 945.
[6] F. Gugumus,,’The Use of accelerated Tests in Evaluation of Antioxidants and Light Stabilisers’, Developments in Polymer Stabilisation - 8, G. Scott, (Ed.), Elsevier, London, 1987, p. 239.
[7] Neumüller, O.-A.: Römpps Chemie-Lexikon, Franckh’sche Verlagshandlung, Stuttgart, 8th Ed., 1979-1988.
[8] B. Radziszewski, Ber. D. Chem. Ges., 10 (1877) p. 70.
[9] L. Zlatkevich, ‘Chemiluminescence in evaluating thermal oxidative stability’, J. Polym. Sci. B, Polym. Phys., 28, 1990, pp. 425-429.
[10] M. Celina, G. George, ‘A heterogeneous model for the thermal oxidation of solid polypropylene from chemiluminescence analysis’, Polym. Degr. Stab., 40, 1993, pp. 323-335.
[11] L. Matisová-Rychlá, J. Rychly, ‘Thermal oxidation of nonstabilised and stabilised polymers and chemiluminescence’, J. Polym. Sci., 42, 2004, pp. 648-660.
[12] D. Lacey, V. Dudler, ‘Chemiluminescence from polypropylene. Part 1: Imaging thermal oxidation of unstabilised film’ Polym. Degrad. Stab., 51, 1996, p. 101.
[13] G. Russel , ‘Deuterium-isotope Effects in the Autoxidation of Aralkyl Hydrocarbons. Mechanism of the Interaction of PEroxy Radicals’, J. Am. Chem. Soc., 79, 1957, p. 3871.
[14] R. Vasiliev, Progr. Reaction. Kinetics, 4, 1967, p. 305.
[15] L. Reich, S. Stivala, ’ Some remarks on the origin of weak chemiluminescence during polymer oxidation’ Makromol. Chem., 103, 1967, p. 74
[16] E. Quinga, G. Mendenhall, ‘Chemiluminescence from Hyponitrite Esters. Excitet Triplet States from Dismutation of Geminate Alkoxyl Radical Pairs’ J. Am. Chem. Soc., 105, 1983, p. 6520.
[17] N. Billingham, E. Then‚ Chemiluminescence from peroxides in polypropylene. Part I: Relation of luminescence to peroxide content’ Polym. Degrad. Stab., 34, 1991, p. 263.
[18] L. Audouin V. Bellenger, A. Tcharkhtchi , J. Verdu‚ ‘Oxyluminescence of Cross-Linked Amine Epoxies: Diglycidylether of Bisphenol A-Diaminodiphenyl Sulfone System’ Polymer Durability. R. Clough, N. Billingham, K. Gillen (Eds). (1996), American Chemical Society: Washington DC, pp. 223-234.
[19] J. Pospíšil, Z. Horák., J. Pilař, N. Billingham, H. Zweifel, S. Nešpůrek, ‘Influence of testing conditions on the performance and durability of polymer stabilisers in thermal oxidation’, Polym. Degr Stab., 82, 2003, p. 156.
[20] M.E. Brown et al. ‘Computational aspects of kinetic analysis. Part A: The ICTAC Kinetics project data, methods and results.’ Thermochim. Acta, 355, 2000, p. 125.
[21] M. Maciejewski, ‘Computational aspects of kinetic analysis. Part B: The ICTAC Kinetics Project - The decomposition kinetics of calcium carbonate revisited, or some tips on survival in the kinetic minefield.’ Thermochim. Acta, 355, 2000, p. 145.
[22] A.K. Burnham, ‘Computational aspects of kinetic analysis. Part D: The ICTAC kinetics project - multi-thermal-history model-fitting methods and their relation to isoconversional methods.’ Thermochimica Acta, 355, 2000, pp. 165-170.
[23] B. Roduit, ‘Computational aspects of kinetic analysis. Part E: The ICTAC Kinetics Project - numerical techniques and kinetics of solid state processes.’ Thermochim. Acta, 355, 2000, p. 171.
[24] H. L. Friedman, ‘Kinetics of thermal degradation of char-forming plastics from thermogravimetry.’ J. Polym. Sci., C, 1965, p. 185.
[25] T. Ozawa, ‘New method of analysing thermogravimetric data.’ Bull. Chem. Soc. Japan, 38, 1965, p. 1881.
[26] J.H. Flynn, L.A. Wall, ‘A quick direct method for the determination of activation energy from thermogravimetric data’ Polym. Lett., 4, 1966, p.323.
[27] P. Brudugeac ‘Differentational non-linear isoconversional procedure for evaluating the activation energy of non-isothermal reactions‘ J. Therm. Anal. Cal., 68, 2002, p. 131.
[28] Advanced Kinetics and Technology Solutions Inc. Siders, Switzerland, http://www.akts.com