RICERCA ITALIANA     

Control and Modeling of Morphology of Semicrystalline Polymers under Realistic Processing Conditions

Università degli Studi di Salerno

Abstract

Aim of this project is to gain a deeper understanding and to build a sufficiently accurate modelling of microstructure evolution during polymer crystallization under processing conditions, i.e. under high pressures and cooling rates and under flow.
The proposed action encompasses many aspects on both the experimental and the theoretical side. In spite of this complexity, the objectives of the project are sharply defined. The Research Units (RU) participating to the project are complementary. Their skills cover all different aspects of the research work plan.
Most of the activity will be carried out with reference to an isotactic polypropylene (iPP) already adopted by the RUs involved in the project in previous research works. This choice is expected to minimize the initial characterization phases.
In the initials stages of the project the evolution of crystallinity and morphology under conditions much closer to processing than those normally available will be studied. To this purpose experimental apparatuses to allow for solidification in quiescent conditions under high cooling rates (up to hundreds of °C/s) and elevated pressure (up to thousands of bars) will be set up. Additionally, crystallization and morphology evolution during solidification will be monitored by recording signals of transmitted and scattered light intensity. Information concerning the morphology of solidified samples, the analysis of the transmitted light intensity during crystallization, the results related to spherulite nucleation and growth rate and to calorimetric investigations will be all fed to a model describing crystallization morphology and kinetics of the alpha phase under quiescent conditions. The competitive kinetics of the mesomorphic phase will be also taken into account. On the other hand, the orientation evolution in shear flow will be analyzed by optical methods (birefringence and light scattering). The results of these investigations will be interpreted in terms of molecular models of various complexity, aiming at predicting and describing also the melt free energy evolution.
Flow effects on crystallization morphology and kinetics will be monitored in rotational cells by means of various techniques (optical microscopy, light scattering, rheological parameters), as well as by direct observation in the proximity of a moving fibre. Furthermore, flow conditions that enhance the transition from spherulitic to fibrillar crystallization will be analysed. Attention will be focused on the quantitative determination of observable variables describing the crystallization process. Predictions of the evolution of these variables will be obtained by molecular models of the polymer molten state. Relevance of molecular weight with reference to flow-induced effects on crystallization morphology and kinetics will be studied by using other isotactic polypropylene samples characterized by different molecular weight distributions.
Based on the above described activity, an integrated model will be assessed to incorporate: i) crystallization kinetics and morphology evolution under quiescent conditions; ii) the molecular flow-induced evolution of melt characteristics; iii) the effects of flow on crystallization kinetics and morphology evolution. This model will be eventually implemented into a software simulation code of the injection moulding process. Model parameters (and to a limited extent the equations of the model) will be revised based on a comparison between model predictions and detailed experimental investigations on the microstructural parameters of specifically targeted injection moulded samples. <<<

Principal Investigator

Giuseppe TITOMANLIO Università degli Studi di SALERNO

Research Objectives

It is an obvious fact that the morphology distribution in polymeric manufacture is determined by the coupling between the thermo-mechanical history of each volume element and the specific characteristics of the polymer considered. Such a morphology is the final stage of a complex process of micro- and macro-structural evolution starting from the molten state. In more details, the evolution of morphology is determined by temperature, pressure and deformation histories applied to the material before, during and after its solidification. Such factors are expected to affect the crystallisation temperature, the rate and the modes of structural evolution at different length scales.
The morphological evolution during polymer crystallisation has been extensively studied in the past both experimentally and theoretically. In most cases, however, only specific aspects of the global problem have been considered. Furthermore, all models rely on experimental information obtained under conditions that are typically far away from those actually realized during the industrial transformation processes.
An integrated experimental activity between the partner laboratories will be developed within this research project on a single polymer, namely, an isotactic polypropylene (iPP). The aim will be to cover, within the same experimental activity, a set of experimental conditions as wide as possible and significantly closer than any other set already available to real processing conditions. This will provide a phenomenological picture of the phenomenon which is unique both for extension and for significance of the parameters. This picture itself is a considerable objective, even if not the main goal of the project. Some experimental activity will be also performed on some iPP specially selected in order to identify the relevance of molecular weight on the crystallisation rate and morphology evolutions under flow conditions. A series of experiments in flow conditions will be carried on.
The results of the multifaceted experimental activity will be transferred into models able to describe specific aspects of the microstructural evolution during polymer crystallization. This activity will include:
- a model describing the quiescent morphology evolution, in the processing window of pressure and thermal histories (for the latter, cooling rates up to several hundreds of degrees per second will be considered). The model will account for spherulite nucleation and growth, and also for the development of the mesomorphic phase;
- a model for the evolution of the melt orientation parameters, aimed at describing the morphological details of crystallization under flow conditions;
- a model describing the effects of free energy and orientation of the amorphous phase onto the kinetics and the morphology evolution of the crystalline phase.
The above sub-models will be synthesized into a global model able to describe the most relevant characteristics of both crystallinity and morphology evolution under process conditions (high cooling rates, high pressures, flow) accounting for the transition from spherulitic to fibrillar morphology and for the formation of a mesomorphic phase. Such a global model will be validated by comparison with results of the analysis of the morphological microstructure as determined in parts produced by injection moulding.
Each sub model itself is a significant goal; the highest objective is the global model since it will be referred to even more severe conditions with respect to the experiments performed for each of the sub-models. A step for a global optimisation is thus planned. During this step the global model results will be compared with the available data of morphology distribution in injection mouldings, and if necessary each sub-model will be re-checked.
The objectives presented above possess a different level of risk. The experimental activities are relatively low risk. Although some of them will require the development of "ad hoc" innovative apparatuses, they are all based on well-known techniques. In some cases, prototypes are already available at the participating laboratories. The development of theoretical sub-models is also characterised by a low level of risk, as it is related and is driven by a large body of experimental information which will work as a guide to define the lines of the modelling activity. The largest degree of risk is associated with the development of the global model, mostly in its validation stage. This is mainly due to the fact that the experimental activity cannot be always carried out under conditions close to real processing. Even in the case of a validation not completely successful, however, it is believed that this project can lead to a substantial step forward in the understanding and the modelling of the microstructure evolution of polymers under processing conditions. <<<

First Results

Work-package A
- Analytical expressions for the dependence of nucleation density, crystal growth rate and overall kinetic constant on isothermal crystallisation temperature (RU1)
- Assessment of athermal nucleation effects (RU1)
- Setting up an innovative light scattering apparatus for monitoring crystallinity and morphology evolution during crystallisation under cooling rates as high as hundreds °C/sec (RU3)
- Setting up an apparatus for sample solidification under high pressure at cooling rates of decades of °C (RU3 )
- Optimised procedures for morphological investigation by means of Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM) and Atomic Force Microscopy (RU1, RU3)
- Quantitative information on morphological features of samples prepared in quiescent conditions at high cooling rates and under pressure (RU1, RU3)
- Correlation between solidification conditions (cooling rate and solidification pressure) and phase distribution, spherulitic dimensions, lamellae thickness, long period (i.e. average size of a domain constituted by a crystalline region and a contiguous amorphous region) in the aforementioned samples (RU1)
- Model for the development of crystallinity and morphology validated over a very wide range of cooling histories and pressures. The model will also account for the competing evolution of different ordered phases under quiescent processing conditions (RU1, RU3)

Work-package B
- Results concerning evolution of melt orientation in shear flow (RU2, RU3)
- Development of a rheo-optical apparatus to measure birefringence and light scattering signals to be correlated with amorphous phase orientation and crystalline phase morphology (RU2)

Work-package C
- Determination of the effects of deformation and deformation rate on crystal nucleation and growth (RU1, RU2)
- Determination, by optical microscopy observations and by monitoring rheological parameters, of the characteristic crystallisation times as a function of orientation resulting from shear flow (RU2, RU3)
- Optimised procedure for quantitative description of semicrystalline morphology in samples crystallized under flow conditions. (RU1)Work-package B
- Results concerning the effect of crystallisation on rheology (RU2)
- Model for the evolution of melt orientation and its free energy (RU2, RU3)
- Model of the effect of crystallinity on rheology validated for linear viscoelastic functions (RU2, RU3)

Work-package C
- Correlation between material's parameters pertaining to nucleation and growth and flow conditions during crystallization (RU1, RU3)
- Correlation between morphological features and flow conditions applied during crystallisation (RU1)
- Relevance of molecular characteristic on flow induced crystallisation of the polymer (RU1, RU3)

Work-package D
- model of crystallisation kinetics and of morphology evolution during crystallisation as function of temperature, pressure and flow histories and of molecular cheracteristics of the polymer (RU1, RU2, RU3)

Work-package E
- Optimised procedure for quantitative morphological investigation of injection moulded samples (RU1, RU3)
- Morphology distribution in injection moulded samples and its relationship to processing conditions (RU1, RU3)

Work-package F
- Injection molding simulation code implementing the model of flow-induced crystallisation developed within work-package D, (RU3)
- Simulation of the injection moulding tests carried out within work-package E, (RU3)
- Assessment of a reliable model for crystallisation kinetics and morphology evolution accounting for pressure and flow effects (RU1, RU2, RU3)
- Final version of injection moulding simulation code capable to predict distribution of orientation, crystalline degree of each phase and morphology in injection mouldings (RU3). <<<

Timescale

24 months

National and international background

Thermoplastic polymers are usually processed in the molten state and solidification takes place under fast cooling during or immediately after flow. Often the polymer is exposed to simultaneous flow and cooling, where the cooling rate can be very high, up to several hundreds of degrees per second. Sometimes, for example during injection moulding, both flow and solidification take place at high pressure (several hundreds of atmospheres). In all cases, process conditions do have profound influence upon the resulting microstructure; in fact, both crystallization kinetics and morphology (size, shape, orientation of crystallites) can be dramatically different with respect to reference isothermal, quiescent, and low pressure conditions. The crystallinity evolution, in turn, affects the polymer rheological response, thus implying further changes to the material solidification behaviour.
For the above reasons, a detailed understanding of transformation processes, and particularly of the injection moulding process, requires a correspondingly detailed description of kinetics and of morphology evolution during crystallisation in a wide range of cooling rates, flow and pressure conditions. The understanding also of the rheological behaviour during solidification is a also a crucial point.
Prediction and control of microstructure during processing represent an important research field in industrial engineering. The recent implementation of the European Union program COST "Structuring of Polymers" and of the coordinated action PIAM, focused on morphology evolution, constitute further evidence that relations among process, structure and properties are frontier studies within the European and worldwide, scientific research.
Crystallisation under quiescent and slowly varying temperature conditions is a relatively well understood phenomenon. Quiescent crystallisation evolves through a typical two-stage pattern: the first stage is the formation of stable nuclei; the second stage is the subsequent crystallite growth. Under quiescent conditions both nucleation and growth depend only on the current temperature, even if some researchers assert that, under fast cooling conditions, nucleation depends also on the rate of change of the temperature. Current kinetic models connect external variables (such as temperature, temperature rate of change, and pressure) to both nucleation and growth rate, and therefore to the material final morphology [1-4]. Nevertheless, present experimental studies on non-isothermal quiescent crystallisation [5-8] consider only low cooling rates, much smaller than industrially relevant ones. In fact, characteristic cooling rates of industrial processes go from about one hundred of degrees per second (in film casting) up to hundreds of degrees per second - recorded at the polymer melt/cold mould surface contact in the injection moulding. In spite of these industrially important conditions, the Differential Scanning Calorimeter (DSC), which is the most widely used instrument in the crystallisation kinetics characterisation, can reach cooling rates of order of one degree per second. When pressure effects on crystallisation are analysed through PVT laboratory apparatuses, the major limit is again the low cooling rate.
In spite of many efforts in modelling the non-isothermal crystallization [5-9], usually the developed models are not sufficiently supported by experimental evidence. Validation of model predictions is limited by the absence of experimental data during crystallization in a wide range of cooling rates and pressures. As an example, the importance of athermal nucleation is still under discussion and experimental results under fast cooling conditions could be useful to discriminate between thermal and athermal nucleation. Also the effect of pressure is still not clearly understood. In some cases (such as for PET and Nylon) pressure seems to promote crystallisation. In others (such as for iPP) the opposite effect is observed. The evaluation of the competitive effects of thermodynamic and kinetic factors is still an open subject of research.
Recently, new experimental techniques have been developed to overcome the above shown limits. A technique to carry out fast cooling tests on thin polymer samples, where the thermal history is recorded, has been developed [10-13]. This technique allows to reach cooling rates of the order of hundreds of degrees per second. Quality and morphology of samples produced with this technique are analysed and the resulting data can be used to calibrate non-isothermal crystallisation models already available in literature [10-16] in a wide range of cooling rates.
An experimental technique, inspired by Ding and Spruiell's works [17], to study non-isothermal crystallisation by means of depolarised light analysis has been recently proposed [18-19]. This technique allows to overcome DSC limitations. In particular, real-time crystallinity evolution - during cooling tests up to several tens of degrees per second - has been already monitored.
High cooling rates during high pressure solidification experiments, which overcome PVT limitations, have been achieved by designing two new experimental techniques [20-22].
Isothermal quiescent crystallisation has been also analysed by comparing the viscoelastic moduli evolution with DSC data collected at the same temperature [23-27]. Carrot et al. [28] used rheological data to describe the isothermal quiescent crystallisation of polyolefin samples with different crystalline morphologies. Rheometric studies have shown that the development of a crystalline phase during flow has, in turn, very relevant effects on the material rheological behaviour. Even the formation of few percent of crystalline phase can cause a several orders of magnitude viscosity increase [29]. In summary, it can be concluded that deep interconnections - particularly in the early crystallisation stages - do exist among polymer crystallinity, thermal history, and flow history. This aspect is particularly relevant in modelling transformation processes as a sudden, considerable increase in the system viscosity corresponds to solidification. In the last decade, crystallisation has been described in terms of sol-gel transition [30], and critical-gel properties have been measured as a function of the thermal history for a number of polymers [30-31]. Up to now, steady-state rheological properties could not be measured because of the crystallinity evolution, possibly accelerated by the application of flow. Very recently, a new technique to measure rheological properties at a fixed degree of crystallinity has been proposed. The method, called "inverse quenching", allowed for the very first time the measurement of the steady-shear viscosity for an isotactic polypropylene as a function of the degree of crystallisation [32].
When flow effects are taken into account crystallinity evolution is a less understood phenomenon. The crystallisation kinetics are increased with respect to quiescent conditions, this phenomenon is known as Flow-Induced Crystallisation (FIC) [33]. The effects of flow on crystallisation kinetics have been analysed for many years (starting from Lagasse and Maxwell's work [34]) and are still an active area of research, thus demonstrating the relevance of this subject. In spite of the many studies dedicated to FIC [35-54], the phenomenon is still only partly understood. iPP spherulitic growth rate (see for example, Alfonso [45] and Monasse and coworkers [46-49]) has been analysed by optical microscopy under both quiescent and shear flow conditions. The effect of shear rate has also been confirmed by recent rheological studies [55].The role of orientation has been observed by Hsiao, and Kornfield with SAXS and rheo-optical techniques [51-54].
The experimental study of elongational flows under well controlled thermal conditions is at a more primitive stage. This is mainly due to the intrinsic experimental difficulties in realising this type of conditions. The few literature studies [56-57] confirm in any case that elongational flow is more effective than shear flow in promoting crystallisation.
Attempts to experimentally characterise FIC in complex flows, where mixed shear and elongation components coexist, are practically absent in the literature. Such conditions, however, are of crucial importance, as they are typical of many industrial processes such as injection moulding. Recently, experimental measurements of crystallinity in polymers subjected to injection moulding, where mixed flow conditions are present, have been presented [32]. The flow, however, is non isothermal, thus making difficult to separate the effect of flow from that of thermal history. Moreover, in this and other cases, the effect of pressure on the crystallisation kinetics (which is sometimes dominant) is not addressed.
Along with the kinetics, also the polymer crystalline morphology developed under flow can completely differ from the quiescent one. An example is given by the well known "shish-kebab" morphology developing in polyolefins subjected to high stretching rates [33]. A second example is given by the so-called "skin-core" morphology observed in injection moulded parts. In this case, both the number and the morphological details of the different layers depend on process conditions [29,58-60], and are determined mainly by the degree of orientation and by the cooling rate. In spite of the strong interest on this subject, a quantitative, systematic understanding and modelling of the microstructure distribution as a function of the operative conditions is still missing.
The common approach to the modelling of FIC considers that the increase of the degree of orientation of the polymer chains due to deformation results in an effective change of the melt free energy. The latter, in turn, affects the crystallisation rate. Based on this concept, several empirical or semi-empirical models have been proposed in the literature [29,61-62]. Although successful in describing FIC phenomena, such models are often limited to the description of particular flow conditions, and generally rely on the estimate of adjustable parameters.
In recent years, increasing attention has been paid to micro-rheological modelling of the FIC. In this case, the modelling route is to calculate the change in free energy induced by the flow in terms of the average orientation and stretching of the polymer chains. The main advantage of such models is that they are, at least in principle, fully predictive. The main disadvantage lies in the often complex level of mathematical representation. Some attempts to describe the crystallisation kinetics on a molecular basis have been recently introduced [63-65], but a fully predictive model for the crystallisation under flow conditions is still missing. In particular, it is crucial for a quantitative model to incorporate the effects of a time-varying end/or complex deformation history. The molecular approach seems particularly promising to describe the early stages of crystallisation, where the orientation of the amorphous phase and the dynamic interactions between the crystalline nuclei and the amorphous phase entangled network should play a crucial role.
The knowledge of the crystallisation kinetics under complex conditions (flow and thermal gradients, pressure) is a necessary prerequisite to an accurate modelling of the polymer transformation processes, in particular injection moulding.
Models have been proposed where the balance equation are coupled with equations describing crystallisation kinetics [66-69]. In some of them (see, for example, Guo [68]), the effect of flow has been also taken into account by assuming that the shear rate only affects the crystallisation induction time. In spite of such numerous efforts, however, the absence of a realistic evolution equation for the crystallisation kinetics and morphology accounting for the coupled effects of flow, cooling rate and pressure has in fact limited significant advancement in the simulation of the injection moulding process. <<<