Universidad de Castilla-La Mancha
Facultad de Ciencias y Tecnologías Químicas
Departamento de Ingeniería Química
DEVELOPMENT OF MICROCAPSULES WITH THERMAL ENERGY STORAGE (TES) CAPABILITY FOR CONCRETE APPLICATIONS
A thesis submitted for the degree of Doctor of Philosophy (PhD)
by
ANNA MARIA SZCZOTOK-PIECHACZEK
Supervisors:
Prof. Juan Francisco Rodríguez Romero
Prof. Anna-Lena Kjøniksen
Ciudad Real, April 2018
Prof. Juan Francisco Rodríguez Romero, Professor of Chemical Engineering at the University of Castilla-La Mancha and Prof. Anna-Lena Kjøniksen, Professor of Materials Science at Østfold University College, certify that:
The research work: "Development of Microcapsules with Thermal Energy Storage (TES) Capability for Concrete Applications", is part of thesis presented by Mrs. Anna Maria Szczotok-Piechaczek to aspire to the degree of Doctor in Chemical Engineering and that has been carried out in the laboratories of the Department of Chemical Engineering of the University of Castilla-La Mancha and Østfold University College under their direction.
And for the record, they sign this certificate in Ciudad Real, at April 16th 2018
Juan Fco. Rodríguez Romero Anna-Lena Kjøniksen
Table of content
Summary ...... vi
1. Introduction ...... 1
1.1. Motivation and background ...... 2
1.2. Objective ...... 3
1.3. Thesis structure ...... 5
2. State of art ...... 6
2.1. Global energy demand ...... 7
2.2. Methods for thermal energy storage ...... 9
2.3. Phase change materials ...... 10
Organic PCMs ...... 11
Inorganic PCMs ...... 11
Eutectics PCMs ...... 12
2.4. Microencapsulation of phase change materials ...... 12
General overview of microencapsulation ...... 12
Microencapsulation by suspension polymerization ...... 16
2.5. Selection criteria of PCMs for passive buildings ...... 23
2.6. Phase change materials in building applications ...... 24
Free cooling ...... 24
Peak load shifting ...... 25
Active building systems ...... 26
Passive buildings system ...... 27
2.7. Integration of phase change materials in passive systems ...... 28
Walls ...... 28
Roofs ...... 31
Floors ...... 32
Windows and shutters ...... 34 Table of content
2.8. Limitations in the PCMs application ...... 36
Thermal conductivity ...... 36
Supercooling ...... 36
Leakage of PCM ...... 37
Flammability ...... 37
References ...... 40
3. Materials and methods ...... 54
3.1. Materials ...... 55
3.2. Experimental setups ...... 57
Synthesis of the microcapsules ...... 57
Bulk polymerization ...... 62
Thermal properties ...... 62
3.3. Experimental procedures ...... 63
Polymerization of the microcapsules ...... 63
Polymerization in mass ...... 64
Synthesis of PNC-HEMA ...... 64
Concrete preparation...... 65
3.4. Characterization techniques and characteristic parameters ...... 65
Differential scanning calorimetry ...... 65
Thermogravimetric analysis ...... 66
Ash determination ...... 67
Density of microcapsules...... 67
Porosity ...... 68
Environmental scanning electron microscopy and energy dispersive x-ray spectrometry ...... 68
Low angle laser light scattering ...... 69 Table of content
Sieving ...... 70
Surface tension and contact angle ...... 70
Gel permeation chromatography ...... 72
Gas chromatography – mass spectrum analysis ...... 72
Fourier transform Infrared spectroscopy ...... 73
X-ray micro-tomography ...... 74
Micro-combustion calorimetry ...... 74
Cone calorimetry ...... 75
Compressive strength test ...... 76
Thermal conductivity ...... 77
Specific heat capacity/latent heat ...... 77
Energy saving aspects ...... 78
Process parameters ...... 79
References ...... 80
4. Optimization of microcapsules recipe and properties ...... 81
4.1. Background ...... 85
4.2. Results on optimization ...... 87
Preliminary experiments ...... 87
Theoretical framework ...... 90
Effect of different suspending agents on microcapsules properties ...... 91
Effect of continuous/discontinuous phase mass ratio on microcapsules properties ...... 99
Effect of stirring rate on microcapsules properties ...... 101
Effect of toluene/PCM mass ratio and the PCM content on microcapsules properties ...... 102
Influence of PVP concentration on microcapsules properties ...... 103 Table of content
Influence of PNC-HEMA incorporation on morphology and encapsulation efficiency ...... 116
Scale-up ...... 121
4.3. Conclusions ...... 123
References ...... 125
5. Synthesis of green microcapsules containing fatty acids as PCM...... 129
5.1. Background ...... 132
5.2. Materials ...... 134
5.3. Results ...... 135
Synthesis of microcapsules in suspension-like polymerization ...... 135
Reactivity of unsaturated fatty acids ...... 145
5.4. Conclusions ...... 151
References ...... 153
6. Synthesis of thermoregulating microcapsules with flame retardant properties .... 158
6.1. Background ...... 161
6.2. Materials ...... 163
6.3. Results ...... 166
Morphology and particle size of the products ...... 166
Thermal energy storage properties of the microcapsules ...... 171
Thermal degradation study of products ...... 173
Flame retardant properties of the microcapsules ...... 178
6.4. Conclusions ...... 182
References ...... 183
7. Thermal and morphological stability of microcapsules ...... 186
7.1. Background ...... 189
7.2. Materials ...... 190 Table of content
7.3. Results ...... 191
Thermal stability ...... 191
Study on neat PCMs ...... 192
Heat treated microcapsules ...... 207
Estimation of the diffusion coefficient of PCM from microcapsules ..... 210
7.4. Conclusions ...... 217
References ...... 219
8. Incorporation of microcapsules into concrete ...... 223
8.1. Background ...... 225
8.2. Materials ...... 226
8.3. Results ...... 228
Compressive strength ...... 228
X-ray micro-tomography ...... 230
SEM analysis ...... 231
Thermal properties ...... 233
Energy saving aspects ...... 236
Fire reaction properties ...... 238
8.4. Conclusions ...... 240
References ...... 241
9. Conclusions and future work ...... 245
9.1. Conclusions ...... 246
9.2. Future work ...... 248
Nomenclature...... 249
Publications and conferences ...... 253
Summary
Summary
This PhD Thesis focuses on the synthesis and characterization of thermoregulating microcapsules for concrete applications in the framework of a wide research program funded by the Research Council of Norway to introduce thermal storage abilities into Portland cement and geopolymer concrete. Microencapsulation permits the thermal features of phase change materials to be utilized while overcoming disadvantages such as easy release (when melted), flammability and volume change during phase transition. Microcapsules containing different phase change materials: paraffins and fatty acids (linoleic acid, erucic acid, oleic acid and palmitic acid) were prepared by a suspension- like polymerization technique using copolymers of styrene and divinylbenzene P(St- DVB) as the shell material.
Laboratory experiments were performed to investigate the influence of the type of suspending agents, continuous/discontinuous phase mass ratio, agitation rate, toluene/PCM mass ratio, PCM amount and suspending agent amount on the properties of the microcapsules. The optimization process was conducted for those materials which encapsulated paraffins and then evaluated in pilot plant scale. The obtained microcapsules revealed high energy storage capacity (> 100 J/g), encapsulation efficiency (> 80%) and a process yield (> 86%). This optimized formulation also ensures high mechanical and thermal resistance due to the highly consolidated crosslinked structure of the polymer shell.
Fatty acids were assayed as PCMs in order to have greener and more sustainable raw material for the microcapsules formulation. Microcapsules with unsaturated fatty acids had lower latent heat than expected at the same time presenting the high particle yield what indicated that these kind of acids played two different roles, as PCM and also as monomers, in the radical polymerization processes. At high initiator concentrations, the unsaturated fatty acids were observed to react being incorporated into the polymeric backbone of the shell.
Another important problem concerning the incorporation of polymer/paraffin microcapsules into building material is the flammability. To prevent this problem, the microcapsules shell formulation were modified by using the co-monomer
vi Summary
hexa(methacryloylethylenedioxy) cyclotriphosphazene (PNC-HEMA) with flame retardant properties. It was found that the incorporation of this flame retardant co- monomer microcapsules exhibit a pomegranate-like structure. Furthermore, the latent heat of the microcapsules increased in the presence of PNC-HEMA. Thermogravimetric analysis performed under atmospheric air confirmed that the PNC-HEMA raised the amount of residue after the burning process, promoting the formation of thermally stable char. Total heat release and peak heat release were lower than for corresponding microcapsules without PNC-HEMA.
Thermal and morphological stability of the microcapsules at high temperatures (up to 240 °C) was observed in-situ. It was concluded that the shape and size of the microcapsules were intact. However, the PCM content decreased with increasing temperature, diminishing to 0 J g-1 at 180 °C. In order to understand this phenomena, a diffusion model of PCM from the microcapsules was developed.
Finally, microcapsules synthesized with the optimal recipe at pilot plant scale were applied in Portland cement concrete (PCC). It was found that the compressive strength of PCC decreased with the addition of microcapsules at temperatures below and above melting temperature of PCM. The thermal conductivity decreased with increasing amount of microcapsules whereas the specific heat capacity increased. A power reduction of 20% could be obtained when 40% of sand was replaced by microcapsules. Cone calorimetry was used in order to evaluate the flammability and the negative effect of incorporation of 10% of MPCM into concrete, however in a heat flux of 50 kW m-2 ignition was not observed, proving little effect of 10% microcapsules on the flammability of concrete.
vii
1. Introduction
CHAPTER 1 INTRODUCTION
Chapter 1
1.1. Motivation and background
The total energy consumption is increasing quickly every year whereas the production of environmentally friendly energy is not enough to ensure the energy demand which causes important environmental consequences. Since the energy consumption will increase rather than decrease, the infrastructure required for electricity distribution will need expensive upgrades to follow with the demands. Energy-saving is therefore desired in order to minimize the negative environmental effect and avoid huge infrastructure investments. Due to narrow temperature region where humans feel comfortable, a huge amounts of energy is used for heating and cooling buildings. Materials such as concrete which have a high thermal mass may help solve this problem by absorbing thermal energy during the daytime and releasing it back during the night. Nevertheless, the thermal energy storage capacity of these materials is limited. In order to improve thermal energy storage capacity of building materials, phase change materials can be used. They are able to absorb, store and release huge amount of energy in a relatively low volume. However, utilizing bulk quantities of PCMs results in low thermal conductivity, solidification around the edges and diminished heat transfer. In order to solve this problem, microencapsulation can be employed. The shell of the microcapsules prevents the leaking of the encapsulated PCMs, interactions with the building materials, improves the heat transfer area per unit volume and ensures the ability to withstand volume change during phase change. Furthermore, microencapsulated PCMs have the advantage that they can be easily and economically incorporated into conventional building materials such as concrete, gypsum or mortar. So far, only microcapsules containing PCM with a hydrophilic shell were incorporated into concrete. Mostly, poly(methyl methacrylate) such as Micronal® supplied by BASF, acrylic shell from CIBA Specialty Chemicals or melamine- formaldehyde shell like Microtek® were used. After incorporation of microencapsulated PCM into concrete, the mechanical properties such as compressive strength decreases. When the microcapsules are added, the compressive strength often become too low to meet the standard requirements of constructive materials. This can be due to a weak microcapsule shell, because of the nature of the polymer shell, utilization of huge amounts
2 Introduction of core material, or the hydrophilic character of the microcapsules which greatly increases the water consumption needed for hydration. This thesis has been developed into the framework of an ambitious research project granted to the research group of Prof. Anna-Lena Kjøniksen at Østfold University College by the Research Council of Norway (project number 238198) and has been carried-out in collaboration with the research group of Prof. Juan F. Rodriguez in the facilities of the Institute of Chemical and Environmental Technology of the University of Castilla - La Mancha.
1.2. Objective
According to the existing environmental requirements for energy saving, the main aim of this thesis is the development of microcapsules containing phase change materials, with high mechanical resistance which permits energy storage after incorporation in concrete.
In order to achieve the aim, the research work was structured into the following specific objectives:
. Optimization and selection of the most suitable recipe for the encapsulation of phase change materials within a polymeric shell having high chemical and mechanical resistance. Selection of the most suitable suspending agent, continuous/discontinuous phase mass ratio, agitation speed, toluene/PCM mass ratio, amount of PCM and amount of suspending agent.
. Scale up of the process. Production of 100 kg of microcapsules.
. Understanding the process parameters responsible for the final morphology and thermal properties in the suspension-like polymerization process.
. Improvement of the sustainability of the microcapsules by using saturated and unsaturated fatty acids. Study of the possible reactivity between monomers and fatty acids.
. Improvement of the fire resistance of the microcapsules. Modification of the polymeric shell by incorporation of a flame retardant monomer.
3 Chapter 1
. In-situ study of the thermal and morphological stability of microcapsules at high temperatures. Estimation of the vapor pressure of commercial PCMs and modeling of the diffusion process of PCM in the microcapsules.
. Application of the obtained materials in the Portland cement concrete and study of the mechanical and thermal properties of the concrete. Evaluation of the flammability of the Portland cement concrete.
Attending to the objectives, the work plan shown in Figure 1.1 was performed for the development of the dissertation.
Synthesis of microcapsules
Variables: suspending agent, Characterization: Optimization of continuous/discontinuous phase SEM, EDAX, LLALS, the recipe and mass ratio, agitation rate, DSC, TGA, Optical properties toluene/PCM mass ratio, PCM tensiometer amount, suspending agent
amount Improvement of Study the adsorption of a surfactant onto the Characterization: the microcapsules, application of TGA, DSC, SEM, environmental EDAX, LALLS, FT-IR, environmentally-friendly PCM, aspect of the GC-MS, GPC, MCC, improved fire resistance, thermal microcapsules Ash determination stability of microcapsules
Characterization: Application of Study on the mechanical and Compressive strength, Guarded hot plates, microcapsules in thermal properties, evaluation of XRD, Environmental concrete the reaction to fire chamber, Cone calorimetry, MCC
Figure 1.1. Work plan performed in the thesis.
4 Introduction
1.3. Thesis structure
The thesis is divided in 9 chapters. Chapter 1 provides a general overview of the work performed in the project, highlighting the motivation and main objectives. Chapter 2 gives a review of different thermal energy storage methods, with a special attention to phase change materials. This chapter explain the suspension-like polymerization technique chosen in the thesis and gives short summary regarding possible applications of phase change materials in the building industry.
Chapter 3 provides details regarding experimental setups, procedures and characterization methods used in the project.
The results have been divided in 5 chapters. Chapter 4 is focused on the optimization of the recipe and the properties of the microcapsules. Additionally, the parameters which influence the morphology of the final microcapsules and the adsorption of suspending agent onto the shell of the microcapsules have been considered.
In Chapter 5, environmentally-friendly fatty acids with different chemical characteristics have be evaluated as phase change materials. The different behavior of saturated and unsaturated fatty acids in relation to the polymerization process and the thermal storage capacity is discussed.
Chapter 6 covers the investigation of the modification of the shell by using a flame retardant monomer. Different proportions of the flame retardant monomer and PCM have been studied. Special focus is given to the possibility of energy saving and fire resistance.
In Chapter 7, the thermal and morphological stability have been investigated. For that purpose, a mathematical model based on the vapor pressure of neat PCM and the diffusion of the PCM though the polymer network has been developed to describe the process of release of PCM from the microcapsules. The experimental data from the thermogravimetric analyses are used to evaluate the vapor pressure of the PCM and to calculate the PCM profiles in the microcapsules with time and temperature.
Finally, Chapter 8 describes the application of selected microcapsules in Portland cement concrete and study their influence on thermal and mechanical properties.
The thesis finishes in Chapter 9, which concludes the research work by describing the most important outcomes and provides recommendations for future work.
5
2. State of art
CHAPTER 2 STATE OF ART
State of art
2.1. Global energy demand
One of the main problems discovered over the last several decades is a high and quickly growing energy demand, which can result in supply difficulties, depletion of resources and destructive environmental impact such as climate change, global warming and ozone depletion. The growing energy demand is mainly due to economic growth, changes in individual sectors contributions to this growth, technology developments, and investor and consumer decisions. It is predicted that the world energy consumption will rise with 28% between 2015 and 2040 [1]. The main sources of energy in Europe are oil, gas and coal, with little impact of renewable resources or nuclear energy [1]. In 2014, renewable energy covered 14% of the global primary energy supply. The renewable energy sources mainly utilized at the current date are traditional biomass like fuelwood and animal waste, modern biomass, bioethanol and biodiesel, and electricity generated from hydro, wind, solar, biomass and geothermal energy. Figure 2.1 shows world energy consumption by energy source.
250 petroleum and other liquids 200 coal natural gas 150 renewables 100
(quadrillion Btu) (quadrillion 50 nuclear
World energy consumption energy World 0 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 Year
Figure 2.1. Prediction of world energy consumption by energy source [1].
The global CO2 emissions increased by an average of 1.9% per year between 1990 and 2014. Moreover, world energy-related carbon dioxide emissions are projected to grow with an average 0.6% per year between 2015 and 2040. Around two thirds of greenhouse gas emissions come from the burning of fossil fuels. In addition, the last part
7 Chapter 2
includes not only CO2 but also methane, nitrous oxide and fluorinated gases, from changes in land use, agriculture, certain energy sector operations and a variety of industrial processes. Figure 2.2 present the increase in the energy-related carbon dioxide emissions between 1990 and 2040.
45 40 35 30 25
20 emission
2 15
10 CO 5 (billion metric metric tons) (billion 0 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 Year
Figure 2.2. Prediction of energy-related carbon dioxide emissions [1].
According to the European Commission, the building industry is responsible for
40% of the energy consumption and 36% of CO2 emissions in the European Union. About 35% of buildings in the European Union are over 50 years old, and these buildings consume an average of 25 liters per square meter per year of heating oil, while new buildings generally need three to five liters per square meter per year less. In that way, it could be possible to reduce the total energy consumption by 5-6% and lower CO2 emissions by about 5% by improving the energy efficiency of buildings [2,3].
In order to satisfy latest European Union’s climate and energy targets “20-20-20” for the year 2020 which assume 20% decreased the greenhouse gases emission, 20% of energy delivered by renewables systems and 20% improvement in energy efficiency, energy storage respond well to these requirements [4]. Among the different types of energy storage systems, thermal, electrochemical and chemical are best suited [5].
8 State of art
2.2. Methods for thermal energy storage
Thermal energy storage (TES) is a technology which improve the balance between supply and demand of energy. The energy storage can be due to a change in internal energy of a material as sensible heat (SHTES), latent heat (LHTES) and thermochemical [6,7]. A comparison between available storage technologies is shown in Figure 2.3.
(c) (a) (b)
Figure 2.3. A comparison between (a) sensible heat and (b) latent heat storage and (c) chemical storage [8].
In sensible heat thermal energy storage (SHTES), the energy is accumulated by increasing the temperature of a solid or liquid, using the heat capacity and the temperature change of the material during the process but without phase change. The quantity of stored heat (Q) is dependent on the specific heat of the material (Cp), the temperature (T) in which the change occur and the quantity of the material (m) [7]:
푇푓 푄 = ∫ 푚퐶푝푑푇 = 푚퐶푝(푇푓 − 푇푖) (2.1) 푇푖
In latent heat thermal energy storage (LHTES), the heat absorption or release takes place when the material undergoes a phase change from solid to liquid or liquid to gas and conversely [7]. The storage capacity of a LHTES system can be describe as:
푇푚 푇푓 푄 = ∫ 푚퐶푝푑푇 + 푚푎푚∆ℎ푚 + ∫ 푚퐶푝푑푇 푇푖 푇푚 (2.2)
= 푚[퐶푠푝(푇푚 − 푇푖) + 푎푚∆ℎ푚 + 퐶푙푝(푇푓 − 푇푚)]
9 Chapter 2
Thermochemical systems are based on the energy absorption and release due to reversible breaking and reforming molecular bonds. The stored heat is dependent on the amount of stored material (m), the endothermic heat of reaction (Δhr), and the extent of conversion (ar) [7]:
푄 = 푚푎푟∆ℎ푟 (2.3)
The classification of the materials used for thermal energy storage was given by Abhat et al. [9] in 1983, and it is shown in Figure 2.4.
Figure 2.4. General classification of materials for thermal energy storage [6,9,10].
2.3. Phase change materials
Phase change material (PCM) is compound that is able to absorb, store and release energy during the phase change within a specific temperature range. When the temperature of the environment rises above the temperature of phase change of the material, the PCM absorbs the heat in an endothermic process and changes the state from solid to liquid. On the other hand, if the temperature drops, the PCM releases its heat of fusion to the environment and return to solid state. It is important to point out that during the phase changes, the temperature of the PCM and therefore that of its surroundings, remains practically constant, reducing the use of air conditioning systems or heating when
10 State of art the system is applied in buildings. PCMs are chemically classified as organic compounds, inorganic compounds and eutectics.
Organic PCMs
Organic phase change materials, are usually divided into paraffin and non-paraffin systems.
Paraffins are obtained from crude oil distillation and constituted of a mixture of different hydrocarbons. They are mainly in liquid or waxy solid state. They are safe, odorless, tasteless, non-toxic and non-corrosive products, chemically stable, with a high heat of fusion (around 200 kJ kg-1), low vapor pressure, good compatibility with other materials and high availability over a large range of melting points (from -10 to 90 °C). On the other hand, they are flammable (with flash point between 150 and 200 °C), have a relatively large volume change during the phase transition and have a low thermal conductivity (around 0.2 W m-1 K-1) [5,8,11,12]. Non-paraffins mostly involve fatty acids and their esters, alcohols and glycols. Due to different chemical structures, these compounds exhibit highly varied properties. They can have some properties in common with paraffins, which makes them particularly attractive. The most suitable group for building applications is fatty acids with melting range from 8 to 65 ºC and latent heat range from 150 to 247 kJ kg-1. Moreover, fatty acids are produced from vegetable and animal oils, and they are therefore more environmentally friendly, but at the same time more expensive. They are less flammable than paraffins but mildly corrosive [7,8].
Inorganic PCMs
Inorganic PCMs are mainly divided into salt hydrates, salt solutions and metals [8,13]. In contrast to organic PCMs, inorganics have higher phase change enthalpy and conductivity (almost twice the thermal conductivity of paraffin). They are also non- flammable. Nevertheless, some disadvantages such as subcooling, water solubility, with corrosive nature, phase separation and lack of thermal stability, limit their application [13]. The most common salt hydrates have a melting temperature between 0 and 99 °C and a latent heat range of 105 to 296 kJ kg-1 [8]. Low-cost hydrated salts include alloys
11 Chapter 2
of inorganic salts and water e.g. CaCl2·6H2O or NaSO4·10H2O. The phase change consist of hydration and dehydration of the salt which can melt to a salt hydrate, having less water or to an anhydrous form where salt and water are separated. During this process, the salts may melt incongruously to form smaller salts, which makes the process irreversible [14]. The incongruent fusion of the salts occurs when the salt is not completely soluble in its water of hydration at the melting point. Consequently, the undissolved salt is deposited at the bottom of the container due to its higher density and does not recombine properly with water when it subsequently solidifies [7,14].
Low melting metals and eutectics alloys are included in the group called “metallics”. They are not well developed as PCM yet. Basically, they present a high heat of fusion per unit volume but low heat of fusion per unit mass, high thermal conductivity, low vapor pressure and low specific heat [5].
Eutectics PCMs
Eutectic materials are classified as organic-organic, inorganic-organic and inorganic-inorganic constituted of at least two components, for which the single substances melt and solidify congruently [11]. Fundamentally, they melt and solidify without segregation, forming a proper mixture which act as single component. The large possibilities of mixtures facilitates the development of important materials for a specific applications, where an exact melting point or latent heat are required. Eutectics are characterized by a sharp melting temperature and high volumetric thermal storage density [15,16].
2.4. Microencapsulation of phase change materials
General overview of microencapsulation
Microencapsulation is a process where a thin shell is created around a microscopic droplet of active substance to produce capsules with useful properties. Solids, liquids and gases can be encapsulated, and the size of microcapsules ranges between 1 –1000 μm, depending on the microencapsulation method. These methods can be categorized into chemical and physical processes. Furthermore, physical processes can be subdivided into
12 State of art physico-chemical and physico-mechanical. Chemical methods consist of polymerization of monomers under conditions in which the polymer in the form of microparticles are obtained in situ. Physical methods are based on previously synthesized polymers, and microparticles are obtained without chemical reaction. The most common methods used for microencapsulation are summarized in Table 2.1 [17].
Table 2.1. Chemical and physical microencapsulation techniques.
Physical processes Chemical processes Physico-chemical Physico-mechanical methods methods Coacervation (simple Interfacial polymerization Co-extrusion and complex) Supercritical fluids Emulsion, dispersion, assisted Spray drying and precipitation, suspension microencapsulation congealing polymerizations (RESS, GAS, PGSS) Polycondensation Sol-gel encapsulation Fluid bed coating Layer-by-layer (L-B-L) Ionic gelation Spinning disk assembly Solvent evaporation Pan coating
Microencapsulation is interesting for numerous applications such as pharmaceuticals [18-20], food industry [21-23], textiles [24,25], electronic devices (optoelectronic devices [26], electrochromic devices and nanoelectronic components [27,28], sensing and catalysis nanocomponents [29], electrophoretic displays [30]) and in the building industry [8,11,31].
Principally, the morphology of microcapsules depends on the core material and the microencapsulation process. According to their inner structure, microcapsules can be categorized in mononuclear, polynuclear, and matrix types as shown in Figure 2.5. When the core material is surrounded by a continuous shell, mononuclear or core-shell microcapsules are formed. Moreover, using two different covers, double-walled particles can be obtained. Polynuclear microcapsules have many cores enclosed within a shell,
13 Chapter 2
whereas in a matrix, the core material is distributed homogeneously in the shell [17] (Figure 2.5).
Figure 2.5. Inner structure of microcapsules.
The mononuclear type of microcapsules is desired, however the mechanism by which those microcapsules are formed is very complex and can be controlled by several factors e.g. physical and chemical properties of the reactants, experimental installations, and microencapsulation methodology. In this way, in addition to spherical microcapsules, other types such as acorn (partial encapsulation) or even complete separation can be obtained (Figure 2.6) [32].
Figure 2.6. Different possible morphologies of the microparticles (a) non-encapsulated, (b) partial encapsulation – acorn, (c) complete encapsulation [32].
14 State of art
The main reasons for encapsulation of materials applied in the building industry are [33]:
. avoiding interaction between the core material and the outside environment, . avoiding the loss of the core material due to volatilization, . reducing the health risks that non-encapsulated materials may present e.g. toxicity, flammability, . avoiding unpleasant odors, . providing a high heat transfer area per unit volume, . controlling volume changes during the phase change, . preventing the leakage of the core material during the phase change, . facilitating the handling of materials for later applications, . the possibility of modifying the physical properties of the microcapsules by shell functionalization.
During the selection of shell material, it is necessary to take into account the compatibility of the shell not only with the PCMs but also with the material where the microcapsules will find the final application [34].
Between all available techniques, the suspension-like polymerization was selected for the microencapsulation of PCM due to important benefits listed below. At the same time, disadvantages compare with other techniques are included [35,36]:
Advantages:
. the dispersing medium (aqueous phase) maintains the viscosity during the polymerization, . aqueous phase allows easy control of temperature and contributes to the elimination of heat, . the level of impurities in the polymerized product is low, . the isolation of the final product is simple, . the process is usually faster than in other polymerizations (mass and solution) and average molecular weights higher than those of bulk polymerization can be obtained.
15 Chapter 2
Disadvantages:
. low productivity, little amount of product is obtained, . polymer builds up on reactor walls, agitators and other surfaces, . wastewater treatments are required, . normally, it cannot be carried out continuously, which limits the usability for commercial large productions.
Microencapsulation by suspension polymerization
Suspension polymerization is a chemical process which involves dispersion of a reactive, organic mixture in the form of droplets in an aqueous phase, producing polymer particles. The organic phase is also called discontinuous phase and constitutes monomers, an active agent (PCM), initiator and optionally a diluent. The suspension-like microencapsulation process is based on the conventional suspension polymerization process but adding the component that is going to be encapsulated. PCM does not interfere in the polymerization, but form micro droplets around which the shell is formed. The aqueous phase (continuous phase) is formed by water and suspending agents or stabilizers. Once the two phases are transferred to the reactor, the discontinuous phase is dispersed by means of vigorous agitation in the continuous phase. The polymerization proceeds in each droplet as bulk polymerization. The reaction is initiated by thermal decomposition of the initiator which is homolytically cleaved, producing two radicals. This method is used with organic peroxides and azo compounds, such as benzoyl peroxide (BPO) and 2,2’-azobisisobutyronitrile (AIBN) [37-39].
There are several basic requirements which must be met to design a successful reaction. The polymerization medium must be immiscible with the organic phase (monomer + PCM) and the initiator must be soluble in the monomer. The organic phase and the continuous phase should have similar densities to avoid phase separation by gravity. The most used polymerization medium is water, since it is cheap, harmless and highly available. Only if an adequate formulation of the raw materials and optimum operating conditions are used, an appropriate product is achieved.
16 State of art
a) Suspending agent
A suspension-like polymerization involve three stages. In the first stage, a liquid– liquid dispersion is made. The liquid organic phase is dispersed as small droplets maintained by the combined action of stirring and a suspending agent. The second stage includes a breakup-coalescence dynamic equilibrium of the monomer–polymer droplets, which define the final particle size. During this process, the droplets breakup due to the impellers shear stress and merge again after colliding with each other. In the last stage, the polymer particles are in solid state and they do not break-up and coalescence any more [38]. Accordingly, the mechanism responsible for the particle size control is the balance between the rate of drop break up and coalescence [40]. The break up and coalescence mechanism is shown in Figure 2.7.
Figure 2.7. Particle formation in suspension polymerization [35].
The suspension-like polymerization ensure the production of microcapsules with a particle size between 0.5 and 1000 μm, which is related to monomer droplets size obtained during dispersion [17]. Accordingly, there are several parameters which determine the final particle size. They are divided in three groups: geometric, operating conditions and raw materials properties. Geometric factors are due to the experimental equipment such as dimensions and type of the reactor, dimension and type of agitator, ratio of diameters between agitator and reactor, bottom clearance of the stirrer, and internal fittings. Operating conditions include agitation speed, polymerization time and temperature as well as continuous/discontinuous phase ratio. Finally, viscosities and densities of the raw materials, and the interfacial tensions between them have also great influence on the final results [36,40,41].
17 Chapter 2
The suspending agent is key player in the processes of emulsion, suspension, and particle and capsule formation at nano- and microscales. The suspending agent or stabilizer form a film or skin around the surface of the droplets and prevent coalescence and agglomeration of the monomer droplets during polymerization [36]. The increased concentration of suspending agent consequently decreases the particle diameter, influences the particle shape, agglomeration or settling [41-46]. Most of the studies available in the literature pay attention on the ability of such compounds to modify the interfacial relations in terms of hydrophobicity, amphoteric properties, and viscosity.
Suspending agents can be divided into two main groups, protective colloids and Pickering emulsifiers. Natural organic protective colloids such as alginates, tragacanth, agar, starch, carboxymethylcellulose (sodium salt), hydroxyethylcellulose, and methylcellulose and synthetic polymers e.g. styrene – maleic anhydride copolymer, poly(methacrylic acid), polyvinylpyrrolidone, poly(vinyl alcohol) or partially hydrolyzed poly(vinyl acetate) are soluble in water, having amphiphilic character which determines their ability to decrease the interfacial tension and form a protective layer at the monomer – water interface. The suspending agent also increases the viscosity of the aqueous phase, which facilitates improved dispersion. However, some studies utilizing small amounts of protective colloid (0.1 wt.%) were performed and the viscosity did not raise significantly. It was concluded that suspension stabilizers are mostly effective when present in the interfacial layers between the water and the monomer droplets, and the tendency for interfacial adsorption is more important than the increase in viscosity of the aqueous phase when choosing a suitable stabilizer [36]. Schematic representation of the adsorption of polyvinylpyrrolidone on the surface of a dispersed particle is shown in Figure 2.8.
Figure 2.8. Schematic representation of the adsorption of polyvinylpyrrolidone on the surface of a dispersed particle based on [47].
18 State of art
Pickering emulsifiers called also powdered dispersants include barium sulfate, talc, aluminum hydroxide, hydroxyapatite, tricalcium phosphate, calcium oxalate, magnesium carbonate, and calcium carbonate. They are finely divided, usually inorganic, insoluble solids wetted by two immiscible liquids, which must also exhibit a certain degree of self-adhesion. Adsorption of low molecular mass surfactants on the solid emulsifier can modify the wettability and make the surface more hydrophobic. One of the advantageous of those compounds is the possibility to dissolve out the dispersant after the polymerization (e.g., in dilute acid), thereby minimizing polymer contamination [36,41].
b) Monomers
Different monomers were chosen for the microencapsulation of PCM by suspension-like polymerization. Mostly, styrene [48,49] and styrene-acrylate co- monomers [50-52] and methyl methacrylate are used [53-55]. The employment of microcapsules to enhance thermal storage abilities in concrete will require enhanced physical resistance due to the aggressiveness of the mixing process and environment exposure. Very few studies about thermoregulating microcapsules from styrene and divinylbenzene have been reported in the literature. The incorporation of a crosslinker into the shell composition will provide to the polymer shell higher mechanical resistance. The greater physical resistance of the crosslinked shell of these materials is necessary for their employment in active thermo-accumulating systems in which the particles have to be also circulated and pumped by ducts as slurries. As far as we know, microcapsules from styrene-divinylbenzene co-polymer were reported by You et al. [56] and Li et al. [57] encapsulating n-octadecane. However, the microcapsules exhibit holes, making them concave in shape. Figure 2.9 presents the reaction scheme of styrene (St) and divinylbenzene (DVB) copolymerization to produce a poly(styrene-co-divinylbenzene) – P(St-DVB).
19 Chapter 2
St DVB P(St-DVB)
Figure 2.9. Reaction scheme of P(St-DVB) synthesis.
c) Phase Change Material
Different phase change materials have been encapsulated by suspension-like polymerization. Most of the studies focus on paraffin-derived compounds such as n- octadecane [56,58-60], n-hexadecane [61,62], tetradecane [48], Rubitherm®RT27 [48], Rubitherm®RT20 [48], paraffin wax [50,63]. On the other hand, it was concluded that this method is not suitable for encapsulation of polar PCMs like polyglycoles due to their hydrophilic nature [48]. The solubility of phase change material in the monomer must be sufficient to ensure the entrapment as core material instead of retaining in the aqueous phase. In addition, a high boiling point is required in order to prevent vaporization at the reaction temperature.
d) Initiator
The initiator utilized in suspension-like polymerization should be soluble in oil, therefore the polymerization is carried out within monomer droplets. The concentration of initiator is usually in the range from 0.1 to 0.5 wt. % based on monomer amount. In that way, monomer droplets contain large number of free radicals (~108) [36]. Accordingly, the kinetic mechanism is identical to that of bulk polymerization. The reaction temperature is limited by the boiling point of the continuous phase, which is water in most of the cases. Thus, the choice of the adequate initiator is controlled by the temperature at which reaction can be performed. The two most common classes of initiators in free radical polymerization are peroxide and azo compounds. These initiators
20 State of art decompose thermally and give free radicals which initiate the polymerization (Figure 2.10).
O O O C O 2 2 2O C O O + O Benzoyl peroxide Benzoyl peroxide radicals Figure 2.10. Free radical initiation of benzoyl peroxide.
The free radical initiator should be relatively stable at room temperature and decompose quickly enough at the reaction conditions. The rate of decomposition (kd) is dependent on the solvent – monomer system used and it is expressed by the initiator half- life time (t1/2), which is defined as the time it takes to decrease the concentration of the initiator to 50% its original value [64]. Since first order decomposition kinetics is assumed for most free radical organic initiators, the half-life can be related to the Arrhenius equation:
푙푛2 푡1/2 = (2.4) 푘푑
−퐸푎/푅푇 푘푑 = 퐴 ∙ 푒 (2.5)
‒1 where t1/2 and kd are half-life (s) and rate of decomposition (s ), respectively; A is the ‒1 Arrhenius frequency factor (s ); Ea the activation energy for the initiator dissociation (J mol‒1); R the universal gas constant (8.314 J mol‒1 K‒1); and T the reaction temperature (K).
As can be concluded from Table 2.2, benzoyl peroxide (BPO) and azobisisobutyronitrile (AIBN) are two the most used initiators. They have activation energies of 122.35 and 130.23 kJ mol‒1, respectively. Therefore, AIBN exhibits shorter half-life than BPO at the same temperature. Accordingly, AIBN is commonly used at 50- 70°C, whereas benzoyl peroxide at 80-95°C [65,66].
The most representative published research regarding microencapsulation of PCM by suspension-like polymerization are summarized in Table 2.2. This table include type of polymer shell, PCM, suspending agent used and initiator.
21 Chapter 2
Table 2.2. Research regarding the microencapsulation of PCM by suspension-like polymerization. Suspending Shell PCM Initiator Reference agent PSt PRS®paraffin wax PVP BPO [48,63,50] PSt Tetradecane PVP BPO [48] PSt Nonadecane PVP BPO [48] PSt Rubitherm®RT20 PVP BPO [48] PSt Rubitherm®RT27 PVP BPO [48] PSt Rubitherm®RT31 PVP BPO [49] P(St – MMA) PRS®paraffin wax PVP BPO [51] P(MMA - EGDMA) n-Octadecane PVA AIBN [54] P(St-DVB) n-Octadecane PVA + PVP AIBN [56,67] P(St- BDDA) n-Octadecane SMA AIBN [68] P(St- DVB-NGD) n-Octadecane SMA AIBN [68] P(St- BDDA -DVB-NGD) n-Octadecane SMA AIBN [68] P(DVB-NGD) n-Octadecane SMA AIBN [68] n-Octadecane + AIBN + PMMA/PAMA SMA [69] Pyrrole BPO
P(MMA) Na2HPO4·7H2O Span-80 BPO [55] P(MAA – BMA) n-Octadecane SMA AIBN [70] P(MAA– BMA) Paraffin wax SMA AIBN [70] PMMA n-Octadecane SMA BPO [71] P(MMA-BDDA) n-Octadecane SMA BPO [71] P(MMA-TMPTA) n-Octadecane SMA BPO [71] P(MMA-PETRA) n-Octadecane SMA BPO [71] P(BMA-co-BA) n-Octadecane SMA AIBN [72] P(BMA-co-BA-co-MAA- n-Octadecane SMA AIBN [72] PETA)
22 State of art
Table 2.2. Research regarding the microencapsulation of PCM by suspension-like polymerization (continuation).
Suspending Shell PCM Initiator Reference agent P(BMA-co-MAA-PETA) n-Octadecane SMA AIBN [72] P(BMA-co-AA-PETA) n-Octadecane SMA AIBN [72] PMMA n-Hexadecane PVA BPO [62] P(MMA-BA) n-Hexadecane PVA BPO [62] PSt Paraffin wax PVP BPO [73] Methocel PEDMA Paraffin mixture AIBN [74] 90 HG P(MMA-TEVS) Paraffin PVA BPO [75] Abbreviations: styrene (St), methyl methacrylate (MMA), 1,4-Butylene glycol diacrylate (BDDA), neopentyl glycol diacrylate (NGD), allyl methacrylate (AMA), n-butyl methacrylate (BMA), sodium salt of styrene-maleic anhydride polymer (SMA), methacrylic acid (MAA), divinylbenzene (DVB), trimethylolpropanetriacrylate (TMPTA), pentaerythritol tetraacrylate (PETRA), acrylic acid (AA), poly(ethylene dimethacrylate) (PEDMA), triethoxyvinylsilane (TEVS), sorbitan trioleate (Span-80), disodium hydrogen phosphate heptahydrate (Na2HPO4·7H2O).
2.5. Selection criteria of PCMs for passive buildings
PCMs are able to decrease the energy demand of cooling and heating systems, stabilizing the indoor temperature fluctuations. Nevertheless, for a desirable material to be effectively implemented in latent heat storage systems, several selection criteria must be considered. They are shown in Figure 2.11 Concerning the thermal comfort for humans, the melting point of the PCM should be in the range of 10 to 30 °C. The precise temperature should be chosen based on average day and night temperatures changes and other climatic limitations [15].
From a thermodynamic point of view, the PCM should have a high latent heat per volume unit, i.e., a small volume of PCM can absorb or release higher amounts of energy, consequently leading to a lighter building envelope [76]. In addition, a large specific heat capacity and high thermal conductivity are required in order to achieve faster thermal responses [77].
23 Chapter 2
Although, the thermodynamic properties are crucial selection criteria for the PCM application, other important features should be taken into account i.e. its chemical aspects, chemical stability, low volume change and little supercooling during the solidification process. Moreover, the PCM should not affect the environment during the manufacture, distribution, installation or operation of the storage system. Accordingly, it is desired that the PCMs are non-toxic, non-corrosive, non-flammable and non-explosive [15].
In order to achieve a long-term stability, the PCM should be resistant to a number of repeated melting/freezing cycles results. And finally the economic aspect is of course important, and a reasonable cost and high availability on the market overshadows all other aspects from an industrial point of view. Accordingly, the annual capital and operating costs have to be lower than those required for commonly used energy supplies.
Figure 2.11. Selection criteria based on PCM characteristics [15].
2.6. Phase change materials in building applications
The ability to increase the thermal comfort and energy savings by optimizing the temperature fluctuations are an interesting benefit of PCMs. The possibilities for building applications can be divided into four main categories: free cooling, peak load shifting, active building systems and passive building systems [11,78].
Free cooling
24 State of art
In free cooling systems, PCMs store outdoor coolness during the night and release it indoor during the day. The PCMs absorb the heat from passing air in a ventilation system or from water in a pipe system during the day, store as latent heat and release it to cool the building when the temperature increases and the need for cooling arises. In order for these systems to work properly it is required that the ambient temperature allows the PCM to freeze and melt during one day [79-81].
Previous studies have reported the use of commercial PCM granules containing paraffin for enhancement of a revolutionary floor-coupled ventilation system. The cooling load during the day was divided between the PCM and a commercial air conditioner. The PCM granules are embedded under the floor and used as the storage material. When the indoor air temperature was set to 28 °C, the cooling load could be reduced by 50% when using a conventional storage system, but by 92% as a result of storing cold in the PCM [82]. Others proposed a theoretical model and a prototype office building with an applied free cooling system in United Kingdom. The investigated system ensured sufficient heat storage to prevent overheating of the building during summer, giving the possibility of reducing the CO2 emissions with around 430 tons per year if it replaced conventional air-conditioning units in 2000 offices around the country [83,84].
Peak load shifting
Peak loads increase during the day putting pressure on the electrical grid. This lead to the necessity of dimensioning the system for higher heating or cooling loads to facilitate the need for heating, ventilation and air conditioning (HVAC) systems. One of the solutions to this problem is shifting the peak load away from the peak hours of electrical demand using PCMs. The PCMs help to divide the peak load throughout the day, reducing the highest peaks. Figure 2.12 shows the reduction and shifting of the peak load by the application of PCM [11].
25 Chapter 2
Figure 2.12. Reduced and shifted peak load by utilization of PCM [11].
Recently, a comprehensive overview of previous studies related to load shifting control strategies using different cold thermal energy storage facilities including building thermal mass, thermal energy storage system and phase change material was reported [85]. It was concluded that phase change materials with its high thermal energy storage capability have been successfully used for shifting load in buildings, reducing peak load between 10% and 57% for concrete and for lightweight constructions, respectively. Moreover, reported studies demonstrated possible cost savings from 11% to 96.6% and improved thermal comfort.
Active building systems
An active storage system works by utilization a forced convection heat transfer and in some cases mass transfer. Particularly, it can be located inside the storage material or storage medium itself circulating through a heat exchanger. This system require one or two tanks for the storage media. Active systems can be classified as direct or indirect systems. In the first one, the heat transfer fluid also acts as the storage medium, whereas in the latter system, a second medium is used for storing the heat. Thus, indirect systems are dual-medium storage systems in which the heat transfer fluid passes through the storage medium for charging and discharging the storage material [86]. Accordingly, integration of PCM with systems such as solar heat pump systems, heat recovery systems and floor heating systems are categorized as active strategies. The combination of those
26 State of art systems not only reduce the peak load as mentioned above, but they can achieve further savings through a reduced electrical demand for HVAC systems [87]. A thermal energy storage active system can be combined with the core of the building such as ceiling, floor, and walls, in external solar facades, suspended ceiling, ventilation system, photovoltaic systems and water tanks [86,87]. The possible ways of the incorporation of TES into buildings are shown in Figure 2.13.
Figure 2.13. Active thermal energy storage system integrated in buildings [86,87].
Passive buildings system
Passive systems are charged and discharged without any mechanical contribution, hence solar radiation, natural convection or temperature difference are used. So far, passive building systems and their application have achieved most interest. In this technology, PCMs are integrated into the building envelopes to increase the thermal mass.
In cold climates buildings require large amounts of insulation to reduce heating loads in the winter, which is the cause of large temperature fluctuations in the summer due to excessive over heating caused by a lack of thermal mass. PCMs which melt during the day time and solidify during nighttime will prevent rooms from overheating during the daytime in the warm months and may also reduce the need for heating during nighttime in the winter [8,11,78,87].
27 Chapter 2
2.7. Integration of phase change materials in passive systems
Walls
Numerus studies about incorporation of PCM into walls have been performed, since this is the most common solution so far. Covering the interior part of the wall, the system can absorb and release heat during a large part of the day and thus enhance the thermal energy storage capacity. Experimental and numerical studies have been conducted, investigating the effect on the indoor climate and the energy demand. In addition, PCM can be introduced inside the material or as a new layer.
Shilei et al. [88] built an ordinary room and a room using PCM gypsum wallboard in the northeast of China. As PCM, capric and lauric acids were mixed with the proportion of 82:18% and applied to the wallboards. They found that the PCM wallboards can diminish indoor air fluctuation and the heat transfer to the outdoor air and thereby keep the warmth. The indoor air fluctuation could be mitigated by 1.15 °C as shown in Figure 2.14.
Figure 2.14. Indoor air temperature of a room with an ordinary wall and with a PCM wall [88].
Similar studies with full-scale test rooms for a typical day in winter, summer and mid-season with PCM wallboards were investigated [89]. The PCM composite consisted of 60% microencapsulated paraffins in a copolymer, creating a flexible sheet. The natural convection mixing of the air was improved, reduced overheating and lower surface temperature on the walls were observed. On the other hand, two stories dwelling located
28 State of art in Greece with gypsum wallboard containing microencapsulated PCM was examined [90]. BASF-Micronal® microcapsules with melting point 23 °C were used. This system worked properly only between September and June, since July and August had temperatures above the solidification point, preventing solidification of the PCM.
The effect of refurbished buildings with PCM plaster integrated on the inner side of the exterior building envelope on energy saving and indoor comfort for cooling season were compared for different cities [91]. Simulation of PCM with different melting temperatures, thickness of the wallboard and the location of the PCM layer were performed. For that purpose, PCM with phase transition ranging from 26 °C to 29 °C were selected. The highest energy saving effect for Ankara was 7.2% whereas other cities (Athens, Naples, Marseille and Seville) achieved between 3 and 4.1%. Moreover, the temperature comfort was prolonged by 15.5% for Seville and 22.9% for Naples. It was also pointed out that during the summer and winter time, the period with temperatures below and above the phase change temperature, would not always be sufficient for the PCM to fully solidify or the heat available during the day would not be enough to completely melt the PCM.
Previous studies have reported thermal tests, economic and environmental evaluations of a lightweight aggregate concrete with macroencapsulated paraffin for applications in residential building walls in Hong Kong [92]. This system was able to decrease the interior temperature at the room center by 4.7 °C, at the internal surface of the panel by 7.5 °C and achieved an indoor temperature reduction of 2.9 °C. Regarding the environmental evaluation, a depletion of 465 kg CO2-eq/year or 12.91 kg CO2- eq/year/m2 was achieved. The application of macroencapsulated paraffin in lightweight aggregate concrete building walls was shown to be profitable after 29 years, while the average life span of a residential building in Hong Kong is 60 years.
At Ulster University studies of the incorporation of 30% by weight of microencapsulated PCM with a melting point 23 °C in clay wallboards were performed. The wallboards were mounted on the internal walls of a test cell and subject to illumination for 6 h by a 28 kW solar simulator. In the reference cell the internal temperature raised over 28 °C within 5 h whereas the clayboard with PCM kept the temperature below 26 °C during the 6 h tested. However, some rupturing of the
29 Chapter 2
microcapsules containing the PCM was observed, which can lead to leakage of PCM from the wallboard [93].
Recently, the project NanoPCM in which University of Castilla La Mancha was involved, developed and tested new insulation components containing PCM in prototype small houses. Rigid polyurethane foam with microcapsules from LDPE-EVA and Rubitherm®RT27 as PCM were installed covering completely the walls and ceiling of monitored prototypes in Spain and Poland. It was found that increasing PCM microcapsules amount could improve the insulation behavior and increase the energy efficiency by 12%. Polyurethane foam with PCM microcapsules stored 5 times more heat than reference material [94,95].
These studies were focused on lightweight constructions where PCM was added as a new layer. However, PCM can be also be incorporated into the material such as in a concrete matrix or in open cell cements called thermocrete [15] and to gypsum mortar.
Initially, simulations and validation of a wall sample incorporated with microencapsulated PCM into the interior of the plaster in a lightweight office with no active cooling system were conducted [96]. The melting temperature range of the PCM was 24 to 26 °C. The results showed that the latent heat stored in the PCM decrease the temperature of the wall surface. Consequently, improved thermal comfort for 6 h longer than with the reference material was observed, and as a result the required cooling load was reduced. Additionally, Cabeza et al. [97] evaluated the thermal properties of concrete with microencapsulated PCM. Two real size concrete cubicles were built to study the effect of the incorporation of a microencapsulated PCM with a melting point of 26 °C. A compressive strength over 25 MPa and a tensile splitting strength over 6 MPa after 28 days were obtained, giving the opportunity of this composite for structural purposes. Moreover, the cubicle without PCM had the maximum temperature 1 °C higher and minimum temperature 2 °C lower than that with PCM and the maximum temperature in the wall with PCM appeared 2 hours later than without PCM.
30 State of art
Roofs
Incorporation of the PCMs to roof systems can absorb the incoming solar energy and the thermal energy from the surroundings to reduce temperature fluctuations inside the building [11].
Recently, the effect of Compact Storage Module panels including PCM with different layer thicknesses (5 and 10 mm) and melting points (23, 25 and 27 °C) on the building HVAC operation based on the Fanger thermal comfort model were tested [98]. The panels were placed inside the walls and roof in order to reduce the HVAC energy consumption of the building. It was concluded that the system containing the PCM with a melting point of 27 °C achieved an annual energy saving of 10–15% in all thermal comfort categories. Moreover, a layer of 10 mm exhibited higher energy savings and lower payback periods than other studied layer thickness.
Several attempts have been made to perform numerical studies with experimental validation the year round thermal management of buildings, where macro-encapsulated PCM was incorporated in a layer on the roof, as shown in Figure 2.15 [99]. This model was used to simulate the behavior of a system with a double layer of macro-encapsulated
PCM. An inorganic salt hydrate consisting of 48% CaCl2 + 4.3% NaCl + 0.4% KCl +
47.3% H2O was used as the PCM. This PCM was encapsulated in a stainless steel panel. A double layer PCM incorporated in the roof was recommended in order to decrease indoor air fluctuations during all seasons.
Figure 2.15. Sketch of a PCM incorporated in a roof system [99].
31 Chapter 2
After that, a numerical investigation to compare thermal performance of different PCM roofs for a dwelling were performed by other authors [100]. The effect of different PCM melting temperatures (30, 34 and 38 °C), PCM layer thicknesses (40-100 mm) and roof slopes (0.05–0.5%) were analyzed using the ANSYS Fluent software. The results showed a 3 h delay in the peak temperatures of the base layer of the PCM roof. Temperature delays of 10 °C, 13 °C and 12 °C were observed for PCM with melting points of 30, 34 and 38 °C, respectively. The PCM layer thickness affected the heat flux and average temperature of the upper surface of the PCM whereas the roof slope did not have significant influence on the thermal performance.
In addition, 1000 m2 of smart thermal insulation panels have been installed by Spanish company, Acciona Construction, and applied in the roof of the Higher Education Institute of the Liège Province in Belgium. The panels consisted in polyurethane foam in which PCM microcapsules were added. The actions implemented in that building refurbishment are expected to reduce energy consumption by up to 75% and return on investment of approximately seven years [101].
Floors
Although, floors usually have limited contact with solar radiation, in areas where the sun shines for large parts of the day they have a large potential and can benefit from incorporation of PCMs.
Xu et al. [102] conducted simulations and experimental validation of the thermal performance of shape-stabilized PCM floor used in a passive floor system during the winter season. The influence of various factors such as thickness of the PCM layer, melting temperature, heat of fusion, thermal conductivity of the PCM and air gaps between the PCM and the covering material on the room thermal performance was studied. It was stated that the most efficient PCM should have a melting temperature equal to the average indoor air temperature of sunny winter days, whereas the latent heat and the thermal conductivity of the PCM should be larger than 120 J g-1 and 0.5 W m-1 K-1, respectively. Moreover, from a construction aspect, the thickness of the shape-stabilized PCM plates used under the floor should not be larger than 20 mm and the air-gap between PCM plats and the floor should be as small as possible.
32 State of art
With the purpose of enhancing the internal comfort without using an air conditioning system, a small-scale experiment studying the thermal performance of a hollow concrete floor panel filled with PCM (Figure 2.16) was performed [103]. The peak load with the PCM floor panel was found to be shifted 7.3 h whereas it was 0.8 h for the reference floor panel without PCM. The thermal energy was stored by PCM in 70% after 6 h and 100% after approximately 8 h. This work proved the usability of a hollow concrete floor panel filled with PCM for a lightweight building envelope to improve the thermal comfort. After that, the same research group, introduced shape-stabilized PCM into the same floor panel to regulate the building’s internal temperature. A numerical model concerning different concentrations of PCM was developed using Comsol Multiphysics and the validation was performed experimentally. The achieved results showed good correlation, reducing the surface temperature variations by approximately 2 °C and peak load shifting was amplified thanks to the incorporation of PCM [104].
(a) (b)
Figure 2.16. Schemes of the utilization of an internal thermal mass in the PCM floor panel during (a) the daytime and (b) the nighttime and the concrete hollow floor slab [103].
33 Chapter 2
Windows and shutters
Windows are part of the building which lead to a higher energy consumption. In summer time high solar radiation causes an increased need for mechanical cooling whereas in winter large parts of the energy escapes through glazed facades, leading to a need for mechanical heating. Glazed facades have low thermal inertia, and there is no way of storing excess heat. The most important issue to overcome when using PCM is ensure sufficient transparency of the window. Thus, transparent PCMs must be used for that purpose. So far, translucent PCMs have been used for PCM windows, since they enable relatively high amounts of visible light to pass through, however they do not offer the same visibility as regular windows [11].
In order to accomplish this requirement, the study on solar transmittance of a commercial grade PCM were carried out [105]. Commercial grade PCM with purity 99.3% and melting point 31.98 °C has been chosen. Pure PCM with a thickness of 4 and 30 mm showed a solar transmittance of 90.7 and 80.3%, respectively. Together with a low thermal conductivity, PCM can be considered as an interesting transparent thermal insulating medium.
Some studies were performed evaluating the thermal performance of glazed insulating units for an indoor office room [106]. The measurement was conducted during six months including time: winter, summer and mid-season. An improved thermal conductivity was observed for most parts of the year with the PCM prototype. Nevertheless, the two systems showed similar properties on thermal comfort during cloudy days. Thus, the importance of selecting the correct melting temperature for the PCM was highlighted. The study was accomplished by a full-scale test on a PCM glazing prototype [107]. The results showed that PCM glazing was able to diminish the specific total daily entering energy in the summer time by more than 50%. Moreover, heat loss reduction during the day in the winter was observed. However, this technology was not effective for achieving solar heat gains.
A different approach was exhibited by evaluating the thermal effectiveness of a PCM curtain integrated into a window system to reduce the solar heat gain [108]. The modeled window system consisted of a glass layer, a cloth-PCM curtain and an air gap in between as shown in Figure 2.17.
34 State of art
Figure 2.17. Sketch of a PCM curtain window [108].
This study developed a numerical model to analyze different PCM thicknesses, melting temperatures and air gaps. Generally, the PCM curtain could effectively decrease the temperature of the inner surface. It was concluded that natural convection in the air gap plays a significant role in transferring heat into the buildings, and that this is increased when the air gaps becomes larger. Combining a 5 cm thickness of the air gap with a 5 mm PCM layer with a melting temperature of 29 °C was found to have the best performance and reduced the heat gain by 16.2%. When increasing the PCM layer to 10 and 15 mm, reductions of 23.8% and 30.9% were achieved, respectively.
Sun protectors with implemented PCMs in window shutters were studied in order to see if the solar heat could be absorbed before it reached the indoor space [109]. The shutters containing PCM could reduce 23.29% of the heat gain through the windows.
The PCMs incorporated in passive structures may enhance the thermal storage capacity and diminish indoor air fluctuation. Moreover, reduced overheating and decreased surface temperature on the walls, may maintain the comfort temperature longer than in reference system. Besides, solar heat gain in warmth climate was diminished by windows shutter containing PCM. From the environmental point of view, the CO2 emission can be decreased. The thermal heat storage depends on the amount of incorporated PCM into the material. The local climate and seasonal variations in temperature should be also taken into account in order to ensure the charging and discharging of the PCM. Thus, the PCM
35 Chapter 2
selection seems to be a key design factor to consider when optimizing performance considering seasonality. Additionally, the solution to use double PCM layers seems to have a potential for year-round performance in climates with heating and cooling needs.
So far, application in walls is the most developed part of the building. However, floor and ceiling also showed great potential. In addition, environmental impact is also omitted through most of the studies [110]. Still, there is little information regarding cost analysis and economic evaluations to date so this issue require further studies.
2.8. Limitations in the PCMs application
As mentioned in the section 2.3 PCMs revealed some disadvantages in the practical application. Solutions trying to overcome any kind of limitation are continuously under development, since thermal energy storage is a promising alternative for saving fossil fuels. The inclusion of various additives is a versatile way of improving the PCMs.
Thermal conductivity
The successful improvement of thermal conductivity has been performed by addition of carbon fibers [111] and carbon brushes [112]. The influence of various carbon nanofillers on the thermal conductivity and energy storage properties of paraffin-based nanocomposites has been investigated. Nevertheless, the addition of nanofillers reduce the phase change enthalpy with minimum effect on the phase change temperature [113]. In addition, the thermal conductivity of PCM can be increased by graphite matrices [114-
117] or TiO2 nanoparticles [118].
Supercooling
Supercooling in inorganic PCM occurs due to poor nucleating properties thus, addition of a chemical additives e.g. borax in nanocrystals, which can behave as nucleation seeds are suggested solutions [7]. The incongruent melting of the salts can be reduced by adding water to the system [119]. To overcome the problem of phase segregation, the use of mechanical agitators has been suggested [14].
36 State of art
Leakage of PCM
On the other hand, the leakage of PCM or diffusive of low viscous liquids through the materials can be solved by encapsulation [11,120]. More impermeable and resistant shells are in continuous development. Currently, the encapsulation is being performed at three size scales, this is nano-, micro-, and macroencapsulation. Numerous studies have been performed in order to find the most suitable materials and technique which accomplish industrial requirements.
Flammability
One of the drawbacks of PCMs is their flammability. However, thanks to their high energy storage possibilities, they are widely tested to be incorporated into the building materials such as walls, floors, ceilings or windows. Nevertheless, it increases the fire risks and cause some limitation in the application which could be avoided if the risks were known and evaluated at design stage [121]. Moreover, taking into account highly improved PCMs usability during microencapsulation into polymer shell, the resistance to fire may be reduced [122].
In order to solve this problem flame retardants can be used. Flame retardants are chemicals added to combustible materials to improve their resistant to ignition. Their goal is to minimize the risk of a fire starting in case of contact with a small heat source or in case that material has ignited, they should slow down combustion and prevent the fire from spreading to other items [123].
In order to improve the fire resistant properties, flame retardants additives may be used in admixture with PCMs, e.g., decabromodiphenyl oxide, octabromodiphenyl oxide, or antimony oxide [124]. The addition of a flame retardant additive to the PCM not only enhance the PCM’s thermal stability but also acts as a nucleating agent. Additionally, fire-retarded form-stable PCM products consisting of paraffin RT21 and propyl ester, high density polyethylene (HDPE) and fire retardants such as magnesium hydroxide, aluminium hydroxide, expanded graphite, ammonium polyphosphate, pentaerythritol, and treated montmorillonite (MMT) have been evaluated [125].
37 Chapter 2
Recently, rigorous safety regulations and flammability requirements have been imposed on building materials to protect the buildings from fire. In order to comply with these rules, some new approaches have been investigated and applied successfully in laboratory tests. Specifically, addition of non-flammable surfaces to the plasterboard, such as aluminum foil and rigid polyvinyl chloride films have been tested. In addition, an insoluble liquid fire retardant (e.g. Fyrol CEF) were applied on the plasterboard after placing the PCM layer. In that way, the insoluble fire retardant substitute part of the PCM and some remains on the surface, providing self-extinguishing characteristics to the plasterboard. Brominated hexadecane and octadecane as PCMs and fire retardant surface coatings have also been used [6,12].
Furthermore, improvement of PCMs were performed by some research groups. Most of them were focused on form-stable PCM prepared by combination of paraffin with high density polyethylene. Cai et al. [126] produced form-stable phase change materials based on paraffin/high density polyethylene composites with addition of halogen-free flame retardants from expandable graphite combined with ammonium polyphosphate and zinc borate. The results indicated a higher amount of charred residue, contributing to improved thermal stability. The flame retardant additives had little effect on the TES capacity of the paraffin. These systems mainly acts in a condensed-phase forming a charred residue with partial gas phase flame retardancy [127]. Similar studies were performed by Zhang et al. [128] who developed paraffin/high density polyethylene/intumescent flame retardant system (IFR) containing ammonium polyphosphate, pentaerythritol and melamine with iron [128] or expanded graphite (EG) [129] as flame retardants. Results showed that the introduction of iron presented synergistic effect with IFR, improving the flame retardancy similarly to EG which formed the first char layer at the onset of combustion. However, in both studies the latent heats of the form-stable PCMs were lower than the theoretical values. After that, Sittisart and Farid [125] developed form–stable PCM containing paraffin (or propyl ester), HDPE and fire retardants such as magnesium hydroxide, aluminium hydroxide, EG, ammonium polyphosphate (APP), pentaerythritol (PER) and treated montmorillonite (MMT). The best results were found for the form-stable PCM which contained APP + PER + MMT and APP + EG since it can self-extinguish by forming a large residue. Moreover, addition of fire retardants to PCM did not impose significant changes to its thermal properties.
38 State of art
Finally, Wang et al. [130] investigated the flame retardancy of shape stabilized phase change materials based on paraffin, high density polyethylene and styrene–butadiene– styrene copolymer (SBS). The effects of organomontmorillonite (OMMT), expanded graphite and crosslinking of the polymer matrix were evaluated. Based on the horizontal burning test, the best flame retardancy was achieved containing OMMT/EG/TBHP, with the longest burning time and a large amount of residue.
A different approach was evaluated by Fang et al. [131] who prepared composites consisting of palmitic acid as PCM and SiO2 as the supporting material with melamine as flame retardant. The presence of melamine in the composites enhanced the thermal stability and flame retardant property. In addition, the microstructure of the charred residue after combustion indicated that the homogeneous and compact charred residue decreased the flammability of the composites.
39 Chapter 2
References
1. U.S. Energy Information Administration (2017) International Energy Outlook 2017. https://www.eia.gov/. Accessed 21.11.2017 2. European Commission (2017) Energy Efficiency: Buildings. https://ec.europa.eu/energy/en/topics/energy-efficiency/buildings. Accessed 21.11.2017 3. EU Directive 2010/31/UE EP S, 2010. 4. European Commission (2017) Europe 2020 strategy. https://ec.europa.eu/. Accessed 15.11.2017 5. Zhang H, Baeyens J, Cáceres G, Degrève J, Lv Y (2016) Thermal energy storage: Recent developments and practical aspects. Progress in Energy and Combustion Science 53 (Supplement C):1-40. doi:https://doi.org/10.1016/j.pecs.2015.10.003 6. Cabeza LF, Castell A, Barreneche C, de Gracia A, Fernández AI (2011) Materials used as PCM in thermal energy storage in buildings: A review. Renewable and Sustainable Energy Reviews 15 (3):1675-1695. doi:https://doi.org/10.1016/j.rser.2010.11.018 7. Sharma A, Tyagi VV, Chen CR, Buddhi D (2009) Review on thermal energy storage with phase change materials and applications. Renewable and Sustainable Energy Reviews 13 (2):318-345. doi:https://doi.org/10.1016/j.rser.2007.10.005 8. Akeiber H, Nejat P, Majid MZA, Wahid MA, Jomehzadeh F, Zeynali Famileh I, Calautit JK, Hughes BR, Zaki SA (2016) A review on phase change material (PCM) for sustainable passive cooling in building envelopes. Renewable and Sustainable Energy Reviews 60:1470-1497. doi:http://dx.doi.org/10.1016/j.rser.2016.03.036 9. Abhat A (1983) Low temperature latent heat thermal energy storage: Heat storage materials. Solar Energy 30 (4):313-332. doi:https://doi.org/10.1016/0038- 092X(83)90186-X 10. Zalba B, Marı́n JM, Cabeza LF, Mehling H (2003) Review on thermal energy storage with phase change: materials, heat transfer analysis and applications. Applied Thermal Engineering 23 (3):251-283. doi:https://doi.org/10.1016/S1359-4311(02)00192-8 11. Kalnæs SE, Jelle BP (2015) Phase change materials and products for building applications: A state-of-the-art review and future research opportunities. Energy and Buildings 94:150-176. doi:http://dx.doi.org/10.1016/j.enbuild.2015.02.023 12. Khudhair AM, Farid MM (2004) A review on energy conservation in building applications with thermal storage by latent heat using phase change materials. Energy
40 State of art
Conversion and Management 45 (2):263-275. doi:https://doi.org/10.1016/S0196- 8904(03)00131-6 13. Ferrer G, Solé A, Barreneche C, Martorell I, Cabeza LF (2015) Review on the methodology used in thermal stability characterization of phase change materials. Renewable and Sustainable Energy Reviews 50 (Supplement C):665-685. doi:https://doi.org/10.1016/j.rser.2015.04.187 14. Baetens R, Jelle BP, Gustavsen A (2010) Phase change materials for building applications: A state-of-the-art review. Energy and Buildings 42 (9):1361-1368. doi:https://doi.org/10.1016/j.enbuild.2010.03.026 15. Zhou D, Zhao CY, Tian Y (2012) Review on thermal energy storage with phase change materials (PCMs) in building applications. Applied Energy 92 (Supplement C):593-605. doi:https://doi.org/10.1016/j.apenergy.2011.08.025 16. Souayfane F, Fardoun F, Biwole P-H (2016) Phase change materials (PCM) for cooling applications in buildings: A review. Energy and Buildings 129 (Supplement C):396-431. doi:https://doi.org/10.1016/j.enbuild.2016.04.006 17. Ghosh SK (2006) Functional coatings: by polymer microencapsulation. John Wiley & Sons, 18. Kumar M (2000) Nano and microparticles as controlled drug delivery devices. J Pharm Pharm Sci 3 (2):234-258 19. Zhao Q, Li B (2008) pH-controlled drug loading and release from biodegradable microcapsules. Nanomedicine: Nanotechnology, Biology and Medicine 4 (4):302-310 20. Desai S, Perkins J, Harrison BS, Sankar J (2010) Understanding release kinetics of biopolymer drug delivery microcapsules for biomedical applications. Materials Science and Engineering: B 168 (1):127-131 21. F. Gibbs SK, Inteaz Alli, Catherine N. Mulligan, Bernard (1999) Encapsulation in the food industry: a review. International journal of food sciences and nutrition 50 (3):213- 224 22. Gharsallaoui A, Roudaut G, Chambin O, Voilley A, Saurel R (2007) Applications of spray-drying in microencapsulation of food ingredients: An overview. Food Research International 40 (9):1107-1121. doi:http://dx.doi.org/10.1016/j.foodres.2007.07.004
41 Chapter 2
23. Estevinho BN, Rocha F, Santos L, Alves A (2013) Microencapsulation with chitosan by spray drying for industry applications – A review. Trends in Food Science & Technology 31 (2):138-155. doi:http://dx.doi.org/10.1016/j.tifs.2013.04.001 24. Nejman A, Cieślak M, Gajdzicki B, Goetzendorf-Grabowska B, Karaszewska A (2014) Methods of PCM microcapsules application and the thermal properties of modified knitted fabric. Thermochimica Acta 589:158-163. doi:http://dx.doi.org/10.1016/j.tca.2014.05.037 25. Sánchez P, Sánchez-Fernandez MV, Romero A, Rodríguez JF, Sánchez-Silva L (2010) Development of thermo-regulating textiles using paraffin wax microcapsules. Thermochimica Acta 498 (1):16-21. doi:http://dx.doi.org/10.1016/j.tca.2009.09.005 26. Reddy KR, Lee K-P, Gopalan AI, Kang H-D (2007) Organosilane modified magnetite nanoparticles/poly(aniline-co-o/m-aminobenzenesulfonic acid) composites: Synthesis and characterization. Reactive and Functional Polymers 67 (10):943-954. doi:http://dx.doi.org/10.1016/j.reactfunctpolym.2007.05.023 27. Reddy KR, Sin BC, Yoo CH, Sohn D, Lee Y (2009) Coating of multiwalled carbon nanotubes with polymer nanospheres through microemulsion polymerization. Journal of Colloid and Interface Science 340 (2):160-165. doi:http://dx.doi.org/10.1016/j.jcis.2009.08.044 28. Reddy KR, Lee K-P, Gopalan AI (2007) Self-assembly directed synthesis of poly (ortho-toluidine)-metal (gold and palladium) composite nanospheres. Journal of nanoscience and nanotechnology 7 (9):3117-3125 29. Hassan M, Reddy KR, Haque E, Minett AI, Gomes VG (2013) High-yield aqueous phase exfoliation of graphene for facile nanocomposite synthesis via emulsion polymerization. Journal of Colloid and Interface Science 410:43-51. doi:http://dx.doi.org/10.1016/j.jcis.2013.08.006 30. Yu F, Fang Y, Wang J, Xu Y, Shi J (2016) Fabrication of compact poly(methyl methacrylate-co-butyl methacrylate-co-acrylic acid) microcapsules for electrophoretic displays by using emulsion droplets as templates. Colloid and Polymer Science 294 (8):1359-1367. doi:10.1007/s00396-016-3901-z 31. Konuklu Y, Ostry M, Paksoy HO, Charvat P (2015) Review on using microencapsulated phase change materials (PCM) in building applications. Energy and Buildings 106:134-155. doi:http://dx.doi.org/10.1016/j.enbuild.2015.07.019
42 State of art
32. Shirin-Abadi AR, Khoee S, Rahim-Abadi MM, Mahdavian AR (2014) Kinetic and thermodynamic correlation for prediction of morphology of nanocapsules with hydrophobic core via miniemulsion polymerization. Colloids and Surfaces A: Physicochemical and Engineering Aspects 462 (Supplement C):18-26. doi:https://doi.org/10.1016/j.colsurfa.2014.08.009 33. Hawlader MNA, Uddin MS, Khin MM (2003) Microencapsulated PCM thermal- energy storage system. Applied Energy 74 (1-2):195-202. doi:10.1016/S0306- 2619(02)00146-0 34. Hawes D, Banu D, Feldman D (1990) Latent heat storage in concrete. II. Solar energy materials 21 (1):61-80 35. Yuan H, Kalfas G, Ray W (1991) Suspension polymerization. Journal of Macromolecular Science, Part C: Polymer Reviews 31 (2-3):215-299 36. Vivaldo-Lima E, Wood PE, Hamielec AE, Penlidis A (1997) An Updated Review on Suspension Polymerization. Industrial & Engineering Chemistry Research 36 (4):939- 965. doi:10.1021/ie960361g 37. Cowie JMG, Arrighi V (2007) Polymers: chemistry and physics of modern materials. CRC press, 38. Nunes DSS, Coutinho FMB (2002) Acrylonitrile–divinylbenzene copolymer beads: influence of pre-polymerization step, stirring conditions and polymerization initiator type on the polymer particle characteristics. European Polymer Journal 38 (6):1159-1165. doi:https://doi.org/10.1016/S0014-3057(01)00292-0 39. Watanabe Y, Ishigaki H, Okada H, Suyama S (1997) New method for determination of free-radical production efficiency of organic peroxides and azo compounds. Polymer Journal 29 (9):733-736 40. Jahanzad F, Sajjadi S, Brooks BW (2005) Characteristic intervals in suspension polymerisation reactors: An experimental and modelling study. Chemical Engineering Science 60 (20):5574-5589. doi:https://doi.org/10.1016/j.ces.2005.04.063 41. Tobita H, Hamielec AE (2000) Polymerization Processes, 2. Modeling of Processes and Reactors. In: Ullmann's Encyclopedia of Industrial Chemistry. Wiley-VCH Verlag GmbH & Co. KGaA. doi:10.1002/14356007.o21_o01.pub2 42. van Zyl AJP, Wet-Roos Dd, Sanderson RD, Klumperman B (2004) The role of surfactant in controlling particle size and stability in the miniemulsion polymerization of
43 Chapter 2
polymeric nanocapsules. European Polymer Journal 40 (12):2717-2725. doi:10.1016/j.eurpolymj.2004.07.021 43. Mayya KS, Bhattacharyya A, Argillier JF (2003) Micro‐encapsulation by complex coacervation: influence of surfactant. Polymer International 52. doi:10.1002/pi.1125 44. Su J-F, Wang L-X, Ren L (2007) Synthesis of polyurethane microPCMs containing n-octadecane by interfacial polycondensation. Colloids and Surfaces A: Physicochemical and Engineering Aspects 299 (1-3):268-275. doi:10.1016/j.colsurfa.2006.11.051 45. Sánchez-Silva L, Rodríguez JF, Sánchez P (2011) Influence of different suspension stabilizers on the preparation of Rubitherm RT31 microcapsules. Colloids and Surfaces A: Physicochemical and Engineering Aspects 390 (1-3):62-66. doi:10.1016/j.colsurfa.2011.09.004 46. Borreguero AM, Carmona M, Sanchez ML, Valverde JL, Rodriguez JF (2010) Improvement of the thermal behaviour of gypsum blocks by the incorporation of microcapsules containing PCMS obtained by suspension polymerization with an optimal core/coating mass ratio. Applied Thermal Engineering 30 (10):1164-1169. doi:10.1016/j.applthermaleng.2010.01.032 47. Winslow FH, Matreyek W (1951) Particle Size in Suspension Polymerization. Industrial & Engineering Chemistry 43 (5):1108-1112. doi:10.1021/ie50497a031 48. Sánchez L, Sánchez P, Lucas A, Carmona M, Rodríguez JF (2007) Microencapsulation of PCMs with a polystyrene shell. Colloid and Polymer Science 285 (12):1377-1385. doi:10.1007/s00396-007-1696-7 49. Sánchez-Silva L, Rodríguez JF, Romero A, Sánchez P (2012) Preparation of coated thermo-regulating textiles using Rubitherm-RT31 microcapsules. Journal of Applied Polymer Science 124 (6):4809-4818. doi:10.1002/app.35546 50. Sánchez L, Lacasa E, Carmona M, Rodríguez JF, Sánchez P (2008) Applying an Experimental Design to Improve the Characteristics of Microcapsules Containing Phase Change Materials for Fabric Uses. Industrial & Engineering Chemistry Research 47 (23):9783-9790. doi:10.1021/ie801107e 51. Sánchez-Silva L, Rodríguez JF, Romero A, Borreguero AM, Carmona M, Sánchez P (2010) Microencapsulation of PCMs with a styrene-methyl methacrylate copolymer shell by suspension-like polymerisation. Chemical Engineering Journal 157 (1):216-222. doi:10.1016/j.cej.2009.12.013
44 State of art
52. Sanchez-Silva L, Tsavalas J, Sundberg D, Sánchez P, Rodriguez JF (2010) Synthesis and characterization of paraffin wax microcapsules with acrylic-based polymer shells. Industrial & Engineering Chemistry Research 49 (23):12204-12211 53. Taguchi Y, Yokoyama H, Kado H, Tanaka M (2007) Preparation of PCM microcapsules by using oil absorbable polymer particles. Colloids and Surfaces A: Physicochemical and Engineering Aspects 301 (1):41-47. doi:https://doi.org/10.1016/j.colsurfa.2006.12.019 54. Chang CC, Tsai YL, Chiu JJ, Chen H (2009) Preparation of phase change materials microcapsules by using PMMA network-silica hybrid shell via sol-gel process. Journal of Applied Polymer Science 112 (3):1850-1857. doi:10.1002/app.29742 55. Huang J, Wang T, Zhu P, Xiao J (2013) Preparation, characterization, and thermal properties of the microencapsulation of a hydrated salt as phase change energy storage materials. Thermochimica Acta 557 (Supplement C):1-6. doi:https://doi.org/10.1016/j.tca.2013.01.019 56. You M, Wang X, Zhang X, Zhang L, Wang J (2011) Microencapsulated n-Octadecane with styrene-divinybenzene co-polymer shells. Journal of Polymer Research 18 (1):49- 58. doi:10.1007/s10965-010-9390-8 57. Li W, Zhang X-x, Wang X-c, Tang G-y, Shi H-f (2012) Fabrication and morphological characterization of microencapsulated phase change materials (MicroPCMs) and macrocapsules containing MicroPCMs for thermal energy storage. Energy 38 (1):249-254. doi:10.1016/j.energy.2011.12.005 58. Qiu X, Li W, Song G, Chu X, Tang G (2012) Microencapsulated n-octadecane with different methylmethacrylate-based copolymer shells as phase change materials for thermal energy storage. Energy 46 (1):188-199. doi:https://doi.org/10.1016/j.energy.2012.08.037 59. Chaiyasat P, Islam MZ, Chaiyasat A (2013) Preparation of poly(divinylbenzene) microencapsulated octadecane by microsuspension polymerization: oil droplets generated by phase inversion emulsification. RSC Advances 3 (26):10202-10207. doi:10.1039/C3RA40802G 60. Qiu X, Song G, Chu X, Li X, Tang G (2013) Preparation, thermal properties and thermal reliabilities of microencapsulated n-octadecane with acrylic-based polymer shells
45 Chapter 2
for thermal energy storage. Thermochimica Acta 551 (Supplement C):136-144. doi:https://doi.org/10.1016/j.tca.2012.10.027 61. Yafei A, Yong J, Jing S, Deqing W (2007) Microencapsulation of n-hexadecane as phase change material by suspension polymerization. e-Polymers 7 (1). doi:10.1515/epoly.2007.7.1.1124 M4 - Citavi 62. Lashgari S, Arabi H, Mahdavian AR, Ambrogi V (2017) Thermal and morphological studies on novel PCM microcapsules containing n-hexadecane as the core in a flexible shell. Applied Energy 190:612-622 63. Sánchez L, Sánchez P, Carmona M, Lucas A, Rodríguez JF (2008) Influence of operation conditions on the microencapsulation of PCMs by means of suspension-like polymerization. Colloid and Polymer Science 286 (8-9):1019-1027. doi:10.1007/s00396- 008-1864-4 64. Denisov ET, Denisova TG, Pokidova TS (2005) Handbook of free radical initiators. John Wiley & Sons, 65. Odian G (2004) Principles of polymerization. John Wiley & Sons, 66. Akzo Nobel Polymer Chemicals (2006) Initiators for high polymers. 2161 BTB Communication 67. You M, Zhang X, Wang J, Wang X (2009) Polyurethane foam containing microencapsulated phase-change materials with styrene–divinybenzene co-polymer shells. Journal of Materials Science 44 (12):3141-3147. doi:10.1007/s10853-009-3418-7 68. Li W, Song G, Tang G, Chu X, Ma S, Liu C (2011) Morphology, structure and thermal stability of microencapsulated phase change material with copolymer shell. Energy 36 (2):785-791. doi:https://doi.org/10.1016/j.energy.2010.12.041 69. Wang H, Wang JP, Wang X, Li W, Zhang X (2013) Preparation and Properties of Microencapsulated Phase Change Materials Containing Two-Phase Core Materials. Industrial & Engineering Chemistry Research 52 (41):14706-14712. doi:10.1021/ie401055r 70. Qiu X, Song G, Chu X, Li X, Tang G (2013) Microencapsulated n-alkane with p(n- butyl methacrylate-co-methacrylic acid) shell as phase change materials for thermal energy storage. Solar Energy 91 (Supplement C):212-220. doi:https://doi.org/10.1016/j.solener.2013.01.022
46 State of art
71. Qiu X, Li W, Song G, Chu X, Tang G (2012) Fabrication and characterization of microencapsulated n-octadecane with different crosslinked methylmethacrylate-based polymer shells. Solar Energy Materials and Solar Cells 98 (Supplement C):283-293. doi:https://doi.org/10.1016/j.solmat.2011.11.018 72. Qiu X, Lu L, Wang J, Tang G, Song G (2014) Preparation and characterization of microencapsulated n-octadecane as phase change material with different n-butyl methacrylate-based copolymer shells. Solar Energy Materials and Solar Cells 128 (Supplement C):102-111. doi:https://doi.org/10.1016/j.solmat.2014.05.020 73. Jamekhorshid A, Sadrameli SM, Bahramian AR (2014) Process optimization and modeling of microencapsulated phase change material using response surface methodology. Applied Thermal Engineering 70 (1):183-189. doi:https://doi.org/10.1016/j.applthermaleng.2014.05.011 74. Feczkó T, Trif L, Horák D (2016) Latent heat storage by silica-coated polymer beads containing organic phase change materials. Solar Energy 132 (Supplement C):405-414. doi:https://doi.org/10.1016/j.solener.2016.03.036 75. Ng D-Q, Tseng Y-L, Shih Y-F, Lian H-Y, Yu Y-H (2017) Synthesis of novel phase change material microcapsule and its application. Polymer 133 (Supplement C):250-262. doi:https://doi.org/10.1016/j.polymer.2017.11.046 76. Barreneche C, Navarro H, Serrano S, Cabeza LF, Fernández AI (2014) New Database on Phase Change Materials for Thermal Energy Storage in Buildings to Help PCM Selection. Energy Procedia 57 (Supplement C):2408-2415. doi:https://doi.org/10.1016/j.egypro.2014.10.249 77. Giro-Paloma J, Martínez M, Cabeza LF, Fernández AI (2016) Types, methods, techniques, and applications for microencapsulated phase change materials (MPCM): A review. Renewable and Sustainable Energy Reviews 53 (Supplement C):1059-1075. doi:https://doi.org/10.1016/j.rser.2015.09.040 78. Zhu N, Ma Z, Wang S (2009) Dynamic characteristics and energy performance of buildings using phase change materials: A review. Energy Conversion and Management 50 (12):3169-3181. doi:https://doi.org/10.1016/j.enconman.2009.08.019 79. Zalba B, Marı́n JM, Cabeza LF, Mehling H (2004) Free-cooling of buildings with phase change materials. International Journal of Refrigeration 27 (8):839-849. doi:https://doi.org/10.1016/j.ijrefrig.2004.03.015
47 Chapter 2
80. Kamali S (2014) Review of free cooling system using phase change material for building. Energy and Buildings 80 (Supplement C):131-136. doi:https://doi.org/10.1016/j.enbuild.2014.05.021 81. Raj VAA, Velraj R (2010) Review on free cooling of buildings using phase change materials. Renewable and Sustainable Energy Reviews 14 (9):2819-2829. doi:https://doi.org/10.1016/j.rser.2010.07.004 82. Nagano K (2007) DEVELOPMENT OF THE PCM FLOOR SUPPLY AIR- CONDITIONING SYSTEM. In: Paksoy HÖ (ed) Thermal Energy Storage for Sustainable Energy Consumption: Fundamentals, Case Studies and Design. Springer Netherlands, Dordrecht, pp 367-373. doi:10.1007/978-1-4020-5290-3_22 83. Turnpenny JR, Etheridge DW, Reay DA (2000) Novel ventilation cooling system for reducing air conditioning in buildings.: Part I: testing and theoretical modelling. Applied Thermal Engineering 20 (11):1019-1037. doi:https://doi.org/10.1016/S1359- 4311(99)00068-X 84. Turnpenny J, Etheridge D, Reay D (2001) Novel ventilation system for reducing air conditioning in buildings. Part II: testing of prototype. Applied Thermal Engineering 21 (12):1203-1217 85. Sun Y, Wang S, Xiao F, Gao D (2013) Peak load shifting control using different cold thermal energy storage facilities in commercial buildings: A review. Energy Conversion and Management 71 (Supplement C):101-114. doi:https://doi.org/10.1016/j.enconman.2013.03.026 86. Navarro L, de Gracia A, Colclough S, Browne M, McCormack SJ, Griffiths P, Cabeza LF (2016) Thermal energy storage in building integrated thermal systems: A review. Part 1. active storage systems. Renewable Energy 88 (Supplement C):526-547. doi:https://doi.org/10.1016/j.renene.2015.11.040 87. de Gracia A, Cabeza LF (2015) Phase change materials and thermal energy storage for buildings. Energy and Buildings 103 (Supplement C):414-419. doi:https://doi.org/10.1016/j.enbuild.2015.06.007 88. Shilei L, Neng Z, Guohui F (2006) Impact of phase change wall room on indoor thermal environment in winter. Energy and Buildings 38 (1):18-24. doi:https://doi.org/10.1016/j.enbuild.2005.02.007
48 State of art
89. Kuznik F, Virgone J (2009) Experimental assessment of a phase change material for wall building use. Applied Energy 86 (10):2038-2046. doi:https://doi.org/10.1016/j.apenergy.2009.01.004 90. Mandilaras I, Stamatiadou M, Katsourinis D, Zannis G, Founti M (2013) Experimental thermal characterization of a Mediterranean residential building with PCM gypsum board walls. Building and Environment 61 (Supplement C):93-103. doi:https://doi.org/10.1016/j.buildenv.2012.12.007 91. Ascione F, Bianco N, De Masi RF, de’ Rossi F, Vanoli GP (2014) Energy refurbishment of existing buildings through the use of phase change materials: Energy savings and indoor comfort in the cooling season. Applied Energy 113 (Supplement C):990-1007. doi:https://doi.org/10.1016/j.apenergy.2013.08.045 92. Memon SA, Cui HZ, Zhang H, Xing F (2015) Utilization of macro encapsulated phase change materials for the development of thermal energy storage and structural lightweight aggregate concrete. Applied Energy 139 (Supplement C):43-55. doi:https://doi.org/10.1016/j.apenergy.2014.11.022 93. Navarro L, de Gracia A, Niall D, Castell A, Browne M, McCormack SJ, Griffiths P, Cabeza LF (2016) Thermal energy storage in building integrated thermal systems: A review. Part 2. Integration as passive system. Renewable Energy 85 (Supplement C):1334-1356. doi:https://doi.org/10.1016/j.renene.2015.06.064 94. Acciona S (2011) New Advanced Insulation Phase Change Materials. EeB PPP Project Review July 2011:26-27 95. ACCIONA Successfully Completes the NANOPCM Project. https://www.youtube.com/watch?v=keG7XKAIa1U. Accessed 17.03.2018 96. Schossig P, Henning HM, Gschwander S, Haussmann T (2005) Micro-encapsulated phase-change materials integrated into construction materials. Solar Energy Materials and Solar Cells 89 (2):297-306. doi:https://doi.org/10.1016/j.solmat.2005.01.017 97. Cabeza LF, Castellón C, Nogués M, Medrano M, Leppers R, Zubillaga O (2007) Use of microencapsulated PCM in concrete walls for energy savings. Energy and Buildings 39 (2):113-119. doi:https://doi.org/10.1016/j.enbuild.2006.03.030 98. Saffari M, de Gracia A, Ushak S, Cabeza LF (2016) Economic impact of integrating PCM as passive system in buildings using Fanger comfort model. Energy and Buildings 112 (Supplement C):159-172. doi:https://doi.org/10.1016/j.enbuild.2015.12.006
49 Chapter 2
99. Pasupathy A, Velraj R (2008) Effect of double layer phase change material in building roof for year round thermal management. Energy and Buildings 40 (3):193-203. doi:https://doi.org/10.1016/j.enbuild.2007.02.016 100. Li D, Zheng Y, Liu C, Wu G (2015) Numerical analysis on thermal performance of roof contained PCM of a single residential building. Energy Conversion and Management 100 (Supplement C):147-156. doi:https://doi.org/10.1016/j.enconman.2015.05.014 101. Construction A ACCIONA Construction installs the world’s First Smart Thermal Insulation. http://www.acciona- construccion.com/pressroom/news/2016/october/acciona-construction-installs-the- world-s-first-smart-thermal-insulation/. Accessed 20.03.2018 102. Xu X, Zhang Y, Lin K, Di H, Yang R (2005) Modeling and simulation on the thermal performance of shape-stabilized phase change material floor used in passive solar buildings. Energy and Buildings 37 (10):1084-1091. doi:https://doi.org/10.1016/j.enbuild.2004.12.016 103. Karim L, Barbeon F, Gegout P, Bontemps A, Royon L (2014) New phase-change material components for thermal management of the light weight envelope of buildings. Energy and Buildings 68 (Part B):703-706. doi:https://doi.org/10.1016/j.enbuild.2013.08.056 104. Royon L, Karim L, Bontemps A (2014) Optimization of PCM embedded in a floor panel developed for thermal management of the lightweight envelope of buildings. Energy and Buildings 82 (Supplement C):385-390. doi:https://doi.org/10.1016/j.enbuild.2014.07.012 105. Jain L, Sharma S (2009) Phase change materials for day lighting and glazed insulation in buildings. Journal of Engineering Science and Technology 4 (3):322-327 106. Goia F, Perino M, Serra V (2013) Improving thermal comfort conditions by means of PCM glazing systems. Energy and Buildings 60 (Supplement C):442-452. doi:https://doi.org/10.1016/j.enbuild.2013.01.029 107. Goia F, Perino M, Serra V (2014) Experimental analysis of the energy performance of a full-scale PCM glazing prototype. Solar Energy 100 (Supplement C):217-233. doi:https://doi.org/10.1016/j.solener.2013.12.002
50 State of art
108. Wang Q, Zhao CY (2015) Parametric investigations of using a PCM curtain for energy efficient buildings. Energy and Buildings 94 (Supplement C):33-42. doi:https://doi.org/10.1016/j.enbuild.2015.02.024 109. Alawadhi EM (2012) Using phase change materials in window shutter to reduce the solar heat gain. Energy and Buildings 47 (Supplement C):421-429. doi:https://doi.org/10.1016/j.enbuild.2011.12.009 110. Mavrigiannaki A, Ampatzi E (2016) Latent heat storage in building elements: A systematic review on properties and contextual performance factors. Renewable and Sustainable Energy Reviews 60 (Supplement C):852-866. doi:https://doi.org/10.1016/j.rser.2016.01.115 111. Frusteri F, Leonardi V, Vasta S, Restuccia G (2005) Thermal conductivity measurement of a PCM based storage system containing carbon fibers. Applied Thermal Engineering 25 (11):1623-1633. doi:https://doi.org/10.1016/j.applthermaleng.2004.10.007 112. Fukai J, Kanou M, Kodama Y, Miyatake O (2000) Thermal conductivity enhancement of energy storage media using carbon fibers. Energy Conversion and Management 41 (14):1543-1556. doi:https://doi.org/10.1016/S0196-8904(99)00166-1 113. Fan L-W, Fang X, Wang X, Zeng Y, Xiao Y-Q, Yu Z-T, Xu X, Hu Y-C, Cen K-F (2013) Effects of various carbon nanofillers on the thermal conductivity and energy storage properties of paraffin-based nanocomposite phase change materials. Applied Energy 110 (Supplement C):163-172. doi:https://doi.org/10.1016/j.apenergy.2013.04.043 114. Py X, Olives R, Mauran S (2001) Paraffin/porous-graphite-matrix composite as a high and constant power thermal storage material. International Journal of Heat and Mass Transfer 44 (14):2727-2737. doi:https://doi.org/10.1016/S0017-9310(00)00309-4 115. Pincemin S, Py X, Olives R, Christ M, Oettinger O (2007) Elaboration of Conductive Thermal Storage Composites Made of Phase Change Materials and Graphite for Solar Plant. Journal of Solar Energy Engineering 130 (1):011005-011005-011005. doi:10.1115/1.2804620 116. Pincemin S, Olives R, Py X, Christ M (2008) Highly conductive composites made of phase change materials and graphite for thermal storage. Solar Energy Materials and Solar Cells 92 (6):603-613. doi:https://doi.org/10.1016/j.solmat.2007.11.010
51 Chapter 2
117. Sarı A, Karaipekli A (2007) Thermal conductivity and latent heat thermal energy storage characteristics of paraffin/expanded graphite composite as phase change material. Applied Thermal Engineering 27 (8):1271-1277. doi:https://doi.org/10.1016/j.applthermaleng.2006.11.004 118. Wang J, Xie H, Guo Z, Guan L, Li Y (2014) Improved thermal properties of paraffin wax by the addition of TiO2 nanoparticles. Applied Thermal Engineering 73 (2):1541- 1547. doi:https://doi.org/10.1016/j.applthermaleng.2014.05.078 119. Biswas DR (1977) Thermal energy storage using sodium sulfate decahydrate and water. Solar Energy 19 (1):99-100 120. Veerakumar C, Sreekumar A (2016) Phase change material based cold thermal energy storage: Materials, techniques and applications – A review. International Journal of Refrigeration 67 (Supplement C):271-289. doi:https://doi.org/10.1016/j.ijrefrig.2015.12.005 121. McLaggan MS, Hadden RM, Gillie M (2017) Flammability assessment of phase change material wall lining and insulation materials with different weight fractions. Energy and Buildings 153 (Supplement C):439-447. doi:https://doi.org/10.1016/j.enbuild.2017.08.012 122. Laoutid F, Bonnaud L, Alexandre M, Lopez-Cuesta JM, Dubois P (2009) New prospects in flame retardant polymer materials: From fundamentals to nanocomposites. Materials Science and Engineering: R: Reports 63 (3):100-125. doi:https://doi.org/10.1016/j.mser.2008.09.002 123. Beard A (2007) Flame Retardants Frequently Asked Questions. European Flame Retardants Association:15 124. Hart RL, Work DE (1995) Flame resistant microencapsulated phase change materials. 125. Sittisart P, Farid MM (2011) Fire retardants for phase change materials. Applied Energy 88 (9):3140-3145. doi:https://doi.org/10.1016/j.apenergy.2011.02.005 126. Cai Y, Wei Q, Huang F, Gao W (2008) Preparation and properties studies of halogen-free flame retardant form-stable phase change materials based on paraffin/high density polyethylene composites. Applied Energy 85 (8):765-775. doi:https://doi.org/10.1016/j.apenergy.2007.10.017
52 State of art
127. Pielichowska K, Pielichowski K (2014) Phase change materials for thermal energy storage. Progress in Materials Science 65 (Supplement C):67-123. doi:https://doi.org/10.1016/j.pmatsci.2014.03.005 128. Zhang P, Hu Y, Song L, Lu H, Wang J, Liu Q (2009) Synergistic effect of iron and intumescent flame retardant on shape-stabilized phase change material. Thermochimica Acta 487 (1):74-79. doi:https://doi.org/10.1016/j.tca.2009.01.006 129. Zhang P, Hu Y, Song L, Ni J, Xing W, Wang J (2010) Effect of expanded graphite on properties of high-density polyethylene/paraffin composite with intumescent flame retardant as a shape-stabilized phase change material. Solar Energy Materials and Solar Cells 94 (2):360-365. doi:https://doi.org/10.1016/j.solmat.2009.10.014 130. Wang J, Wang Y, Yang R (2015) Flame retardance property of shape-stabilized phase change materials. Solar Energy Materials and Solar Cells 140 (Supplement C):439- 445. doi:https://doi.org/10.1016/j.solmat.2015.05.007 131. Fang G, Li H, Chen Z, Liu X (2011) Preparation and properties of palmitic acid/SiO2 composites with flame retardant as thermal energy storage materials. Solar Energy Materials and Solar Cells 95 (7):1875-1881. doi:https://doi.org/10.1016/j.solmat.2011.02.010
53
3. Materials and methods
CHAPTER 3 MATERIALS AND METHODS
Materials and methods
3.1. Materials
Chemical compounds utilized in each process as well as their purity and supplier are presented in Table 3.1.
Table 3.1. List of the chemical substances employed in each process.
Reagent Purity Supplier Synthesis of the microcapsules
Divinylbenzene (C10H10) 80% isomers Sigma-Aldrich
Styrene (C8H8) 99% Sigma-Aldrich Hexa(methacryloylethylenedioxy) - - cyclotriphosphazene (P3N3O6(C6O2H11)6) Rubitherm®RT31 - Rubitherm GmbH Rubitherm®RT27 - Rubitherm GmbH Rubitherm®RT21 - Rubitherm GmbH Rubitherm®RT6 - Rubitherm GmbH
Linoleic acid (C18H32O2) 60-77% Sigma-Aldrich
Oleic acid (C18H34O2) 90% Sigma-Aldrich
Erucic acid (C22H42O2) 90% Sigma-Aldrich
Palmitic acid (C16H32O2) 98% Sigma-Aldrich
Toluene (C6H5CH3) 99.99% BDH Prolabo
Benzoyl peroxide (C14H10O4) 97% Panreac Arabic gum reagent grade Sigma-Aldrich Polyvinylalcohol analytical grade Sigma-Aldrich –1 (10-98, Mw 61000 g mol ) ((C2H4O)x) Polyvinylpyrrolidone reagent grade Sigma-Aldrich, –1 (K30, Mw 40000 g mol ) (C6H9NO)n Polyvinylpyrrolidone reagent grade Chem-Impex –1 (Mw 40000 g mol ) (C6H9NO)n
Sodium dodecylsulfate (NaC12H25SO4) pharma grade Panreac
55 Chapter 3
Table 3.1. List of the chemical substances employed in each process (continuation).
Synthesis of the hexa(methacryloylethylenedioxy) cyclotriphosphazene
Phosphonitrilic chloride trimer ((NPCl2)3) 99% Sigma-Aldrich
2-Hydroxyethyl methacrylate (C6H10O3) 98% Sigma-Aldrich
Pyridine (C5H5N) 99.5% Merck KGaA Preparation of Portland Cement Concrete Portland cement II - Norcem Fly ash Class F Norcem Dynamon SR-N - MAPEI Sand - Gunnar Holth AS Gravel - Skolt Pukkverk AS Purification, storage and other Sodium hydroxide (NaOH) >99% Panreac
Calcium chloride anhydrous (CaCl2) 95% Panreac
Nitrogen 5.0 (N2) 99.999% Air Liquid Air synthetic 99.99% Air Liquid Ultra-pure water (1 μS cm–1) Milli-Q -
Ethanol (C2H6O) 99% Panreac
Acetone (CH3COCH3) 99.5% Panreac
Polyethylene glycol 400, HO(C2H4O)nH PS (for syntheis) Panreac
Potassium carbonate (K2CO3) 99% Fluka Sodium chloride (NaCl) 99.5% Sigma-Aldrich Hydrochloric acid 2N (HCl) - AVS Titrinorm Hexadecane 99% VWR
Diiodomethane (CH2I2) 99% Sigma-Aldrich
Ethylene glycol (C2H6O2) 99.8% Sigma-Aldrich
The monomers, styrene and divinylbenzene were purified by washing with an aqueous sodium hydroxide solution (1.25 N) with calcium chloride as a desiccant. Vacuum distillation was applied to remove the inhibitor from 2-hydroxyethyl methacrylate. The remaining reagents were used as received, without further purification.
56 Materials and methods
3.2. Experimental setups
Synthesis of the microcapsules
a) Continuous phase lab scale
Microcapsules were prepared by a suspension-like polymerization technique. The experimental setup is shown in Figure. 3.1. It includes 0.5 L borosilicate glass jacketed reactor heated by a thermostatic bath with circulation (Ultraterm-200, SELECTA) containing polyethylene glycol 400 as the heating oil, which is a suitable liquid in the temperature range from room temperature to 200 °C. The reactor is equipped with a reflux condenser, a nitrogen gas inlet tube and a digital overhead stirrer (Heidolph RZR 2021) with the ability of agitation speeds between 40 and 2000 rpm.
Figure 3.1. Experimental setup for the synthesis of the microcapsules: (a) general view; (b) schematic diagram: (1) reactor vessel, (2) reactor outlet, (3) thermostatic bath, (4) reflux condenser, (5) nitrogen gas inlet tube (6) gas scrubber, (7) overhead stirrer (8) six-bladed Rushton turbine impeller.
The reflux condenser is associated with a cool water intake to avoid loss of reagents. The nitrogen gas inlet tube is connected to a gas scrubber (which indicates the nitrogen supply to the reaction medium, necessary to maintain an inert atmosphere). The agitator powers a six-bladed Rushton turbine impeller, was built according to the
57 Chapter 3
standards [1]. The stirring system is sealed using a stirrer bearing SVL® made of Teflon. The reactor outlet is located at the bottom of the reactor vessel allowing discharge of the reaction product. In order to ensure safe handling of the chemicals used in the reactions, the polymerization system was isolated by a fume hood.
b) Continuous phase pilot plant
The pilot plant scale setup, shown in Figure 3.2, was employed with the purpose of enlarge the production of microcapsules. This plant is composed of a reactor with a capacity of 100 L, which is geometrically proportional to that used on the laboratory scale, equipped with digital stirring rate and temperature control.
Figure 3.2. View of the pilot plant scale reactor.
The dimensions of the Rushton turbine impeller were based on the standard stirrer configuration reported by Shäfer et al. [1] for continuous stirred tank reactors. A schematic diagram of the reaction vessel and dimensions of the impeller are shown in Figure 3.3. In addition, dimensions of the design parameters are summarized in Table 3.2 for lab and pilot plant scales.
58 Materials and methods
d2
d1/D =1/3 H/D = 1/1
h/d1 = 1
W/d1 = 1/5 H L/d1 = 1/4
W d2/d1 = 3/4 d1 h
L D
Figure 3.3. Configuration of the reaction tank and impeller [2]. D: reactor diameter, d1: impeller diameter, d2: central disk diameter, W: blade width, L: blade length, H: liquid level, h: distance between the tank bottom and the impeller.
Table 3.2. Design parameters for laboratory and pilot scale vessels (dimensions in mm).
Parameter 0.5 L reactor 100 L reactor D 87.00 430.00
d1 29.00 143.00
d2 21.75 107.00 W 5.80 29.00 L 7.25 36.00
c) Discontinuous phase container for lab scale
The discontinuous phase was preheated in a 0.5 L Pyrex container shown in Figure 3.4. The container has a peripheral mouth at the top for the addition of the raw materials and an inlet valve for nitrogen supply in order to ensure inert reaction environment. Furthermore, it has an outlet valve at the bottom to control the addition of the discontinuous phase into the reactor. This phase was stirred by a magnetic stirrer (3 cm length) and temperature controlled with a hot plate stirrer Agimatic-ED (SELECTA), which is able to heat up to 350 ºC and works in a range from 60 to 1600 rpm.
59 Chapter 3
(2) (1)
(3)
Figure 3.4. Pyrex container for the preparation of the discontinuous phase: (1) reagent addition,
(2) N2 inlet, (3) solution outlet.
d) Discontinuous phase reactor for pilot plant
The discontinuous phase in pilot plant was preheated in a 60 L stainless steel reactor shown in Figure 3.5.
Figure 3.5. Stainless steel 60 L reactor for the preparation of the discontinuous phase.
The reactor is geometrically proportional to that used for the continuous phase. It has Rushton impellers and side impellers. Nitrogen is delivered through an inlet value located in the lid. A reflux condenser is added to avoid loss of reagents.
60 Materials and methods
e) Purification of the synthesized microcapsules
Depending on the production scale, the purification of the resultant microcapsules was carried out by using vacuum filtration and centrifugation for lab and pilot plant, respectively. . Vacuum filter system For laboratory scale production, vacuum filtration was used. The filtration system consists of a Buchner funnel, a 1 L vacuum filtration flask, a filter paper (Whatman), an adjoining cone, and an electrical vacuum pump.
Microcapsules together with Milli-Q water are fed in the funnel and separation of the beads and the liquid residue is achieved due to the vacuum. Clean microcapsules are collected in the filters, and the waste leaves the funnel through the bottom outlet to the flask.
. Filtering centrifuge In pilot plant scale production, a RINA model (from Riera Nadeu) filtering centrifuge (Figure 3.6) was utilized. It is powered up by an electric motor of 50 Hz with speed rate of 3000 rpm, and a frequency inverter Altivar 31A.
The centrifuge is fed from the top and the liquid waste is collected through the bottom of the housing. The removable filtering bag in which microcapsules are retained has a permeability of 27 L m‒2 h‒1.
(4)
(3) (2) (1)
(5)
Figure 3.6. Filtering centrifuge system. (1) Filtering centrifuge, (2) electric motor, (3) frequency inverter, (4) suspension inlet, (5) residue outlet.
61 Chapter 3
Bulk polymerization
Bulk polymerization was carried out in a 100 ml plastic container with a lid, equipped with a magnetic stirrer and temperature adjustment controlled by a hot plate stirrer Agimatic-ED (SELECTA). The plastic container was immersed in the oil bath. Nitrogen was delivered through a hole in the lid.
Thermal properties
a) Guarded hot plates method
In order to characterize the thermal performance of the concrete samples with microencapsulated phase change material (MPCM), the guarded hot plates method was utilized. The test is based on recording of temperature variations and heat fluxes exchanged through the sample. The guarded hot plates system is presented in Figure 3.7. The experimental setup includes two aluminum plate heat exchangers connected to thermal regulated baths. The concrete sample is sandwiched between two aluminum plate heat exchangers and it is insulated by a 40 mm thick polyethylene expanded foam. This insulated cover will decrease the heat transfer from the lateral side face of the sample into the surrounding environment. Therefore, the heat transfer through the concrete sample can be calculated assuming one dimensional thermal conditions.
Figure 3.7. Guarded hot plates system.
62 Materials and methods
Heat flux sensors (Captecv, France) and K-type thermocouples (TC Ltd., UK) are introduced on both sides of the sample to measure the temperature fluctuations and heat fluxes through sample. The data are recorded by a multichannel multimeter (LR8410-20 Hioki, Japan) to which all sensors are connected.
b) Environmental chamber
Figure 3.8 illustrates the thermal testing system which was used to investigate the influence of MPCM on the thermal properties of Portland cement concrete. An environmental chamber (VT³ 4250, Vötsch, Germany) and Laird temperature regulator (AA150-Laird Technologies) were employed to simulate the ambient outdoor temperature Tout (t) and the temperature inside the test box (Troom). A small test box with inner dimensions of 600 × 800 × 600 mm was made of 50 mm panels of polyethylene expanded foam. It is placed inside an environmental chamber in a rectangular hole (200 × 200 mm) at the middle of the top insulation panel.
Figure 3.8. The thermal performance testing system and sketch of cross-section of system.
3.3. Experimental procedures
Polymerization of the microcapsules
Microencapsulation was performed by a suspension-like polymerization technique using the experimental setup shown in Figure 3.1. The synthesis process includes two phases: a continuous phase containing water and the suspending agents and a discontinuous phase containing the monomers, core material, initiator and diluent.
63 Chapter 3
Initially, the continuous phase is transferred to the reactor (in established proportions) at a fixed temperature and agitation that allow us to obtain the desired microcapsules. Additionally, the water shut-off valve and the nitrogen valve are opened to ensure the condensation of the up flowing vapors and an inert atmosphere in the system. When the temperature is stabilized, the discontinuous phase is prepared in the glass vessel shown in Figure 3.4. The phase change material, monomers, diluent and initiator are premixed at 250 rpm and heated to 50 ºC, allowing an efficient operation and minimizing the idle time. The discontinuous phase is then transferred into the continuous phase and maintained under vigorous agitation at a constant temperature. When the reaction is completed, stirring and heating are stopped and the water and nitrogen valves are closed. The product is subsequently discharged into a glass beaker, separated from liquid by filtration and purified by repeated washing with ethanol (using the purification systems explained in the section 3.2.1 preceding experimental procedure) and finally dried at room temperature for at least 24 h.
Polymerization in mass
Polymerization in mass were performed at 80 °C under nitrogen atmosphere. Monomers and initiator are weighted and transferred to the plastic container. The container is immersed in an oil bath, previously heated to the required temperature. The total reaction time is 5 hours.
Synthesis of PNC-HEMA
The reaction is prepared in a polymerization equipment shown in Figure 3.1. The synthesis consists of three stages, and it was prepared according to Anzai et al. [3]. First, the established amount of the phosphonitrile chloride trimer (PNC) is mixed with the solvent and introduced into the reactor. The stirring is regulated at a speed of 400 rpm and a temperature of 50 °C. Secondly, the required amounts of the monomer 2- hydroxyethyl methacrylate (HEMA) is added together with the solvent to the same reactor. Finally, the catalyst is dissolved in the solvent and added dropwise to the reactor with two other solutions.
64 Materials and methods
The reaction is carried out for 5 days to ensure that the 6 chlorine atoms of the PNC can be substituted by the HEMA. Subsequently, the product is cooled to aid in the filtration of the formed salts. Afterwards, a series of washes of the obtained product is performed. This operation is carried out in a separatory funnel in order to divide the two immiscible phases that are formed. Three types of washing procedures are carried out: the first utilizes 50 ml of 2N HCl, the second a 5% solution of K2CO3 and the last one uses a 5% solution of NaCl. To eliminate the amounts of solvent, catalyst and water that the flame retardant may contain after washing, a rotary evaporator is used to obtain the final PNC-HEMA product.
Concrete preparation
Initially, all components of Portland cement concrete (PCC) are weighted using a digital balance (BERGMAN) with an accuracy of 0.1 g. Portland cement, sand and gravel are mixed together for 2 min. Afterwards, water with admixture are partially added and mixed for 1 min. Finally, the MPCMs are added to the concrete mixture and mix next 2 minutes. The MPCM is added as the last component in order to avoid the damage of the MPCM during the mixing process. After mixing, PCC is casted into molds at a size of 10 × 10 × 10 cm for mechanical test and 20 × 20 × 2.53 cm for the thermal test. Half-filled molds are compacted by means of a steel pestle (25 times) and the same procedure was repeated after the molds were filled all the way up. After casting, concrete samples are pre-cured at ambient temperature with a relatively humidity of 90% for 24 h, then demolded and cured in water at either 20 °C or at 40 °C for 1, 7, 14, and 28 days.
3.4. Characterization techniques and characteristic parameters
Differential scanning calorimetry
Measurements of melting point and latent heat storage capacities of different materials were performed in a differential scanning calorimetry (DSC) model DSC Q100 from TA Instruments (Figure 3.9), equipped with a refrigerated cooling system and nitrogen as the purge gas. Measurements were carried out in the temperature range from −40 to 80 °C with a heating and cooling rate of 3 °C min-1.
65 Chapter 3
Figure 3.9. Differential scanning calorimeter.
Thermogravimetric analysis
a) Ramp studies
The thermal stability, amount of PCM and toluene content of the synthesized microcapsules were obtained by using a TA instruments SDT Q600 Simultaneous DSC- TGA from room temperature to 600 °C at a heating rate of 10 °C min-1 under a nitrogen or air atmosphere. This equipment presents a calorimetric precision of ±2 % (based on metal standards) and a balance sensitivity of 0.1 μg.
b) Stepwise studies
Measurements were carried out using the same TA instruments SDT Q600 Simultaneous DSC-TGA (Figure 3.10). Samples were placed in tared platinum pan (internal diameter of 6 mm and internal height of 3.9 mm). The substances were melted and mixed before analysis in order to ensure homogeneity. The crucible was filled with 10-15 mg of material. In order to obtain the same and known sample surface area, the liquids were introduced in small drops in the center of crucible. Measurements were conducted under nitrogen flow rate 100 ml min−1. Isothermal conditions were kept
66 Materials and methods constant during 10 min in each stepwise for a temperature range from 323 to 523 K. The analyses were triplicated for each material.
Figure 3.10. Thermogravimetric balance.
Ash determination
Ash determination was performed based on ISO 3451-1. The crucibles were heated it in the muffle furnace at the test temperature until constant mass, cooled in the desiccator to room temperature and weighed on the analytical balance. 2g of sample was introduced into the crucible. The crucible was heated directly on the burner to burn slowly until volatile products have been driven off. The crucibles were placed into the muffle furnace preheated to the 600 °C and calcined for 1h. After that, the crucibles were cooled in the desiccator until room temperature was reached and weighed on the analytical balance.
Density of microcapsules
The densities of microcapsules were determined by helium pycnometer (Micromeritics Accupyc 1330). Samples amount between 0.6 and 1.0 g were used. Each analysis was performed three times and average values will be reported.
67 Chapter 3
Porosity
The porous structure was analyzed by adsorption–desorption N2 isotherms at 77 K in an automatic Micromeritics Asap 2020 and was determined by the Brunauer– Emmett–Teller (BET) method. The porosity was calculated using density of microcapsules (ρMC) and pore volume (Vpore).
푉푝표푟푒푠 푃표푟표푠𝑖푡푦 (%) = × 100% 1 (3.1) ( + 푉푝표푟푒푠) 휌푀퐶
Environmental scanning electron microscopy and energy dispersive x-ray spectrometry
The morphology and the surface features of the microcapsules were observed by using Quanta 250 (FEI Company) with a tungsten filament operating at a working potential 12.5 kV or 15 kV (Figure 3.11). The Secondary Electron detector (LFD - Large Field Detector) and Backscattered Electron Detector (BSED) were applied for imaging. The electron beam resolution at this vacuum mode is 2.0-3.0 nm at 30 kV. The scanning electron microscope is equipped with an EDAX Apollo X (AMETEK), an energy dispersive x-ray spectrometer (EDX), which analyze the chemical composition of the samples with the detection limits about 1000 ppm or 0.1 wt%. The heat treated test of microcapsules were performed in situ from ambient temperature (20 °C) to 250 °C at heating rate 3 °C min-1 and pressure 60 Pa.
68 Materials and methods
Figure 3.11. Scanning electron microscope.
Low angle laser light scattering
Volume average particle size (dv0.5) and number average particle size (dn0.5) as well as particles size distribution (PSD) of the microcapsules were determined by Low Angel Laser Light Scattering (LALLS) laser diffraction, utilizing a Malvern Mastersizer 2000 equipped with a Scirocco 2000 unit (Figure 3.12) for analyzing dispersions of the particles in air and a software that uses the Mie theory to analyze the experimental data.
It is important to notice that dv0.5 and dn0.5 represent median values, which are defined as the value that splits the distribution with half above and half below this diameter.
Figure 3.12. Low Angel Laser Light Scattering laser diffraction.
69 Chapter 3
Sieving
The percentage of microcapsules with size larger than 2000 μm were obtained by sieving the synthesized product using CISA test sieves of different aperture mesh (from 250 to 2000 μm).
Surface tension and contact angle
The surface tension and interfacial tension of water solutions and PCM were determined using a Theta optical tensiometer (Attension, Espoo, Finland).
Figure 3.13. Theta optical tensiometer.
Water phase is named as Wy, where y correspond to H2O or the suspending agent used. Oil phase is called ORT27, and polymer phase is called as Pz where z means polymer composition. The pendant drop method and reversed pendant drop method were utilized for surface tension and interfacial tension measurements, respectively. The measurements were performed at 30 °C in order to keep PCM in a molten state. The drop shapes were fitted with the Young–Laplace equation (equation 3.2) [4]. Surface or interfacial tensions were measured directly from the pendant drop shape analysis. They can be related to the drop shape by:
∆휌 푔(푅2) 훾 = 0 (3.2) 푖 훽 where γi are surface tensions of the pure materials, for the three phases, polymer (p), oil
(o) and water (w), Δρ is density difference between fluids, g is gravitational constant, R0 is the radius of the drop curvature at vertex and β is the shape factor.
70 Materials and methods
A contact angle () and consequently surface free energy (γp) of the polymer were measured by the sessile drop method. Owens, Wendt, Rabel and Kaelble (OWRK)/Fowkes model was used to perform calculations of the surface free energy based on the two-component surface energy theory for the polymer with respect to each one of the liquids, oil or water phase [5]. Equation 3.3 shows the formula for the liquid o although it can be extended for the liquid w:
퐷 퐷 1/2 푃 푃 1/2 (훾푝 훾표 ) + (훾푝 훾표 ) = 훾표(푐표푠휃 + 1)/2 (3.3) where D and P are dispersive and polar fractions, respectively. Water and diiodomethane were employed as standards for surface free energy measurement [6].
A polymer film was prepared in situ onto a glass slide. The glass slide was cleaned with ethanol and immersed in the solution consisting of monomers mixed with initiator. The slide was wetted by the solution for 5 min and placed in an oven at 80 °C. Before deposition of the drops, impurities from the polymer surface were removed by ethanol. The drop deposition onto the polymer film is sketched in Figure 3.14.
Figure 3.14. Schematic of a liquid drop on the sample surface [5].
The interfacial tension between water/polymer or oil/polymer can be calculated based on
Young’s equation (equation 3.4) where γop or γpw, γp, γo, γw and correspond to interfacial tension between liquids (oil, water phase) and polymer, surface free energy of polymer, surface tension of liquid and contact angle between liquid and solid, respectively:
훾표푝 = 훾푝 − 훾표푐표푠휃 (3.4a)
훾푝푤 = 훾푝 − 훾푤푐표푠휃 (3.4b)
For each combination at least three measurements were made.
Surface polarities for liquids and solids were calculated using equation 3.5 and 3.6, respectively.
푃 % 푠푢푟푓푎푐푒 푝표푙푎푟𝑖푡푦 (표𝑖푙) = 100 ∗ 훾표 /훾표 (3.5a)
71 Chapter 3
푃 % 푠푢푟푓푎푐푒 푝표푙푎푟𝑖푡푦 (푤푎푡푒푟) = 100 ∗ 훾푤/훾푤 (3.5b) 푃 % 푠푢푟푓푎푐푒 푝표푙푎푟𝑖푡푦 (푝표푙푦푚푒푟) = 100 ∗ 훾푝 /훾푝 (3.6)
Gel permeation chromatography
Gel permeation chromatography (GPC) was used to determine the molecular weight distribution of the oligomers and polymers. Measurements were performed with a Viscotek chromatograph (Figure 3.15), having refractive index detector and two columns (Styragel HR2 and Styragel HR0.5) at 35 °C at a flowrate of 1 mL min-1 and using tetrahydrofuran (THF) as eluent. The samples were dissolved in THF at a concentration of 10 mg mL-1 for 24h, and the dissolved part of the samples was analyzed.
Figure 3.16. Gel permeation chromatograph.
Gas chromatography – mass spectrum analysis
The molecular weight of Rubitherm®RT6 was analyzed by GC-MS using a Thermo Scientific DSQ II Series Single Quadrupole GC-MS (Figure 3.16) with a NIST05-MS library. The column was a HP-5MS UI. The temperature ramp was 40 °C for 5 min, 15 °C min−1 up to 200 °C, hold time 5 min and then 15 °C min−1 up to 250 °C and hold time 5 min. The inlet, source and transfer line temperatures were 200, 200 and 300 °C, respectively.
72 Materials and methods
Figure 3.16. Gas chromatograph with mass spectroscopy.
Fourier transform Infrared spectroscopy
Fourier transform infrared spectroscopy (FT-IR) was utilized to identify the characteristic functional groups present in the raw materials and final products. Spectra were obtained on a Varian 640-IR FT-IR spectrometer (Figure 3.17) equipped with an Attenuated Total Reflectance (ATR) accessory (with a wavelength accuracy and precision >0.01 cm and >0.005 cm, respectively). All infrared spectra were collected using 16 scans and 8 cm‒1 resolution in the wavelength range of 4000 to 500 cm-1.
Figure 3.17. Infrared spectrometer.
73 Chapter 3
X-ray micro-tomography
The internal microstructure of Portland cement concrete with and without microcapsules were investigated using X-ray tomography images. The X-ray micro- tomography images from cross-section of samples were obtained using a Skyscan 1172 CT scanner from Bruker (Figure 3.18) with 85 kV incident radiation, 400 ms exposure time per frame and 0.5° rotation step. The samples after 28 days of curing were prepared in cylindrical form of 1 cm diameter and 1 cm height.
Figure 3.18. X-ray micro-tomography scanner.
Micro-combustion calorimetry
Micro-combustion calorimetry (MCC) was performed in FTT Micro Calorimeter (Fire Testing Technology Ltd., United Kingdom) as shown in Figure 3.19. Approximately 8 to 14 mg of each sample was weighed with an analytical balance and placed in a MCC for rapid pyrolysis. A sample was placed into a heated interface continuously purged with pure nitrogen. The system was programmed to heat 1 °C s-1 from 75 to 600 °C. After pyrolysis, the volatilized decomposition products were transferred to a combustion furnace set at 900 °C where pure oxygen was added and the decomposition products were completely combusted. The amount of consumed oxygen was measured with an oxygen analyzer and used to calculate a heat release rate (HRR).
74 Materials and methods
Figure 3.19. Micro-combustion calorimeter.
Cone calorimetry
In order to determine the influence of MPCM on the flammability of the Portland cement concrete (PCC), cone calorimetry was used. Cone calorimeter supplied by Fire Testing Technology Ltd., United Kingdom (Figure 3.20) was used. The specimens made of Portland cement concrete without and with 10% of MPCM and microcapsules with flame retardant (MPCM-FR) as a replacement of sand were exposed to heat flux of 50 kW m-2, corresponding to 750 °C at the cone surface. The tests were performed according to IS0 5660. The samples with the size 10 x10 cm were analyzed in a horizontal position, with the exhaust duct flow rate set at 0.025 m3 s-1. The duration of the tests was set until 2 min after any flaming or combustion signs ceased or until 60 min elapsed. Before testing, the samples were wrapped in a single layer of aluminum foil, covering the unexposed surfaces with the shiny side towards the specimen. For each concrete mix, three specimens were tested.
75 Chapter 3
Figure 3.20. Cone calorimeter.
Compressive strength test
The compressive strength tests were conducted according to standard EN 12390- 3. The compressive strength of concrete samples was determined using digital compressive strength test machine (Form + Test Machine) with compression capacity of 3000 kN. Each test cube was exposed to a force at a loading rate of 0.8 kN/s until it failed. The compressive strength tests were carried out at 20 °C and 40 °C for PCC samples containing 0, 10 and 20% of MPCM at curing times of 1, 7, 14, and 28 days. For the measurements at 20 °C, three specimens were left in the room for 1 h in order to remove free water from the surfaces, before they were weighed and tested. In order to perform the measurements at 40 °C, the compressive strength machine was isolated thermally and connected to a heating chamber by means of an isolated tube to keep the temperature of the machine constant at 40 °C. Before the test, three samples were kept in a heating chamber at 40 °C for 1 h to remove free water from the surfaces while keeping the temperature of the cubes constant, and immediately afterward the cubes were weighed and tested. The reported results at both temperatures are the average of the three measurements.
76 Materials and methods
Thermal conductivity
The thermal conductivity of the concrete samples was determined at 20 and 32 °C thus, below and above melting range of PCM. They are denoted as solid thermal conductivity, kS when measurement was performed below melting point and liquid thermal conductivity, kL when measurement was performed above melting point. First, both aluminum plate heat exchangers (as presented in section 3.2.3) were kept at a constant temperature Tinit until the heat fluxes were established and thermal steady-state condition was reached. Then, a temperature variation was imposed on the top aluminum plate heat exchanger from Tinit to Tend, kept at Tend while the other aluminum plate heat exchanger was kept at Tinit and until a thermal steady state was reached. Finally, the average temperature on the top (Ttop) and bottom (Tbottom) faces of the block and the average heat fluxes (φave) on both faces were recorded and thermal conductivity (k) was calculated via the following relationship:
휑푎푣푒푑 푘 = (3.7) 퐴(푇푡표푝 − 푇푏표푡푡표푚) where A and d are the area and the thickness of the concrete block, respectively. In the experiments the dimension of the concrete specimens were A = 400 cm2 and d = 2.53 ±
0.02 cm. For thermal conductivity in solid state, Tinit and Tend were set at 5 and 10 °C, respectively. Whereas in order to calculate the thermal conductivity in the liquid state, values of Tinit and Tend of 45 and 50 °C were utilized.
Specific heat capacity/latent heat
The latent heat and the specific heat capacity of the concrete samples were determined by the testing system shown in Figure 3.8. The sample was initially isothermal -1 at Tinit and then heated by increasing the temperature 10 °C h of both aluminum plate heat exchangers from Tinit to Tend using oil thermostatic baths. Tinit and Tend were set as 5
°C and 45 °C, respectively. The average heat fluxes (φave) and temperature on both faces of concrete specimens (Ttop and Tbottom) were determined via heat flux sensors and thermocouples, respectively. The solid specific heat capacity, Cp-solid and the liquid specific heat capacity, Cp-liquid were measured in the temperature range of 10–15 °C and 35–40 °C, respectively. The specific heat is given by:
77 Chapter 3
퐴휑 퐶 = 푎푣푒 푝 푑푇 (3.8) 푚 푑푡 where Cp and m are specific heat capacity and the mass of sample, respectively.
Energy saving aspects
In order to evaluate the energy saving possibility, thermal system shown in Figure
3.9 and described in section 3.2.3b was employed. Room temperature (Troom) was set constant at 23 °C throughout the experiment while the outdoor temperature Tout (t) was imposed as a sinusoidal function of time (equation 3.9). The maximum (Tmax) and minimum (Tmin) outdoor temperatures during a day were 40 ºC and 10 ºC, respectively.
The maximum outdoor temperature Tmax was set at 14:00.
푇 + 푇 푇 − 푇 휋 2휋 푇 (푡) = 푚푎푥 푚푖푛 + 푚푎푥 푚푖푛 sin ( 푡 − ) (3.9) 표푢푡 2 2 43200 3
Initially, indoor and outdoor temperatures were set at 23 ºC for 2 hours to achieve a steady-state condition. Thereafter, the outdoor temperature cycles were run continuously for 72 hours. The temperature was measured across the sample by thermocouple installed at different depths of the concrete wall in steps of 25 mm. Moreover, the indoor and outdoor temperatures were measured at different positions in the test box and environmental chamber. Data were saved every 60 s using a multichannel multimeter (LR8410-20 Hioki, Japan). The heat fluxes on both surfaces of the sample were recorded by heat flux sensors. It was assumed that: (i) the insulation panels of the test room are perfectly thermally insulated, (ii) the heat will be transferred to the test box via the concrete sample and will be compensated by the temperature regulator to maintain a room temperature of 23 ℃. Consequently, the energy supplied to the temperature regulator can be calculated as the total of the heating power consumption when the indoor surface temperature is lower than room temperature and the cooling power consumption when the indoor surface temperature is higher than room temperature: 24ℎ ∫ |휑푖푛푑표표푟|푑푡 푃 = 0 (3.10) 3600 ∙ 103
78 Materials and methods
where φindoor is the heat flux on the indoor side of the sample. The power reduction Pr can be counted as: 푃 − 푃 푃푟 = 푃퐶퐶 푃퐶퐶푥 ∙ 100% (3.11) 푃푃퐶퐶 where PPCC and PPCCx are the power consumption of the heating and cooling system for one working day for Portland cement concrete without MPCM and with amount of MPCM respectively.
Process parameters
a) Yield
The microcapsules yield (ηr) was determined by considering the maximum amount of the product as that constituted by the polymer from monomers P(St-DVB)MC, and encapsulated PCM (PCMMC) by:
푃퐶푀MC + P(St − DVB)MC ηr = (3.12) 푃퐶푀feed+(St − DVB)feed where PCMfeed, (St-DVB)feed are the weights of the PCM and monomers fed to the reactor, respectively.
b) Paraffin content
The paraffin content (CPCM) in the microcapsule was calculated based on the enthalpy values:
∆HMC CPCM (%) = × 100% (3.13) ∆H PCM
-1 where ΔHMC is the enthalpy for the analyzed microcapsules (J g ) and ΔHPCM is the enthalpy of pure PCM.
c) Encapsulation efficiency
The encapsulation efficiency (EE) can be calculated from the relationship between the PCM inside the total microcapsules (PCMMC) and the PCM fed (PCMfeed):
PCMMC EE (%) = × 100% (3.14) PCMfeed
79 Chapter 3
References
1. Schäfer M, Karasözen B, Uludağ Y, Yapıcı K, Uğur Ö (2005) Numerical method for optimizing stirrer configurations. Computers & Chemical Engineering 30 (2):183-190. doi:https://doi.org/10.1016/j.compchemeng.2005.08.016 2. Kysela B, #x161, Konfr, #x161, t J, #x159, #xed, Fo, #x159, t I, Ch, #xe1, ra Z, #x11b (2014) CFD Simulation of the Discharge Flow from Standard Rushton Impeller. International Journal of Chemical Engineering 2014:7. doi:10.1155/2014/706149 3. Anzai M, Ohashi M (1984) Studies on the reaction product of hexachlorocyclotriphosphazene and 2-hydroxyethyl methacrylate and on the physical properties of its polymer. The Journal of Nihon University School of Dentistry 26 (2):109-118 4. Tasker AL, Hitchcock JP, He L, Baxter EA, Biggs S, Cayre OJ (2016) The effect of surfactant chain length on the morphology of poly(methyl methacrylate) microcapsules for fragrance oil encapsulation. Journal of Colloid and Interface Science 484 (Supplement C):10-16. doi:https://doi.org/10.1016/j.jcis.2016.08.058 5. Rulison C (2000) Two-component surface energy characterization as a predictor of wettability and dispersability. KRUSS Application note AN213:1-22 6. Fernandez V, Khayet M (2015) Evaluation of the surface free energy of plant surfaces: toward standardizing the procedure. Frontiers in Plant Science 6 (510). doi:10.3389/fpls.2015.00510
80
4. Optimization of microcapsules recipe and properties
OPTIMIZATION OF MICROCAPSULES CHAPTER 4 RECIPE AND PROPERTIES
Based on the article:
. Szczotok, A. M.; Garrido, I.; Carmona, M.; Kjøniksen, A.-L.; Rodriguez, J. F., Predicting microcapsules morphology and encapsulation efficiency by combining the spreading coefficient theory and surface polarity. Submitted to the Colloids and Surfaces A: Physicochemical and Engineering Aspects.
. Szczotok, A. M.; Carmona, M.; Kjøniksen, A.-L.; Rodriguez, J. F., Equilibrium adsorption of polyvinylpyrrolidone and its role on thermoregulating microcapsules synthesis process. Colloid and Polymer Science 2017, 295 (5), 783- 792.
Optimization of microcapsules recipe and properties
Abstract
Microencapsulation of phase change materials (PCMs) with styrene (St) and divinylbenzene (DVB) shells by suspension-like polymerization has been carried out. The optimization of the process was performed by varying: suspending agents, discontinuous/continuous phase mass ratio, agitation rate, toluene/PCM mass ratio and the amount of PCM in the recipe. The influence of process variables on the surface morphology, particle size, encapsulation efficiency and the process yield has been studied. A concentration of polyvinylpyrrolidone (PVP) of 5.03% was found to be most suitable to achieve the best properties in particle size distribution, thermal energy storage, encapsulation efficiency and process yield. Finally, using the proper suspending agent, agitation rate and the mass ratios between continuous to discontinuous phases, and porogen to PCM, allowed the production of microcapsules with a high energy storage capacity (> 100 J/g), encapsulation efficiency (> 80%) and process yield (> 86%). This formulation also ensures high mechanical and thermal resistance due to the crosslinked structure of the polymer shell. The morphology and encapsulation efficiency of thermoregulating microcapsules based on the spreading coefficient theory and surface polarity were evaluated. Contact angles and interfacial tensions were measured, and the results were discussed with respect to the internal structure as well as the encapsulation efficiency of the microcapsules. Taking into account the chemical nature of the employed components, coefficient theory predicts a core-shell structure, whereas observations revealed that a matrix morphology was obtained. A core-shell structure was favored when the polarity of the polymer was increased and the compatibility between the core and shell decreased. The type of suspending agent and its proportion do not have a decisive influence on the final characteristics of microcapsules. Shells with various polarities from styrene (St), divinylbenzene (DVB) and hexa(methacryloylethylenedioxy) cyclotriphosphazene (PNC-HEMA), introduced to improve the particle fire retardancy have direct impact on the experimental morphology. Only microcapsules with higher surface polarity exhibited a core-shell structure.
The role of polyvinylpyrrolidone (PVP) in the formation of the microcapsules shell has also been studied. The results confirmed that a large amount of PVP is
83 Chapter 4
incorporated into the polymeric network that forms the shell. This incorporation of PVP into the shell can be interpreted as an adsorption process where an adsorbate is retained by an adsorbent material. Experimental data were perfectly fitted by the Langmuir model, obtaining at a confidence level of 95% and values of 192.9 ± 0.4 g/kg and 0.18 ± 0.11 m3/kg for the maximum adsorption capacity and the equilibrium constant, respectively.
84 Optimization of microcapsules recipe and properties
4.1. Background
The past several decades have seen an amazing and rapid progress in the available technologies for the microencapsulation of phase change materials (PCMs). The combination of the advantages of the PCMs and microencapsulation makes these new materials promising to overcome the limitations of the growing energy demand.
The encapsulation of phase change materials of different nature has been carried out by several authors and is summarized in Table 2.2 of the previous chapter. In this first approach, paraffins were selected because of their excellent thermal properties, easy and large availability, and low price. Furthermore, if paraffin leaks out from the microcapsules, it will not corrode the steel reinforcements in concrete. Accordingly, paraffin based microcapsules are a unique type of microcapsules that has achieved an emerging introduction in the building sector. On the other hand, several polymers have been employed as shell materials including melamine-formaldehyde [1,2], polystyrene [3-5], poly(methyl methacrylate) [6-8] and its copolymers [7,9-11]. The right choice of the shell material for the encapsulation of PCM will play a crucial role, providing the microcapsules with enough chemical and mechanical resistance to resist the aggressive mixing process in the concrete preparation.
While the incorporation of these kinds of polymeric microcapsules in “soft” building materials such as gypsum and polyurethane foams has been extensively studied, the incorporation into concrete has not been sufficiently explored. The structural function of concrete and the aggressive mixing process of fabrication require of harder microcapsules. So the physical strength of the microcapsule shell for concrete has to be improved introducing crosslinking agents into the formulation to create a stable 3-D network. For this purpose divinylbenzene (DVB) was chosen to be copolymerized with styrene (St). Poly(styrene-divinylbenzene) - P(St-DVB) exhibits a hydrophobic nature, minimizing the water adsorption and will therefore not affect the water required for hydration process in the CSH phase of concrete formation too much. Consequently, crosslinked polystyrene seems to be an interesting alternative for the microcapsules shell.
Most previous studies focus on the ability of suspending agents to modify the interfacial relations in terms of hydrophobicity, amphoteric properties, and viscosity. Nevertheless, very few of such studies have considered not only the suspending agent as
85 Chapter 4
a skin of the particle but a constitutive portion of the material formed (particles or capsules). The adsorption of nonionic surfactants onto solid surfaces has been studied previously [12-15], observing that a large amount of the suspending agent was loaded onto the solid material. Smith et al. [15] found a high affinity adsorption of PVP onto polystyrene lattices in water, where the maximum adsorption capacity was independent on the PVP molecular weight. On the contrary, when they used lattices in 0.5 N NaCl, the adsorption of PVP onto polystyrene lattices increased with the molecular weight. Geffroy et al. [14] studied the adsorption of different nonionic surfactants formed by combining alkyl groups C8 or C12 with an ethylene oxide oligomer. They found that the maximum adsorption capacity varied with the size of the polar head group. Hence, it is possible that surfactant agents could be incorporated to the microcapsules containing thermoregulating materials, and the quantity may depend on the molecular weight and polarity of the surfactants. Thus, apart of its tensioactive function this compound can be incorporated as a constitutive part of the microcapsule shell. This incorporation will increase the apparent microcapsules yield. However, there is a general lack of knowledge on the way in which the inclusion of the tensioactive agents influences not only the size or shape but also the intrinsic properties of the microparticles or microcapsules. In this chapter the applicability of PVP and toluene for producing thermoregulating microcapsules from poly(styrene-divinylbenzene) containing Rubitherm®RT27 is examined. In addition, the effect of different types of suspension stabilizers (AG, PVP, SDS and AG/PVP mixture), the influence of the concentration of the suspension agents on the product properties: thermal properties, morphology and size distribution and also on the process yield were analyzed. Additionally, a deep prospection of the influence of PVP in the properties of the final product has been done. This study has demonstrated that the PVP is incorporated into the polymeric shell as a constituent part of the matrix following an adsorption-like process. Synthesis utilizing increasing amounts of PVP were examined and fitted to the Langmuir adsorption model. Furthermore, the morphology and encapsulation efficiency of microcapsules synthesized by using additional co-monomer hexa(methacryloylethylenedioxy) cyclotriphosphazene (PNC-HEMA) with higher surface polarity and flame retardant properties has been
86 Optimization of microcapsules recipe and properties included in this chapter. However, a more extensive study about thermal and fire resistant characteristics will be provided in Chapter 6.
4.2. Results on optimization
Preliminary experiments
Taking into account previous experience in our research group, regarding the synthesis of microcapsules from poly(styrene-co-divinylbenzene) (P(St-DVB)) by means of suspension-like polymerization containing extractant agent, an organic solvent (toluene) was used as a diluent of the discontinuous phase, in order to decrease the reaction rate [16]. Initially, the microcapsules were synthesized by using the recipe developed by Alcazar et al. [16] in which the extractant agent was replaced by phase change material. AG and PVA were applied as a mixture of suspending agents. Theoretically the microcapsules should be composed of 33.17 wt.% of PCM, 54.26 wt.% of styrene and 12.57 wt.% of divinylbenzene. Previous to characterization, the product was sieved due to high agglomeration and 80.86 wt.% of the product was rejected. Thus, the final amount of microcapsules useful in following analyses became low although the initial yield of the reaction was high (82.87%). Figure 4.1 shows the SEM image of the microcapsules. It can be noticed that the remaining part of the microcapsules were spherical and smooth exhibiting a particle size around 1 mm. Nevertheless, the thermal properties are the most important parameter in synthesis of thermoregulating microcapsules. The latent heat of 29.61 J g-1 correspond to a paraffin content of only 17.30% and an encapsulation efficiency 48.34%. This is only half of the theoretically expected amount whereas, in the cited study the core material constituted 34.86% of the final product. Based on these results, the mixture of AG and PVA as suspending agents was discarded. However, this experiment prove that the encapsulation of PCM into a P(St-DVB) shell is possible using toluene as solvent. Taking into account the future application as a concrete additive, it was mandatory to find another suspending agent in order to improve dispersibility of single drops in the continuous phase. In addition, the initial amount of core material should be increased to improve the thermal energy storage. By changing the mass ratio of the monomers, a higher mechanical resistance may be achieved.
87 Chapter 4
Figure 4.1. SEM micrographs of microcapsules synthesized by using AG:PVA as suspending agent.
In order to optimize the thermal and physical properties of the microcapsules and the process yield, a new series of experiments were performed. The experiments are shown in Table 4.1, starting from the studies determining the most suitable suspending agent and ending by finding the optimal suspending agent concentration within 0.87 and 14.99% in total mass.
88 Optimization of microcapsules recipe and properties
Table 4.1. Experimental conditions used for obtaining microcapsules with Rubitherm®RT27.
-
0 5
10 20
PNC
HEMA (%)
1.86 1.73 0.87 3.41 6.59 8.10 5.03
14.99
agent (%)
Suspending
33 50 68
(%)
PCM PCM
3.00 1.92
Tol/PCM Tol/PCM
mass mass ratio
rate 400 800
1200
Agitation
discont.
5.55 3.24
13.69 11.09
Cont./
phase mass ratio
AG
type
SDS PVP
AG:PVP
AG:PVA
Suspending agent
1.73)
-
DVB)
-
HEMA) HEMA)
HEMA)
- -
-
P(St
-
MC(PVP
0.87) 3.41) 6.59) 8.10)
14.99)
- - - -
-
MC
5%PNC
10%PNC 20%PNC
-
MC(3.00 and50%)
- -
Study
MC(AG)
MC(5.55)
MC(SDS)
5.03)/
Alcazar et al.
DVB
-
MC(AG:PVP)
DVB DVB
MC(1200 rpm) -
MC(PVP MC(PVP MC(PVP MC(PVP
- -
MC(PVP
MC(PVP)/ MC(11.09)
P(St
P(St P(St
MC(3.24)/ MC(400 rpm)
-
- -
MC(PVP
MC
MC MC
MC(800 rpm)/
MC(1.92 and 68%)/
89 Chapter 4
Theoretical framework
The shell composition has a great influence on the morphology during microencapsulation. In addition, the morphology has a direct effect on the thermal properties of the microcapsules. The relationship between morphology and encapsulation efficiency related to the interfacial properties of the polymeric systems are investigated.
The morphology of microcapsules containing a hydrophobic core can be affected by thermodynamic and kinetic factors. The kinetics includes the rate of polymerization and phase separation while the thermodynamics are related to the interfacial tensions changes between the three phases [17].
Spreading coefficients can be calculated by the Equation 4.1., where three spreading coefficients can be obtained for each interfacial system:
푆푝 = 훾푤표 − (훾표푝 + 훾푤푝) (4.1a)
푆표 = 훾푤푝 − (훾표푝 + 훾푤표) (4.1b)
푆푤 = 훾표푝 − (훾푤푝 + 훾푤표) (4.1c) where o, w and p are the three phases oil, water and polymer, respectively. γow, γpw and
γop are the water/oil, water/polymer and oil/polymer interfacial tensions, respectively. If
훾표푤 > 훾푝푤, there are only three possible morphology configurations and they are illustrated in Table 4.2. For the condition described by the inequality (4.2), a core–shell morphology is expected, from inequality (4.3) acorn-shaped morphologies occur, whereas from the inequality (4.4), droplet separation takes place.
Table 4.2. Different possible morphologies of the microparticles [17,18].
Predicted configuration Cross-section view Expected morphology complete encapsulation capsule 푆표 < 0; 푆푤 < 0; 푆 > 0 (4.2) 푝 partial encapsulation acorn 푆표 < 0; 푆푤 < 0; 푆 < 0 (4.3) 푝 non-encapsulated sphere 푆표 < 0; 푆푤 > 0; 푆 < 0 (4.4) 푝
90 Optimization of microcapsules recipe and properties
Effect of different suspending agents on microcapsules properties
The following reactions were performed keeping the amount of suspending agent constant at 1.73 wt.% and with a monomers/PCM mass ratio at 1. Arabic gum (AG) and polyvinylpyrrolidone (PVP) were employed as natural and synthetic polymeric suspending agents, respectively. Sodium dodecylsulfate (SDS) was evaluated as an anionic suspending agent.
a) Morphology
SEM images of the microcapsules obtained in presence of different suspending agents are shown in Figure 4.2. The microcapsules were spherical and smooth when prepared by either of the suspending agents. Moreover, in the right hand column of Figure 4.2, cross-sections of the microcapsules are presented. All composites exhibit a matrix structure independently on the suspending agent used. These results are not in agreement with the predictions, where core-shell structures were expected.
(a) (b)
(c) (d)
91 Chapter 4
(e) (f)
(g) (h)
Figure 4.2. SEM micrographs of microcapsules synthesized by different stabilizers: (a) MC(AG), (b) cross-section MC(AG), (c) MC(AG:PVP), (d) cross-section MC(AG:PVP), (e) MC(PVP), (f) cross-section MC(PVP), (g) MC(SDS), (h) cross-section MC(SDS). T=80 °C; C/D phase mass ratio= 11.09; agitation rate= 400 rpm; Tol/PCM= 3.00; PCM= 50%, SA= 1.73%.
In Table 4.3, surface tension, surface polarity and interfacial tension between the three phases, oil, water and polymer, are tabulated. As expected, the surface tension of the aqueous phase in presence of different suspending agents decreased compared to distilled water (70.41 mN m-1) and the highest surface tension decline was observed when SDS was added to the water.
92 Optimization of microcapsules recipe and properties
Table 4.3. Surface tension, surface polarity and interfacial tensions for the oil, polymer and various aqueous phases under study.
Surface tension Surface polarity ow pw op Phase (mN m-1) (%) (mN m-1) (mN m-1) (mN m-1)
ORT27 29.26 0.00 -
P(St-DVB) 35.79 2.78 - 6.52
WH2O 70.41 63.10 34.99 39.14
WAG:PVP 60.77 47.93 32.10 24.54
WAG 61.88 48.61 33.63 21.51 -
WPVP 60.50 46.83 31.52 18.37
WSDS 33.40 40.84 7.60 13.98
It was found that the spreading coefficient of the oil phase in all experiments was lower than zero. In order to study the inequalities, the values of the spreading coefficients for the polymer and water phases are shown in Figure 4.3.
S (mN m-1) So<0; Sw<0; Sp>0 p 12 Impossible conditions 9 AG:PVPP(St-DVB) H O 6 2 P(St-DVB) AGP(St-DVB)
3 PVPP(St-DVB) SDS 0 P(St-DVB) -60 -40 -20 0 20 40 60 -3 -1 Sw (mN m ) -6
-9
-12 So<0; Sw<0; Sp<0 So<0; Sw>0; Sp>0 -15
Figure 4.3. Graphical representation of spreading coefficients and expected morphologies.
93 Chapter 4
It is observed that the spreading coefficient of the polymer Sp varied greatly along the Y-axis depending on the suspending agent. Only in the case of SDS, the interfacial tension between the oil and water phases was low enough that Sp was negative. For the other suspending agents, the oil/water interfacial tension was high enough to ensure positive values of Sp. Consequently, for microcapsules synthesized by using AG, PVP and AG:PVP the predominant structure should be core-shell, whereas acorn shape was preferred for MC(SDS). In order to check the validity of the predicted structures of the microcapsules, they were compared with microscope images. The microscopic observations (matrix structures) are not in agreement with the predictions, where core- shell structures were expected.
b) Particle size
Microcapsules synthesized by using AG and AG:PVP are much larger than those obtained from single PVP and SDS as suspending agents. In the latter, powder materials were obtained. The products synthesized by AG and AG:PVP were sieved because the upper measuring limit of the LALLS is 2000 µm. The highly agglomerated material with a diameter greater than 2000 µm was mechanically separated from the main sample, obtaining 37.60 and 64.03 wt.% for AG and AG:PVP, respectively. The smaller size fractions of the materials were analyzed by laser diffraction and the results are shown in Figure 4.4.
25 25 MC(AG) MC(AG) (a) MC(AG:PVP) (b) MC(AG:PVP) MC(PVP) MC(PVP) 20 20 MC(SDS) MC(SDS)
15 15
Volume (%) Number (%) Number 10 10
5 5
0 0 0 500 1000 1500 2000 2500 0 500 1000 1500 2000 2500 Particle size (m) Particle Size (m)
Figure 4.4. Particle size distribution of synthesized microcapsules with a different stabilizers (a) number average; (b) volume average.
94 Optimization of microcapsules recipe and properties
The volume average particle size for the synthesized MC(AG), MC(AG:PVP), MC(SDS) and MC(PVP) were 1082, 1040, 571 and 301 µm, respectively. The lowest average particle size was obtained for PVP and agreed perfectly with previous results reported by Sanchez et al. [9] for microcapsules synthesized from polystyrene (PSt). They reported that the particle size increases with the viscosity of the continuous phase and also, with the reaction time previous to the identity point [19]. In agreement with the SEM images and sieves analysis, the laser diffraction results showed that the particle size for microcapsules obtained by using AG, AG:PVP and SDS were larger than those obtained for microcapsules synthesized by using PVP. The larger measured particle size for microcapsules obtained by using SDS can be explained by the agglomeration of the single particles. This was also confirmed by the result obtained for the number average particle size where values for AG, AG:PVP, SDS and PVP were 905, 865, 15 and 174 µm, respectively. This illustrates that the volume average size for the microcapsules synthesized by using SDS were dominated by relatively few large clusters, while most of the particles were much smaller. In order to obtain a product containing only single particles without huge agglomerates, a large amount of this surfactant must be used. Unfortunately, large amount of SDS creates foaming problems in the reaction systems.
The particle size of a microcapsule is also affected by the interfacial tensions reached during synthesis and is dependent on the type of suspending agent, and core material as well as the monomer used. The interfacial tension between oil and water (γow) decreased (Table 4.2) in the presence of suspending agent [20]. The highest reduction of oil/water interfacial tension was observed for SDS which allowed us to produce a uniform protective layer, followed by PVP. Consequently, microcapsules exhibited a decrease in particle size when the interfacial tension values became lower for the assigned suspending agents. Pronounced coalescence demonstrate poor colloidal stability of AG and AG:PVP in this system.
c) Thermal properties
In Table 4.4, latent heat (H), paraffin content (CPCM) and encapsulation efficiency (EE) of obtained microcapsules are tabulated. The highest heat storage capacity (66.37 J g-1) was obtained for the sample where PVP was used as a stabilizer. For this sample, the encapsulation efficiency was 69.77% and paraffin content was 38.77%. For
95 Chapter 4
comparison, Sanchez et al. [3] synthesized microcapsules by using PSt shell where the paraffin content was 28.69 wt.%. Hence, by changing the recipe it is possible to increase the paraffin content to obtain microcapsules with the desired characteristics.
When the interfacial tension between the polymeric and aqueous phase decreases, the tendency of polymer to migrate through the water and droplet interface increases [21]. Consequently, the encapsulation efficiency increases. As can be seen in Table 4.3, the lowest surface polarities were obtained for water with PVP and SDS. Therefore, in presence of these suspending agents in the water phase, the polymer was more prone to create a shell around RT27. Accordingly, the encapsulation efficiency was higher than for the other suspending agents, which were 69.77, 63.99, 61.96 and 61.64 for MC(PVP), MC(SDS), MC(AG) and MC(AG:PVP), respectively.
Table 4.4. Properties of microcapsules using different suspending agents.
-1 Sample ΔH (J g ) CPCM (%) EE (%) ηr (%)
MC(AG) 56.27 32.87 61.96 80.98
MC(AG:PVP) 53.53 31.27 61.64 63.07
MC(PVP) 66.37 38.77 69.77 84.89
MC(SDS) 64.19 37.49 63.99 81.99
The yield of the microcapsules for the products obtained by using AG, AG:PVP, SDS and PVP were 80.98, 63.07, 81.99 and 84.89%, respectively. The highest microcapsules yield 84.89% was obtained by using the PVP. These results are contrary to those reported by Alcazar et al. [16] using PVP, who found a large amount of non- spherical particles (23.29%). This suggests that the core materials exhibit a large influence on the microencapsulation process. According to the results above, microcapsules from P(St-DVB) containing the paraffin Rubitherm®RT27 can be satisfactory utilized to facilitate better thermal comfort for building applications. Based on the particle size and particle size distribution, the yield of the microcapsules, and the thermal energy storage, PVP seems to be the most suitable suspending agent to accomplish the production of these thermoregulating materials. PVP was therefore used
96 Optimization of microcapsules recipe and properties in the studies of how the other processing parameters were influencing the produced microcapsules.
d) Thermal stability
Thermogravimetric analyses were carried out in order to confirm the absence of toluene and study the thermal stability of materials, Figure 4.5 shows the TGA curves for the pure PCM, P(St-DVB), PVP, AG, SDS whereas Figure 4.6 presents TGA results for microcapsules synthesized with different suspending agents.
Figure 4.5. TGA curves for the P(St-DVB), neat Rubitherm®RT27 and suspending agents.
97 Chapter 4
Figure 4.5 shows that the paraffin had the highest volatility among the pure materials, shortly afterwards the decomposition processes of SDS, AG, P(St-DVB) and PVP took place. It is noticed that the paraffin and SDS presented one weight loss peak from 150 to 250 °C and from 200 to 260 °C, respectively but the polymer materials revealed two weight loss ranges. Regarding AG and PVP, the first weight loss was caused by the presence of water, therefore the peak was observed at 100 ºC while for P(St-DVB) the change was related with the evaporation of monomer and low molecular weight polymer and it was registered from 100 to 200 ºC. The second weight loss observed mainly at temperatures 310.67, 431.87 and 422.07 ºC for AG, PVP and P(St-DVB), respectively was related to the polymer degradation. Finally, the amount of residue remaining after 500 ºC suggest that paraffin was fully evaporated while the SDS and AG were the most stable with residue contents higher than 20%.
Figure 4.6. TGA curves for the different studied materials containing PCM synthesized by using the different suspending agents.
Figure 4.6 shows that all synthesized microcapsules present two weight losses. The first one was related with the paraffin content. However, in the case of SDS, it is possible that this surfactant agent could be incorporated into the microcapsules. It is
98 Optimization of microcapsules recipe and properties difficult to differ between these because SDS evaporates at the same temperature range as the paraffin. In the same way, the second weight loss corresponded mainly with the degradation of PVP and the copolymer.
Effect of continuous/discontinuous phase mass ratio on microcapsules properties
In order to increase efficiency of the reaction, the continuous/discontinuous phase mass ratio was decreased from 11.09 to 3.24 for the microcapsules synthesized by using PVP. Figure 4.7 shows the SEM images for microcapsules obtained by using water/organic phase mass ratio of 11.09, 5.55 and 3.24, respectively.
(a) (b)
(c)
Figure 4.7. SEM images of microcapsules synthesized by changing the continuous/discontinuous phase mass ratio: (a) MC(11.09); (b) MC(5.55); (c) MC(3.24). T=80 °C; agitation rate= 400 rpm; Tol/PCM= 3.00; PCM= 50%, SA= 1.73%.
As can be seen, spherical and smooth microcapsules were obtained in each synthesis. It was also observed that, the lower the continuous/discontinuous mass ratio,
99 Chapter 4
the higher the particle size of the microcapsules. This is due to the difficulty of achieving a proper distribution of the oil phase into the continuous phase. Accordingly, the coalescence of the monomer increases previous to the polymer formation, favoring the production of larger particles [22]. This difference in particle size was confirmed by LALLS (Figure 4.8).
14 16 (a) MC(11.09) (b) MC(11.09) MC(5.55) MC(5.55) 12 14 MC(3.24) MC(3.24)
12 10
10 8 8 6
Volume (%) Number (%) Number 6 4 4
2 2
0 0 0 500 1000 1500 2000 0 500 1000 1500 2000 2500 Particle Size (m) Particle Size (m)
Figure 4.8. Particle size distribution for microcapsules synthesized by using different continuous/discontinuous phase mass ratio (a) number average; (b) volume average.
LALLS showed that there was no difference between the particle size distribution in number and volume when the continuous/discontinuous phase mass ratio was increased from 3.24 to 5.55. On the contrary, a large shift was observed in the particle size of the microcapsules when the continuous/discontinuous phase mass ratio was 11.09, obtaining the lowest number and volume particles size distribution. It is also important to point out that both the yield of the microcapsules and the encapsulation efficiency were highest for the larger continuous/discontinuous phase mass ratio, (84 and 69%, respectively). The total product obtained using the lowest continuous/discontinuous mass ratio is three times higher than that obtained by using the highest mass ratio. Hence, the lowest continuous/discontinuous mass ratio was selected due to higher amount of product per reaction. This reduces the generation of a large amounts of waste water, which have to be further treated. In addition, the process yield and the encapsulation efficiency could be improved by changing the other reaction conditions.
100 Optimization of microcapsules recipe and properties
Effect of stirring rate on microcapsules properties
Figure 4.9 shows the influence of changing the agitation rate from 400 to 1200 rpm on the volume average particle size (a) and its distribution (b).
1000 (a) dv 600 (b) MC(400rpm) 900 0.5 20 MC(800rpm) dn MC(1200rpm) 800 0.5 500
700 15 400
m) m) 600
(
(
0.5 0.5 500 300 10
dv
dn 400 (%) Volume 200 300 5 100 200
100 0 0 400 600 800 1000 1200 0 500 1000 1500 2000 Agitation speed (rpm) Particle Size (m)
Figure 4.9. Particle size of synthesized microcapsules with a different stirring speed (a) influence of the stirring speed on volume and number average particle size; (b) volume particle size distribution. T=80 °C; C/D phase mass ratio= 3.24; Tol/PCM= 3.00; PCM= 50%, SA= 1.73%.
As expected, the volume average particle size and number average particle size decreased by increasing the agitation rate. The volume average was higher, indicating that a high amount of particles having large size and weight were present in the product. This behavior is due to the direct influence of this operating variable on the drop formation and their aggregation [4,23]. Figure 4.9b shows unimodal and bimodal particle size distributions depending of the agitation rate. A low agitation rate resulted in a unimodal particle size distribution. However, it had the disadvantage providing a wide distribution. On the other hand, the two higher agitation rates promoted the formation of bimodal size distributions. Of these, the distribution obtained using 800 rpm was the narrowest. Accordingly, an agitation rate of 800 rpm could be considered the adequate for synthesizing microcapsules with a particle size lower than 500 μm. This selection (800 rpm) is also confirmed by the highest values of EE, CPCM and r at this agitation rate. The values obtained at the agitation rates 400, 800 and 1200 rpm were respectively: (i) EE
101 Chapter 4
51.57, 60.34 and 57.44, (ii) CPCM 29.63, 32.41 and 32.37 and (iii) r 75.78, 80.17 and 78.72%.
Effect of toluene/PCM mass ratio and the PCM content on microcapsules properties
In order to decrease the cost of the process and increase the environmentally- friendly character of the reaction, the mass ratio between toluene and PCM was decreased from 3 to 1.92 and the PCM content was increased from 50 to 68.5% with respect to the total mass of monomers and PCM. SEM micrographs for the microcapsules obtained by changing the porogen/PCM mass ratio and amount of PCM are shown in Figure 4.10.
(a) (b)
Figure 4.10. SEM micrographs of microcapsules synthesized by using changed toluene/PCM mass ratio and amount of PCM (a) 3.00 and 50%, (b) 1.92 and 68.5%, respectively. T=80 °C; C/D phase mass ratio= 3.24; agitation rate= 800 rpm; SA= 1.73%.
As can be seen, a lower amount of toluene and higher amount of PCM caused slight increase in the particle size. This observation is in agreement with previous findings [4]. Nevertheless, microcapsules exhibited still spherical and smooth shape. The effect of these conditions on EE and r are shown in Figure 4.11.
102 Optimization of microcapsules recipe and properties
Figure 4.11. The influence of the toluene/PCM mass ratio and amount of PCM on the yield of
the process (r) and encapsulation efficiency (EE).
Accordingly, a lower toluene/PCM mass ratio and a higher amount of PCM, increase the encapsulation efficiency and the yield of the process. Due to these results, the following reactions were performed using 68.5 % of PCM in the discontinuous phase and a toluene/PCM mass ratio of 1.92.
Influence of PVP concentration on microcapsules properties
With the aim of examining the effect of PVP concentration on the microcapsules properties, several concentrations (0.87, 1.73, 3.41, 5.03, 6.59, 8.10 and 14.99%) have been studied. The reactions have been performed holding continuous/discontinuous mass ratio at 3.24, an agitation rate of 800 rpm, a toluene/PCM mass ratio of 1.92 and 68% of PCM with respect to the total amount of monomers and PCM used.
a) Morphology
SEM images for some microcapsules synthesized at different concentrations of PVP are depicted in Figure 4.12. MC(PVP-0.87) has an uneven, irregular shape, making
103 Chapter 4
it difficult to clearly define the structure (Figure 4.12a). Accordingly, this amount of surfactant is not enough to ensure a good dispersion of the discontinuous phase into the continuous one. All the other products have a spherical shape and a smooth surface, also indicating that the use of toluene as porogen prevent the formation of holes. According to the SEM photographs, the optimal PVP amount must be between 5.03 and 8.10%.
(a) (b)
(c) (d)
(e) (f)
104 Optimization of microcapsules recipe and properties
(g)
Figure 4.12. SEM micrographs of microcapsules synthesized by different amount of stabilizers: (a) MC(PVP-0.87); (b) MC(PVP-1.73); (c) MC(PVP-3.41); (d) MC(PVP-5.03); (e) MC(PVP- 6.59); (f) MC(PVP-8.10); (g) MC(PVP-14.99). T=80 °C; C/D phase mass ratio= 3.24; agitation rate= 800 rpm; Tol/PCM= 1.92; PCM= 68%.
In order to evaluate morphology, the interfacial properties were determined and shown in Table 4.5.
Table 4.5. Surface tension, surface polarity and interfacial tensions for the various aqueous phases under study.
Surface tension Surface polarity ow pw Phase (mN m-1) (%) (mN m-1) (mN m-1)
WPVP-0.87 60.63 59.29 33.28 31.40
WPVP-1.73 60.50 46.83 32.55 18.37
WPVP-3.41 60.68 45.40 30.44 17.42
WPVP-5.03 61.83 43.72 28.78 15.84
WPVP-6.59 60.11 37.22 29.63 14.26
WPVP-8.10 59.75 23.73 28.58 13.94
The surface tension of aqueous phases in presence of PVP was not affected by the concentration, which is in agreement with a previous study performed by Bolten and Turk
[24]. The interfacial tension between oil and water (γow) decreased as the PVP concentration was raised from 0.87% to 3.41%, after which γow remained practically constant up to a PVP concentration of 8.10%. The water/polymer interfacial tension (γpw) was reduced with the addition of suspending agent and became close to those obtained by using the SDS solution. Therefore, the surface polarity of the water phase decreases
105 Chapter 4
systematically with increasing amounts of PVP. Spreading coefficients and prediction of morphology were determined for microcapsules synthesized in presence of different amount of PVP and is graphically illustrated in Figure 4.13. An acorn structure for MC(PVP-0.87) and a core-shell structure for the other microcapsules were expected.
-1 Sp (mN m ) 8 Impossible conditions 6 PVP-0.87 PVP-5.03 4 PVP-1.73 PVP-6.59 2 PVP-3.41 PVP-8.10 So<0; Sw<0; Sp>0 0 -60 -40 -20 0 20 40 60 -2 -1 Sw (mN m ) -4 -6 -8 S <0; S >0; S >0 So<0; Sw<0; Sp<0 o w p
Figure 4.13. Graphical representation of spreading coefficients and expected morphologies for different PVP concentrations.
Cross-sections of some microcapsules synthesized by changing the concentration of PVP are depicted in Figure 4.14. As mentioned before MC(PVP-0.87) had the structure difficult to define. The other samples exhibited a matrix configuration, similar to those obtained for the products synthesized by using different suspending agents. Although, thermodynamic theory tends to predict core-shell structure, Figure 4.3 and 4.14 clearly show that this is not the case for these samples. From a kinetic perspective, these results indicate a relatively high polymerization rate of styrene and divinylbenzene as co- monomers in which propagation proceed faster than phase separation. Moreover, the viscosity of the oil phase increased due to polymerization of the monomers and made the diffusion more difficult [25]. Neither type of suspending agent, nor the concentration of suspending agent in the water phase have a decisive influence on the final morphology of the microcapsules. Accordingly, P(St-DVB) exhibits little driving force for phase
106 Optimization of microcapsules recipe and properties separation of the polymer from PCM resulting in a homogenous structure in which the core is uniformly distributed in the polymer.
(a) (b)
(c)
Figure 4.14. SEM micrographs of microcapsules synthesized by different amount of stabilizers: (a) cross-section MC(PVP-0.87); (b) cross-section MC(PVP-5.03); c) cross-section MC(PVP- 8.10).
b) Particle size
It can be observed that all products seemed to be quite homogeneous in particle size. The particle size was decreasing with the amount of PVP, except for the highest PVP concentration (PVP-14.99). This indicates that a large amount of suspending agent exerts an undesirable effect of particle coagulation. An increase in the amount of stabilizer from a 1.73 to 8.10%, resulted in a product with the desired properties: smaller particle size, spherical and regular shape and also a smooth surface. The volume and number average particle size distribution of the different products has been analyzed by laser diffraction and the results are shown in Figure 4.15. Figure 4.15a indicates that all products present unimodal volume average distributions. As
107 Chapter 4
expected, the widest distribution was observed for the irregular product obtained using 0.87% of PVP. However, Figure 4.15b shows that the number average particle size distribution for the microcapsules synthesized by 6.59 and 14.99% of PVP were bimodal. These results confirmed that an excess of suspending agent favors the formation of smaller particles, which agglomerates during the reaction time, promoting the formation of larger particles. Hence, in order to obtain unimodal distributions, 5.03% of PVP should be used.
20 (a) MC(PVP-0.87) 20 (b) MC(PVP-0.87)
10 10
0 0 20 MC(PVP-1.73) 20 MC(PVP-1.73)
10 10
0 0 20 MC(PVP-3.41) 20 MC(PVP-3.41)
10 10
0 0 20 MC(PVP-5.03) 20 MC(PVP-5.03)
10 10
0 0 20 MC(PVP-6.59) 20 MC(PVP-6.59)
Volume (%) Volume
Number (%) Number
10 10
0 0 20 MC(PVP-8.10) 20 MC(PVP-8.10)
10 10
0 0 20 MC(PVP-14.99) 20 MC(PVP-14.99)
10 10
0 0 0 500 1000 1500 2000 0 20 40 60 100 150 200 250 300 Particle size (m) Particle size (m)
Figure 4.15. Particle size distribution for microcapsules with different amount of PVP (a) volume average; (b) number average.
108 Optimization of microcapsules recipe and properties
Figure 4.16 shows the influence of the amount of PVP on the volume average and number average particle size. As can be seen, the volume and number average sizes do not follow the same trend. While the behavior of the number average increased with the PVP amount and seemed to be nearly linear, the volume average presented a minimum at PVP-9.14. The number average sizes are more sensitive to the presence of small particles, while the volume average sizes are biased toward large particles. Accordingly, the results suggest that as the concentration of PVP is increased up to 8.10%, fewer of the largest particles were produced (reduction in dv0.5) combined with fewer very small particles
(increase in dn0.5). The result of this is a narrower size distribution.
350 350
300 300
250 250
)
)
m
m
200 200
(
(
dv0.5
0.5 0.5 150 150 dn0.5
dv
dn 100 100
50 50
0 0 0 2 4 6 8 10 12 14 16 PVP amount (%)
Figure 4.16. Influence of PVP concentration on the volume average (dv0.5) and number average
(dn0.5) particle size.
The minimum in the volume average sizes and the increase in the number average sizes might be related to the solubility of PVP in water. According to the data sheet from the PVP supplier, at a molecular weight of 40 000 g mol-1 the solubility is 100 mg mL-1 at 25 ºC. This is lower than the 233 mg mL-1 of the PVP-14.99. At the highest concentration some of the PVP will therefore be in solid state, dispersed as small particles in the solvent. These solid particles might act as nuclei on which the polymer drops grow, leading to the formation larger microcapsules. These results were not in agreement with those reported by Ma et al. [26], obtaining stable particle size from 21.6 to 20.9 µm when the PVP (K30, Mw = 40 000 g
109 Chapter 4
mol-1) concentration was changed from 2.0 to 7.0 g in 225 g of water, respectively. In addition, they obtained an increase in the amount of coagulum for a PVP concentration higher than 4.0 g in 225 g of water, a phenomenon that only appeared in our case for the largest studied PVP amount.
c) Thermal properties
Figure 4.17 shows latent heat (H) of the microcapsules synthesized with different amounts of PVP. The results indicate that, microcapsules with thermal energy storage capacity higher than 80 J g-1 can be produced using this technology. A maximum value 101.80 J g-1 was reached for the microcapsules synthesized using a 5.03% of PVP. This do not follow the observation of Li at al. [27] where a two-step miniemulsion polymerization method resulted in an increase in the latent heat of the microcapsules from 114.6 to 143.7 J/g when the amount of surfactant was changed from 0.05 to 0.20 g in water. This indicates that the thermal properties of the microcapsules are mainly dependent on the microencapsulation technology.
100
80
)
-1 60
H (J g 40
20
0 0.87 1.73 3.41 5.03 6.59 8.1 14.99 PVP amount (%) Figure 4.17. Latent heat of the microcapsules as function of the PVP mass ratio.
According to the latent heat, 5.03% of PVP can be considered as the optimal PVP amount to produce microcapsules having a large TES capacity. Nevertheless, before taking a final decision about the optimal recipe, the effect of PVP mass ratio on other
110 Optimization of microcapsules recipe and properties important variables was studied. As stated in the introduction, the final goal of this study is to clarify the role of the suspending agent not only on the particle size but also on other important variables such as the encapsulation efficiency, the paraffin content and the microcapsule yield. The paraffin content (CPCM) and the microcapsules yield (ηr) are depicted in Figure 4.18 as function of the PVP concentration whereas the encapsulation efficiency is shown in Figure 4.19.
A slight increase of EE and ηr with increasing amount of PVP was observed whereas the CPCM seemed to be almost stable at 53%, although the highest value of 59.46% was obtained for 5.03% of PVP.
100 100
90 80
80 60
(%) C (%) 70 PCM
(%)
r (%) r
PCM 40
C 60
20 50
40 0 0 2 4 6 8 10 12 14 16 PVP amount (%)
Fig. 4.18. Effect of PVP amount on microcapsules yield (ηr), paraffin content (CPCM) and encapsulation efficiency (EE).
In Figure 4.18, the influence of PVP concentration on surface polarity and encapsulation efficiency is shown. The standard deviation of the encapsulation efficiency for studied microcapsules was between 3 and 4%. With increasing amount of PVP, the surface tension of the water phase decreased and the encapsulation efficiency raised slightly.
111 Chapter 4
100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30
Surface polarity (%) polarity Surface
20 20 (%) efficiency Encapsulation
10 Surface polarity 10 Encapsulation efficiency 0 0 0 1 2 3 4 5 6 7 8 9 PVP amount (%)
Figure 4.19. Influence of PVP amount on surface polarity and encapsulation efficiency.
The highest values of the encapsulation efficiency (77.59%) and microcapsule yield (84.28%) were found for the products using 5.03 and 8.10% of PVP, respectively. These results are contrary to those reported by Khakzad et al. [2] encapsulating hexadecane in a melamine formaldehyde shell by using the in situ dispersion polymerization technique in aqueous media. They observed a decrease in the encapsulation efficiency from 121 to 84.4%, increasing the poly(vinyl alcohol) concentration from 1 to 8 wt.%. In the case of the yield, Ma et al. [26] synthesizing polystyrene-polyacrylamide composite microspheres from water/oil/water emulsion and further suspension polymerization reported a maximum of 90.2% when they used 5 g of PVP in 225 g water. Hence, there is no rule that allows establishing the optimum concentration of the stabilizer in the manufacture of microcapsules, as it is dependent on the microencapsulation technology. Accordingly, by changing the PVP amount, it is possible to increase the paraffin content and to obtain microcapsules with the desired characteristics.
d) Adsorption of PVP onto microcapsules shell
We have reasons to think that PVP is retained by the microcapsule shell in a similar way in which an adsorption material retain an adsorbate.
112 Optimization of microcapsules recipe and properties
In order to examine the presence of PVP in the microcapsules, thermal analyses were performed. The TGA curves for the pure Rubitherm®RT27, St-DVB copolymer, pure PVP and some microcapsules obtained at different PVP concentrations are shown in Figure 4.20.
100 P(St-DVB) pure RT27 pure PVP 80 MC(PVP-1.73) MC(PVP-5.03) MC(PVP-8.10) 60 MC(PVP-14.99)
40
Weight (%) Weight
20
0 0 100 200 300 400 500 550 600 Temperature (C)
Figure 4.20. TGA curves for the studied materials: P(St-DVB) copolymer, pure Rubitherm®RT27, PVP and microcapsules synthesized by using the different PVP amount.
Figure 4.20 shows that the pure material with the greatest volatility is Rubitherm®RT27, followed by the evaporation of the Rubitherm®RT27 encapsulated by a polymeric shell and finally, the decomposition processes of P(St-DVB) and PVP. Accordingly, MC(PVP-5.03) was the type of microcapsules with the highest content of encapsulated PCM, followed by MC(PVP-8.10) and with a minimum for MC(PVP- 14.99). This indicate that for a higher amount of PVP the paraffin content in the microcapsules might be slightly lower. The second weight loss in the microcapsule curve was due to the polymer degradation (PVP and P(St-DVB)). Therefore, the degradation temperatures of PVP and P(St-DVB) suggest that TGA is not a good technique to differ between the presence of PVP or P(St-DVB) in the microcapsules. The amount of residue obtained in this TGA after 500 ºC shows that in the case of microcapsules, the residue
113 Chapter 4
increases with the mass ratio of PVP used in the recipe, explaining why the microcapsules produced at a high mass ratio of PVP have a lower paraffin content.
As stated in the introduction, one of the final goals of this chapter is to clarify the role of the suspending agent not only on the particle size, encapsulation efficiency, paraffin content and the microcapsule yield, but also as a compound forming part of the microcapsules. The hypothesis is that the suspending agent remains as a constitutive part of the particle after the synthesis, with significant influence on its final properties.
The presence of PVP in the microcapsules (f SA) can be quantified from the DSC analyses and weighing the total amount of product (PMC) obtained in each reaction. It was assumed a 100% of monomer conversion.
푃푀퐶 = 푃퐶푀푀퐶 + 푃(푆푡 − 퐷푉퐵)푀퐶 + 푓 푆퐴 (4.5) where SA is the total amount of the suspending agent used in the synthesis.
In this way, the fraction of suspending agent that constitutes the microcapsules f can be calculated by:
푃 − 푃퐶푀 − 푃(푆푡 − 퐷푉퐵) 푓 = 푀퐶 푀퐶 푀퐶 (4.6) 푆퐴
Once values of f are calculated, the amount of PVP on the microcapsules (Г in g -1 kg ) and the concentration of PVP in the bulk solution at the end of the process (Cb in kg m-3) can be estimated by Equations 4.7 and 4.8, respectively.
푓푆퐴 Г = (4.7) 푃푀퐶
(1 − 푓)푆퐴 퐶 = (4.8) 푏 푉 where V is the total volume of the bulk solution.
Figure 4.21 shows the relationship between Г and Cb. PVP was strongly taken up by P(St-DVB) even at a very low concentration in the water/toluene media. This behavior is characteristic of sorbents that exhibit a high affinity for the solute in adsorption systems [15]. As for most conventional adsorption processes, a maximum value of incorporated
114 Optimization of microcapsules recipe and properties
PVP was reached. The interaction established between the P(St-DVB) and PVP can be explained by the observations of Smith et al. [15]. They studied the adsorption of PVP onto polystyrene lattice and concluded that “in water, interaction with the PSt occurred through the PVP hydrophobic methylene/methane groups and the positive dipole of the amide nitrogen of the pyrrolidone ring. The negative dipole associated with the amide oxygen is directed away from the surface into the solution”.
200
150
)
-1 100
g kg
(
50 Experimental data Langmuir model 0 0 20 40 60 80 100 120 140 160 180 200 -3 Cb (kg m )
Figure 4.21. Adsorption isotherm of PVP onto P(St-DVB) in water/toluene media at 80 ºC.
In order to fit these data, the Langmuir model (Equation 4.9) usable for monolayer materials was selected [12].
Γ 퐾퐶푏 max = (4.9) Γ 1 + 퐾퐶푏 where Гmax is the maximum retention capacity of the P(St-DVB) and K is the equilibrium constant of the system, defined as the ratio between adsorption and desorption rate constants, respectively.
Experimental data were fitted to Equation 4.9 in order to obtain the two unknown parameters Гmax and K. For that purpose, a fitting tool for solving non-linear equations based on Marquardt’s algorithm was utilized. The fitting values of Гmax and K and their
115 Chapter 4
confidence interval using a confidence level of 95% (α=0.05) were 192.9±0.4 g/kg and 0.18± 0.11 m3/kg, respectively. The proposed model gave a good fit to the experimental data, illustrating that the distribution of the PVP between the solid and liquid phases at the end of the polymerization process followed a Langmuir type trend.
The value of Г for MC(PVP-5.03) was also confirmed by EDAX analyzing the nitrogen content of the microcapsules. In this way, EDAX analyses were performed in MC(PVP-5.03) on the external surface and at the center of the microcapsule. A small difference was found between the external and internal nitrogen contents 1.46 and 2.30 wt.%, respectively. These values were within the respective theoretical nitrogen content (2.28 wt.%) obtained from the adsorption curve (180.81 g kg-1) and the nitrogen content of the PVP (12.60 wt.%). Figure 4.22 presents EDAX mapping images for carbon and nitrogen. The uniform distribution of nitrogen through the microcapsules also indicates that the adsorption of PVP takes place in the whole microcapsule structure.
Accordingly, thermoregulating microcapsules from P(St-DVB) could contain up to a 19.3 wt.% of surfactant when the percentage of PVP of the total mass is higher than 5.03 wt.%.
C N
Figure 4.22. EDAX mapping pictures showing overall SEM image and elemental regions of carbon and nitrogen.
Influence of PNC-HEMA incorporation on morphology and encapsulation efficiency
Microcapsules with an additional co-monomer, PNC-HEMA, were synthesized with the aim to improve the fire resistance of microcapsules. A full characteristic of this
116 Optimization of microcapsules recipe and properties material and the microcapsules is given in Chapter 6. However, this monomer exhibits important features which are interesting to consider in this chapter. PNC-HEMA presents more hydrophilic and polar nature than styrene and divinylbenzene due to heteroatomic structure. Accordingly, the effect of the increasing the amount of PNC-HEMA (5, 10 and 20% with respect to the total amount of monomers) has been analyzed considering interfacial properties (Table 4.6). All products exhibited positive values of Sp, indicating core-shell morphologies (Figure 4.23).
Table 4.6. Surface tension, surface polarity and interfacial tensions for the various polymers and aqueous phases PVP-5.03 under study.
Surface free energy Surface pw op Phase of polymer polarity (mN m-1) (mN m-1) (mN m-1) (%)
Different polymers in aqueous phase WPVP-5.03
P(St-DVB) 35.79 2.78 15.84 6.52
P(St-DVB-5%PNC-HEMA) 40.95 4.84 14.73 11.98
P(St-DVB-10%PNC-HEMA) 41.95 7.74 14.23 13.01
P(St-DVB-20%PNC-HEMA) 42.10 9.76 13.95 13.24
-1 Sp (mN m ) 12 P(St-DVB-20% PNC-HEMA) 9 P(St-DVB-10% PNC-HEMA) P(St-DVB-5% PNC-HEMA)6
3 Impossible conditions So<0; Sw<0; Sp>0 0 -40 -30 -20 -10 0 10 20 30 40 -3 -1 Sw (mN m ) -6
-9
-12 S <0; S >0; S >0 So<0; Sw<0; Sp<0 o w p -15
Figure 4.23. Graphical representation of spreading coefficients and expected morphologies for different PNC-HEMA concentrations.
117 Chapter 4
In Figure 4.24, SEM images of microcapsules obtained by addition of PNC- HEMA and their cross-sections are shown. Microcapsules were wrinkled, which can be attributed to the shrinkage occurring during polymerization due to the density difference between monomers (<1 g cm-1) and polymers (> 1 g cm-1). In addition, the volume of the core material was reduced during the phase change from molten to solid [28]. Nevertheless, only for these microcapsules the predicted morphology (core-shell) was in an agreement with the experimental results. The microcapsules exhibited a morphology similar to pomegranate-like structure where the core consists of seed-like spheres with a thin continuous shell. Similar morphologies have been reported previously when acrylic [29,7] and vinyl [25] monomers used. This is usually observed when monomers of different reactivity are utilized for polymerization [25]. The PCM is separated from the polymer creating nanograins.
(a) (b)
(c) (d)
118 Optimization of microcapsules recipe and properties
(e) (f)
Figure 4.24. SEM images for microcapsules containing different amount of PNC-HEMA (a) MC-P(St-DVB-5%PNC-HEMA); (b) cross-section MC-P(St-DVB-5%PNC-HEMA); (c) MC- P(St-DVB-10%PNC-HEMA); (d) cross-section MC-P(St-DVB-10%PNC-HEMA); (e) MC- P(St-DVB-20%PNC-HEMA); (f) cross-section MC-P(St-DVB-20%PNC-HEMA). T=80 °C; C/D phase mass ratio= 3.24; agitation rate= 800 rpm; Tol/PCM= 1.92; PCM= 68%.
It is important to point out that in Figures 4.3, 4.13 and 4.23 nearly all points (Sw,
Sp) were located in the same quarter on the chart, predicting a core-shell structure of microcapsules. However, the morphology was observed to vary between matrix and core- shell for different microcapsule compositions. Accordingly, there is not an agreement between the predicted morphology based only on the spreading coefficients and the experimental observations. Such a discrepancy between predicted and experimental morphology has been reported previously for emulsion solvent evaporation and mini emulsion polymerization [30-33].
From a thermodynamic perspective, when the surface polarities percentages of the liquid and polymer are closer to each other, the phases become more compatible and the free energy is reduced. Accordingly, determination of these percentages allows prediction of whether the immersion is unfavorable or favorable [34]. The surface polarities of the core and shell materials were very close to each other and relatively hydrophobic. The compatibility between them slowed down the phase separation and migration of oligomeric chains from the oil phase to the external part of the drop, leading to difficulties in core-shell formation. In Figure 4.25, we can observe that surface polarity increased with the concentration of PNC-HEMA in the polymer structure. The diffusion of the monomer though oil phase becomes faster when the polarity of the shell becomes higher.
119 Chapter 4
Higher shell polarity ensures reduced compatibility between the core and shell, which improve the affinity between the shell and the water phase. Accordingly, it allowed the core-shell structure formation. Similar phenomena was reported by Lashgari [7] who observed matrix morphology for microcapsules synthesized by using poly(methyl methacrylate-co-butyl acrylate) as a shell and multi-nucleus for poly(methyl methacrylate), where the polarity indexes were 0.01 and 4.9, respectively.
The encapsulation efficiency increased significantly when PNC-HEMA was used as an additional co-monomer. An encapsulation efficiency of 77.59% for microcapsules from styrene and divinylbenzene was obtained. Whereas, for MC(St-DVB-5%PNC- HEMA), MC(St-DVB-10%PNC-HEMA) and MC(St-DVB-20%PNC-HEMA) yielded 88.06, 86.83 and 85.52%, respectively. A drop in encapsulation efficiency was observed when the content of PNC-HEMA was raised. This can be caused by a weaker and more shrunken shell for PNC-HEMA-rich microcapsules.
20 20 Surface polarity 18 Water/polymer interfacial tension 18
16 16
14 14
)
12 12 -1
10 10
(mN m (mN
8 8 wp
6 6
Surface polarity (%) polarity Surface
4 4
2 2
0 0 0 2 4 6 8 10 12 14 16 18 20 PNC-HEMA amount (%)
Figure 4.25. Influence of PVP amount and PNC-HEMA amount on water-polymer interfacial tension.
Non-encapsulated PCM can be reduced to a minimum when a morphology close to core shell is expected. However, perfect core-shell structures are not necessarily more effective than matrix–like structures. Although microcapsules with core-shell structure
120 Optimization of microcapsules recipe and properties exhibited higher encapsulation efficiency, the shell thickness is very low, resulting in weaker microcapsules that are easier damaged in practical applications.
Scale-up
Once the recipe for the production of microcapsules containing PCM was established, the material was synthesized in a pilot scale plant in a 100 L reactor and stirrer geometrically proportional to those used in the laboratory scale experiments (described in section 3.2.1 of Chapter 3). For comparison the reaction by using PVP-1.73 and PVP-8.10 were also performed. Spherical beads with smooth surface were obtained, as can be observed in Figure 4.26.
(a) (b)
Figure 4.26. ESEM micrographs of MC(PVP-5.03) synthesized in pilot plant: (a) general view, (b) magnification view.
It was found that the products have smaller volume particle size by using the 100 L reactor than the lab scale reactor (Figure 4.27). This could be related with the energy dissipation rate. If the energy delivered by the stirrer at both scales were the same, the particle size should be equal [9,35]. The particle size of the microcapsules was smaller in the pilot plant, even though a much slower stirring rate was used at pilot plant scale (300 rpm) than at laboratory scale (800 rpm). This means that the geometrically similar to a Rushton stirrer utilized in the pilot plant is more efficient delivering energy to the bulk. In general, when increasing the amount of PVP in pilot plant scale, all products exhibited a larger number particle size but a lower volume particle size than those obtained at lab
121 Chapter 4
scale. Accordingly the two size distributions are closer to each other, indicating the possibility of synthesizing relatively monodisperse materials. The volume and number averages were closer to each other than for the lab-scale experiments, which is indicative of a narrower size distribution. Two main conclusions can be extracted from these observations: i) for devices that are geometrically similar, the delivery of power is more effective at large scale, ii) increasing the scale using the same mass ratio it is possible to reach a more monodisperse product.
300 300 Lab scale Pilot scale
dv0.5 dv0.5
250 dn0.5 dn0.5 250
200 200
m)
m)
( ( 150 150
0.5
0.5
dv
dn 100 100
50 50
0 0 0 2 4 6 8 10 PVP amount (%)
Figure 4.27. Influence of PVP concentration on the volume average (dv0.5) and number average
(dn0.5) particle size.
As expected, values of CPCM, EE and ηr obtained at pilot plant scale were similar to those reached at laboratory scale but requiring a lower amount of suspending agent for leading the same encapsulation efficiency and microcapsules yield (78.86 and 79.99%, respectively for 5.03% of PVP used). The effect of PVP amount on microcapsules yield and paraffin content is shown in Figure 4.28. Thus, it was possible to scale up the production of microcapsules owing to the robustness of the formulation.
122 Optimization of microcapsules recipe and properties
100 100
90 80 80
(%) 60 70
EE
(%)
r
and and 60 40
PCM C 50 Lab scale Pilot scale 20 40 EE EE CPCM CPCM 30 r r 0 0 2 4 6 8 10 PVP amount (%)
Figure 4.28. Effect of PVP amount on microcapsules yield (ηr), paraffin content (CPCM) and encapsulation efficiency (EE).
4.3. Conclusions
Microcapsules based on styrene crosslinked by divinylbenzene containing Rubitherm®RT27 were fabricated by suspension-like polymerization in the presence of a diluent in order to reduce the reaction rate. The effect of agitation rate, the kind of suspending agent and its concentration, the PCM content and the mass ratios of the continuous/discontinuous phases and porogen/PCM on the microencapsulation process were studied.
It was found that optimizing the concentration of PVP allowed an increase of the encapsulation efficiency and the yield of the process while keeping the physical characteristics of the microcapsules. Nevertheless, this parameter must be combined with a proper agitation rate (800 rpm), an optimum mass ratio between the continuous/discontinuous phases (3.24), a porogen/PCM mass ratio (1.92) and a PCM percentage of 68.5% with respect to the total monomers and paraffin. 5.03% PVP was selected as the best concentration to accomplish the production of microcapsules containing Rubitherm®RT27 as a phase change material.
This optimization procedure allowed the synthesis of microcapsules that can be utilized in building materials that provide a comfortable indoor temperature, due to their
123 Chapter 4
morphology, high thermal energy storage capacity (> 100 J g-1) and an appropriate melting temperature (24°C).
Study of interfacial properties were performed in order to explain how the differences in morphology, particle size and encapsulation efficiency depend on the type and amount of the suspending agent. Based on spreading coefficient calculations, a core- shell morphology was expected in most cases. However, the predictions were not accurate with respect to the appearance of the microcapsules. A matrix structure was observed for all microcapsules obtained for the tested suspending agents and for different amounts of PVP when a P(St-DVB) shell was used. A core-shell morphology was favored when PNC-HEMA was added. Accordingly, the morphology is controlled by the polarity of the shell rather than the type and amount of suspending agent. Optimization of these parameters is crucial for a proper dispersion of the phases. When the PVP concentration was varied, the encapsulation efficiency increased slightly with decreasing surface polarity of the water phase, which also became closer to the surface polarity of the polymer. The encapsulation efficiency increased significantly when PNC-HEMA was utilized. Therefore, conditions in which thermodynamic aspect favor multi-nucleus structure might improve encapsulation efficiency.
An important incorporation of the surfactant agent into the thermoregulating microcapsules was confirmed. It was found that PVP is retained by the P(St-DVB) microcapsules. The distribution of the PVP between the solid and liquid phases at the end of the polymerization process follows a Langmuir type trend. The PVP distribution data were fitted by the Langmuir model obtaining a value of 192.9 g kg-1 and 0.18 m3 kg-1 for the maximum retention capacity and the equilibrium constant, respectively. The adsorption of surfactant agents on polymeric materials has been reported previously by other authors [13,14] but as far as we know not quantified and modeled for thermoregulating microcapsules.
Finally, the robustness of the process was checked at pilot plant scale, obtaining more monodisperse materials, a latent heat of 96.1 J g-1, an encapsulation efficiency of 79.0% and a microcapsules yield of 80.0%. These characteristics were similar to those obtained at laboratory scale, although the particle size distribution was improved at pilot plant scale.
124 Optimization of microcapsules recipe and properties
References
1. Zhang H, Wang X (2009) Fabrication and performances of microencapsulated phase change materials based on n-octadecane core and resorcinol-modified melamine– formaldehyde shell. Colloids and Surfaces A: Physicochemical and Engineering Aspects 332 (2-3):129-138. doi:10.1016/j.colsurfa.2008.09.013 M4 - Citavi 2. Khakzad F, Alinejad Z, Shirin-Abadi AR, Ghasemi M, Mahdavian AR (2014) Optimization of parameters in preparation of PCM microcapsules based on melamine formaldehyde through dispersion polymerization. Colloid and Polymer Science 292 (2):355-368. doi:10.1007/s00396-013-3076-9 3. Sánchez L, Sánchez P, Lucas A, Carmona M, Rodríguez JF (2007) Microencapsulation of PCMs with a polystyrene shell. Colloid and Polymer Science 285 (12):1377-1385. doi:10.1007/s00396-007-1696-7 4. Sánchez L, Sánchez P, Carmona M, Lucas A, Rodríguez JF (2008) Influence of operation conditions on the microencapsulation of PCMs by means of suspension-like polymerization. Colloid and Polymer Science 286 (8-9):1019-1027. doi:10.1007/s00396- 008-1864-4 5. Sánchez L, Lacasa E, Carmona M, Rodríguez JF, Sánchez P (2008) Applying an Experimental Design to Improve the Characteristics of Microcapsules Containing Phase Change Materials for Fabric Uses. Industrial & Engineering Chemistry Research 47 (23):9783-9790. doi:10.1021/ie801107e 6. Huang J, Wang T, Zhu P, Xiao J (2013) Preparation, characterization, and thermal properties of the microencapsulation of a hydrated salt as phase change energy storage materials. Thermochimica Acta 557 (Supplement C):1-6. doi:https://doi.org/10.1016/j.tca.2013.01.019 7. Lashgari S, Arabi H, Mahdavian AR, Ambrogi V (2017) Thermal and morphological studies on novel PCM microcapsules containing n-hexadecane as the core in a flexible shell. Applied Energy 190:612-622 8. Qiu X, Li W, Song G, Chu X, Tang G (2012) Fabrication and characterization of microencapsulated n-octadecane with different crosslinked methylmethacrylate-based polymer shells. Solar Energy Materials and Solar Cells 98 (Supplement C):283-293. doi:https://doi.org/10.1016/j.solmat.2011.11.018
125 Chapter 4
9. Sánchez-Silva L, Rodríguez JF, Romero A, Borreguero AM, Carmona M, Sánchez P (2010) Microencapsulation of PCMs with a styrene-methyl methacrylate copolymer shell by suspension-like polymerisation. Chemical Engineering Journal 157 (1):216-222. doi:10.1016/j.cej.2009.12.013 10. Chang CC, Tsai YL, Chiu JJ, Chen H (2009) Preparation of phase change materials microcapsules by using PMMA network-silica hybrid shell via sol-gel process. Journal of Applied Polymer Science 112 (3):1850-1857. doi:10.1002/app.29742 11. Qiu X, Song G, Chu X, Li X, Tang G (2013) Microencapsulated n-alkane with p(n- butyl methacrylate-co-methacrylic acid) shell as phase change materials for thermal energy storage. Solar Energy 91 (Supplement C):212-220. doi:https://doi.org/10.1016/j.solener.2013.01.022 12. Tabor RF, Eastoe J, Dowding PJ (2010) A two-step model for surfactant adsorption at solid surfaces. Journal of Colloid and Interface Science 346 (2):424-428. doi:10.1016/j.jcis.2010.03.047 13. Esumi K, Takamine K, Ono M, Osada T, Ichikawa S (1993) The Interaction of Poly(vinylpyrrolidone) and Solid Particles in Ethanol. Journal of Colloid and Interface Science 161 (2):321-324. doi:10.1006/jcis.1993.1473 14. Geffroy C, Cohen Stuart MA, Wong K, Cabane B, Bergeron V (2000) Adsorption of Nonionic Surfactants onto Polystyrene. Langmuir 16 (16):6422-6430. doi:10.1021/la0000769 15. Smith JN, Meadows J, Williams PA (1996) Adsorption of Polyvinylpyrrolidone onto Polystyrene Latices and the Effect on Colloid Stability. Langmuir 12 (16):3773-3778. doi:10.1021/la950933m 16. Alcázar Á, Lucas A, Carmona M, Rodríguez JF (2011) Synthesis of sulphonated microcapsules of P(St–DVB) containing di(2-ethylhexyl)phosphoric acid. Reactive and Functional Polymers 71 (8):891-898. doi:10.1016/j.reactfunctpolym.2011.05.008 17. Shirin-Abadi AR, Khoee S, Rahim-Abadi MM, Mahdavian AR (2014) Kinetic and thermodynamic correlation for prediction of morphology of nanocapsules with hydrophobic core via miniemulsion polymerization. Colloids and Surfaces A: Physicochemical and Engineering Aspects 462 (Supplement C):18-26. doi:https://doi.org/10.1016/j.colsurfa.2014.08.009
126 Optimization of microcapsules recipe and properties
18. Torza S, Mason SG (1970) Three-phase interactions in shear and electrical fields. Journal of Colloid and Interface Science 33 (1):67-83. doi:http://dx.doi.org/10.1016/0021-9797(70)90073-1 19. Sánchez-Silva L, Rodríguez JF, Sánchez P (2011) Influence of different suspension stabilizers on the preparation of Rubitherm RT31 microcapsules. Colloids and Surfaces A: Physicochemical and Engineering Aspects 390 (1-3):62-66. doi:10.1016/j.colsurfa.2011.09.004 20. Al-Shannaq R, Farid M, Al-Muhtaseb S, Kurdi J (2015) Emulsion stability and cross- linking of PMMA microcapsules containing phase change materials. Solar Energy Materials and Solar Cells 132:311-318. doi:http://dx.doi.org/10.1016/j.solmat.2014.08.036 21. Luo Y, Zhou X (2004) Nanoencapsulation of a hydrophobic compound by a miniemulsion polymerization process. Journal of Polymer Science Part A: Polymer Chemistry 42 (9):2145-2154 22. Wang J-M, Zheng W, Song Q-W, Zhu H, Zhou Y (2009) Preparation and Characterization of Natural Fragrant Microcapsules. Journal of Fiber Bioengineering and Informatics 1 (4):293-300. doi:10.3993/jfbi03200907 23. Zhang XX, Fan YF, Tao XM, Yick KL (2004) Fabrication and properties of microcapsules and nanocapsules containing n-octadecane. Materials Chemistry and Physics 88 (2-3):300-307. doi:10.1016/j.matchemphys.2004.06.043 24. Bolten D, Türk M (2011) Experimental Study on the Surface Tension, Density, and Viscosity of Aqueous Poly(vinylpyrrolidone) Solutions. Journal of Chemical & Engineering Data 56 (3):582-588. doi:10.1021/je101277c 25. Liu Q, Duan Y, Shen S, Zhou Z (2011) A simple route to prepare pomegranate-like polystyrene-based microspheres with high porosity. Polymer International 60 (9):1287- 1290. doi:10.1002/pi.3138 26. Ma G-H, Sone H, Omi S (2004) Preparation of Uniform-Sized Polystyrene−Polyacrylamide Composite Microspheres from a W/O/W Emulsion by Membrane Emulsification Technique and Subsequent Suspension Polymerization. Macromolecules 37 (8):2954-2964. doi:10.1021/ma035316g M4 - Citavi
127 Chapter 4
27. Li MG, Zhang Y, Xu YH, Zhang D (2011) Effect of different amounts of surfactant on characteristics of nanoencapsulated phase-change materials. Polymer Bulletin 67 (3):541-552. doi:10.1007/s00289-011-0492-1 28. Chaiyasat P, Suzuki T, Minami H, Okubo M (2009) Thermal properties of hexadecane encapsulated in poly(divinylbenzene) particles. Journal of Applied Polymer Science 112 (6):3257-3266. doi:10.1002/app.29648 29. Sanchez-Silva L, Tsavalas J, Sundberg D, Sánchez P, Rodriguez JF (2010) Synthesis and characterization of paraffin wax microcapsules with acrylic-based polymer shells. Industrial & Engineering Chemistry Research 49 (23):12204-12211 30. Loxley A, Vincent B (1998) Preparation of Poly(methylmethacrylate) Microcapsules with Liquid Cores. Journal of Colloid and Interface Science 208 (1):49-62. doi:https://doi.org/10.1006/jcis.1998.5698 31. Pisani E, Fattal E, Paris J, Ringard C, Rosilio V, Tsapis N (2008) Surfactant dependent morphology of polymeric capsules of perfluorooctyl bromide: influence of polymer adsorption at the dichloromethane–water interface. Journal of Colloid and Interface Science 326 (1):66-71 32. van Zyl AJ, Sanderson RD, de Wet-Roos D, Klumperman B (2003) Core/shell particles containing liquid cores: morphology prediction, synthesis, and characterization. Macromolecules 36 (23):8621-8629 33. Feczkó T, Kardos AF, Németh B, Trif L, Gyenis J (2014) Microencapsulation of n- hexadecane phase change material by ethyl cellulose polymer. Polymer Bulletin 71 (12):3289-3304. doi:10.1007/s00289-014-1250-y 34. Rulison C (2000) Two-component surface energy characterization as a predictor of wettability and dispersability. KRUSS Application note AN213:1-22 35. Sánchez-Silva L, Carmona M, de Lucas A, Sánchez P, Rodríguez JF (2010) Scale-up of a suspension-like polymerization process for the microencapsulation of phase change materials. Journal of Microencapsulation 27 (7):583-593. doi:10.3109/02652048.2010.501394
128
5. Synthesis of green microcapsules containing fatty acids as PCM
SYNTHESIS OF GREEN
CHAPTER 5 MICROCAPSULES CONTAINING FATTY
ACIDS AS PCM
Based on article
. Szczotok, A. M.; Carmona, M.; Kjøniksen, A.-L.; Rodriguez, J. F., The role of radical polymerization in the production of thermoregulating microcapsules or polymers from saturated and unsaturated fatty acids. Journal of Applied Polymer Science 2018, 135 (10), 45970-n/a.
Synthesis of green microcapsules containing fatty acids as PCM
Abstract
The microencapsulation of linoleic, oleic, erucic and palmitic acids from styrene and divinylbenzene were studied by using the suspension-like polymerization technique. All materials exhibited a spherical shape, with a particle size between 166 and 416 μm. The phase change material (PCM) content decreased in the presence of double bonds in the fatty acid molecules. The latent heat of the microcapsules containing saturated palmitic acid was highest (123.30 J g-1). The lowest latent heat was observed for linoleic acid (LinA). Latent heat values from unsaturated fatty acid materials and the high particle yield indicated that these kinds of acids play two different roles in the radical polymerization processes; both as PCM and also as monomers. At high initiator concentrations, the unsaturated fatty acids were observed to react. This was confirmed by FTIR where the peak assigned to the C=C bond disappears in the spectrum of the microcapsules.
131 Chapter 5
5.1. Background
Fatty acids have similar heat of fusion as paraffins. They are produced from sustainable vegetable and animal sources, they are non-pollutant, non-toxic and considered as environmentally friendly with a suitable temperature range for building applications [1]. During phase change the volume change of fatty acids are less than for other PCMs [2]. The specific application of phase change materials (PCMs) is imposed by their phase transition temperature. Compounds with melting points above 30 °C are usually described for solar passive heating [3,4] and domestic water heating applications [5]. Fatty acids with a solid-liquid transition temperature between 20-30 °C are useful for building and greenhouse heating systems, depending on the climate requirements [6,7]. Finally, low melting temperature materials can find applications in cool storage systems [8].
In recent years, different studies containing pure alkanes [9-12] and commercial paraffins [13-16] were performed. Nevertheless, few researchers have shown interest in the application of fatty acids as PCMs. Saturated fatty acids such as stearic acid (SA), palmitic acid (PA), myristic acid (MA), lauric acid (LauA), caprylic acid (CA) and decanoic acid (DA) have been used. Wang et al. [17] and Alkan et al. [18] have prepared composites by entrapping SA, PA, MA and LauA into poly(methyl metacrylate) blends. The maximum entrapped proportion (51.8 wt.%) was obtained for the SA. Konuklu et al. microencapsulated DA with poly(melamine–urea–formaldehyde) [19], and saturated CA in urea-formaldehyde, melamine-formaldehyde [2], or polyethylacrylate shell with ethylene glycol dimethacrylate as a crosslinking agent. The best latent heat storage capacities for DA and CA were 88.0 and 93.9 J g-1, respectively [20]. Bao et al. [21] have synthesized microcapsules of LauA as the core and melamine/formaldehyde resin as the shell by in situ polymerization with a latent heat of 84.96 J g-1, which corresponds to 46.20% by weight of core content. Values close to that were reported by Konuklu et al. using a shell with the same characteristic. On the other hand, Özonur et al. [22] prepared microcapsules of natural coco fatty acid mixture by means of complex coacervation methods, using natural shell gelatin–gum Arabic. Finally, Giro-Paloma et al. [23] produced nanocapsules containing PA from poly(styrene-co-ethylacrylate) by (emulsion co-polymerization), with a thermal energy storage (TES) capacity of 97.93 J g-1.
132 Synthesis of green microcapsules containing fatty acids as PCM
It is possible to achieve form-stable and microencapsulated saturated fatty acids having a melting temperature ranged between 13 and 69 °C [2,17-23]. Form-stable PCM is a material which consist of the PCM that serves as a latent heat storage material and the polymer which acts as a supporting material, preventing leakage of the melted paraffin due to good structural strength. The linkage of the fatty acid to the polymer backbone allows to retain the phase transition thermal exchange properties of the fatty acid that is side attached to the polymer, while the polymer structure maintain the shape during the phase transition of the attached chain [24].
Microencapsulation of unsaturated fatty acids by spray-drying, complex coacervation and extrusion methods for food industry has been reported previously [25]. However, the shell materials used in these studies are easily degraded and therefore not suitable for building applications. Only studies about the behavior in bulk polymerization of unsaturated fatty acids with vinyl monomer in the presence of benzoyl peroxide (BPO) as radical initiator have been reported [26-28]. These studies demonstrated the reactivity of conjugated linoleic acid (LinA) to form copolymers. In addition, the significant role of oleic acid (OA) in the polymerization reactions was illustrated but without incorporation into the polymer product. Other authors have used oleic acid as a surfactant replacement in the emulsion polymerization of styrene without copolymerization. The non-reactivity of oleic acid was explained by Jeong et al. [29] and Laible et al. [30] by resonance stabilization of its radicals. Khan et al. [31] reported synthesis of nanostructured brush from polystyrene and carbon nanotubes (CNT) functionalized with oleic acid.
There is a lack of information regarding encapsulation possibilities of unsaturated fatty acids by suspension polymerization techniques. The influence of the chemical structure on physical and thermal properties of microcapsules in free radical polymerization has not been investigated previously [32]. Furthermore, preparation of microcapsules with poly(styrene-divinylbenzene) shell (P(St-DVB)) has not been reported. It is important to point out that a large number of unsaturated fatty acids exhibit an environmental temperature range that make them suitable for building applications. Particularly interesting are low melting point fatty acids that can be used in cool storage systems. Moreover, the commercial PCMs are based on fossil fuels, thus renewable products like fatty acids are important alternatives for sustainable energy saving.
133 Chapter 5
Due to the poor reactivity of unsaturated fatty acids, they are promising candidates for core materials in microcapsules produced by suspension polymerization. In this chapter, four different fatty acids: the two mentioned above (LinA and OA) and two additional ones (erucic acid (EA) and palmitic acid (PA)) were selected as examples of polyunsaturated, monounsaturated and saturated acids. These PCMs have melting temperatures between -13.55 and 62.89 ºC and heat of fusion ranged between 87.74 and 191.50 J g-1. The encapsulation of this kind of PCMs were compared with the microencapsulation of commercial paraffins (Rubitherm®RT27, Rubitherm®RT21 and Rubitherm®RT6), and with the results reported in literature for saturated acids. Finally, the reactivity of the polyunsaturated fatty acid was studied in bulk radical polymerization reactions using vinyl monomers.
5.2. Materials
Microcapsules were prepared by a suspension-like polymerization technique, following a recipe obtained in the optimization process reported in Chapter 4 which is also shown in Table 5.1.
Table 5.1. Recipe for the synthesis of microcapsules containing PCM and experimental conditions.
Raw Material wt.% Continuous phase Water (Mili-Q) 72.0 Polyvinylpyrrolidone (PVP) 5.0 Discontinuous phase Phase Change Material (PCM) 7.0 Styrene (St) 2.0 Divinylbenzene (DVB) 2.0 Toluene 13.0 Benzoyl Peroxide (BPO) 1.0 Reaction temperature (°C) 80 Stirring rate (rpm) 800 Reaction time (h) 5
134 Synthesis of green microcapsules containing fatty acids as PCM
5.3. Results
Synthesis of microcapsules in suspension-like polymerization
The microencapsulation of fatty acids with a P(St-DVB) shell were achieved by the suspension-like polymerization technique. In this study, fatty acids presenting different chemical characteristics and thermal properties were used. The chemical structures of fatty acids, melting points and latent heats of PCMs are summarized in Table 5.2.
Table 5.2. Chemical structures of fatty acids and thermal properties of different PCMs.
Phase change Melting Latent Condensed structural formula material point (°C) heat (J g-1)
Linoleic acid (LinA) CH3(CH2)3(CH2CH=CH)2(CH2)7COOH -13.55 87.74
Oleic acid (OA) CH3(CH2)7CH=CH(CH2)7COOH 10.66 128.72
Erucic acid (EA) CH3(CH2)7CH=CH(CH2)11COOH 28.40 122.64
Palmitic acid (PA) CH3(CH2)14COOH 62.89 191.53 Rubitherm®RT6 - 6.79 162.21 Rubitherm®RT21 - 17.31 145.04 Rubitherm®RT27 - 24.60 171.21
The physical aspect of the microcapsules containing fatty acids is shown in Figure 5.1. The nomenclature used for the different samples will be the following: for microcapsules containing linoleic acid (MC(LinA)), oleic acid (MC(OA)), erucic acid (MC(EA)) and palmitic acid (MC(PA)). As can be seen, the microcapsules prepared with fatty acids have regular and spherical shape with smooth surface, although, MC(LinA) and MC(PA) were slightly agglomerated.
135 Chapter 5
(a) (b)
(c) (d)
(e) (f)
(g) (h)
Figure 5.1. SEM images of microcapsules containing different fatty acids with magnification x100 left and x400 right (a) MC(LinA); (b) x400 MC(LinA); (c) MC(OA); (d) x400 MC(OA); (e) MC(EA); (f) x400 MC(EA); (g) MC(PA); (h) x400 MC(PA).
136 Synthesis of green microcapsules containing fatty acids as PCM
The volume and number particle size distribution (PSD) of the microcapsules are shown in Figure 5.2. The particles prepared with LinA exhibits the widest PSD while the one synthesized with PA has the narrowest PSD. The volume and number average size depend on the kind of fatty acid employed and follow the order: MC(PA) 14 25 MC(PA) (a) (b) MC(PA) 12 MC(EA) MC(EA) MC(OA) 20 MC(OA) 10 MC(LinA) MC(LinA) 8 15 6 10 Volume (%) Number (%) 4 5 2 0 0 0 200 400 600 800 1000 1200 1400 0 10 20 30 40 50 200 400 600 800 Particle size (m) Particle size (m) Figure 5.2. Particle size distribution for microcapsules containing different fatty acids (a) volume average, (b) number average. Table 5.3 shows the latent heat and melting points of the materials obtained using fatty acids and paraffins Rubitherm®RT6 (MC(RT6)), Rubitherm®RT21 (MC(RT21)) and Rubitherm®RT27 (MC(RT27)), determined by utilizing DSC. Table 5.3. Properties of microcapsules using different core materials. -1 Sample Tm (°C) ΔHMC (J g ) CPCM (%) Microcapsules containing fatty acids MC(LinA) 15.68 3.54 4.03 MC(OA) 5.80 47.09 36.59 MC(EA) 25.56 63.86 52.09 MC(PA) 62.08 123.30 64.39 137 Chapter 5 Table 5.3. Properties of microcapsules using different core materials (continuation). -1 Sample Tm (°C) ΔHMC (J g ) CPCM (%) Microcapsules containing paraffins MC(RT6) 6.41 87.44 53.91 MC(RT21) 15.58 94.40 65.10 MC(RT27) 24.39 101.80 59.46 The latent heat of the materials increases with the melting point of the fatty acids. -1 Accordingly, MC(PA) shows the highest ΔHMC (123.3 J g ) and CPCM (64.39%). The PCM content and latent heat of the MC(PA) are higher than reported by Giro-Paloma et al. [23] encapsulating PA in poly(styrene-co-ethylacrylate) (47.79 wt.% and 97.93 J g-1). Interestingly, the heat storage capacity achieved by using unsaturated fatty acids decreases markedly for the monounsaturated fatty acids (OA and EA), and is reduced to almost nothing for the polyunsaturated LinA. Nevertheless, Jimenez et al. [33] were able to encapsulate 85% of LinA in a whey protein concentrate shell by using physical spray- drying. Accordingly, the presence of double bonds in the fatty acid affect the thermal properties of the microcapsules, probably because they are participating in the shell polymerization process. It seems clear that the greatest latent heat of PA is related to its saturated nature. Edirisinghe et al. [34] found that the reactivity of unsaturated double bonds in bulk radical polymerization depends on the number of double bonds and the chain length of the fatty acid. Keeping this in mind, unsaturated fatty acids can perform two different roles in the radical polymerization process, one as PCM and another as monomer. The proposed mechanism for the unsaturated fatty acids reaction is depicted in Figure 5.3. The low latent heat of the microcapsules containing non-conjugated LinA indicates that this fatty acid acts mainly as a crosslinking agent, forming part of the polymer shell. Accordingly, the presence of double bonds make that the fatty acid behave as a monomer instead of remaining as the reaction media for the vinyl monomers. This alter the TES properties of the original acid. Comparing the values of ΔHMC and CPCM for MC(OA) and MC(EA), it is evident that the latent heat of the microcapsules rises significantly with increasing chain length (EA > OA). A longer chain will have higher hydrophobicity and melting temperature of the PCM. The wt. fraction of unreacted PCM 138 Synthesis of green microcapsules containing fatty acids as PCM in MC(OA) and MC(EA) is different due to higher reactivity of OA compared with EA. Since hydrophobic compounds have a higher interfacial tension against water, they prefer to stay in the inner core of the droplet. Accordingly, based on the hydrophobic properties, EA is expected to be easier to encapsulate than OA [35]. Figure 5.3. Proposed polymerization reaction of a monounsaturated fatty acid. 139 Chapter 5 In order to check the influence of the melting temperature of the PCM on the encapsulation procedure, three additional materials with different melting temperatures (Rubitherm®RT6, Rubitherm®RT21 and Rubitherm®RT27) were encapsulated. Figure 4 shows the relation between the latent heat of microcapsules (MC) and form-stable materials (FS) with the melting temperature of the core material. All the reported microcapsules and form-stable materials containing fatty acids were included. PCM content (%) -20 -10 0 10 20 30 40 50 60 70 80 Melting temperature PCM content 240 dependency dependency 240 MC(DA) [19] MC(DA) [19] 220 MC(LauA) [20] MC(LauA) [20] 220 FS(PA) [18] FS(PA) [18] 200 FS(LauA) [18] FS(LauA) [18] 200 MC with paraffin MC with paraffin 180 MC with fatty acids MC with fatty acids 180 ) ) MC(CA) [29] MC(CA) [29] -1 -1 FS(SA) [18] FS(SA) [18] 160 FS(MA) [18] FS(MA) [18] 160 140 140 MC(PA) 120 120 MC(RT21) MC(RT21)MC(RT27) 100 MC(RT6) MC(RT6) 100 MC(RT27) 80 80 g (J heat Latent Latent heat (J g (J heat Latent MC(EA) 60 MC(OA) MC(EA) 60 MC(OA) 40 40 20 20 MC(LinA) MC(LinA) 0 0 -20 -10 0 10 20 30 40 50 60 70 80 Melting temperature of PCM (C) Figure 5.4. Influence of melting temperature and content of the PCM on the latent heat of microcapsules (MC) and form stable materials (FS) with stearic acid (SA), palmitic acid (PA), myristic acid (MA), lauric acid (LauA), caprylic acid (CA), decanoic acid (DA), oleic acid (OA), erucic acid (EA) and linoleic acid (LinA). As can be seen, the latent heat of both fatty acids and paraffins increases with the melting temperature of the PCM. It is also observed, although depending on the material type that a lower amount of PCM is encapsulated when the melting temperature decreases. In this way, the lowest value of latent heat observed in the case of MC(LinA) can be attributed to its higher reactivity. Thus, the microencapsulation of paraffins by this 140 Synthesis of green microcapsules containing fatty acids as PCM technique is strongly dependent on the melting temperature of the core material, whereas the encapsulation of fatty acids is affected by both reactivity and melting temperature. This also indicates that the microencapsulation process of a saturated fatty acid is equal to that of a paraffin. On the other hand, the reported data for thermoregulating materials containing unsaturated fatty acids obtained for the form-stable method resulted in materials with a larger latent heat than achieved from microencapsulation. This is because the form-stable PCM works as a mixture, while the latent heat of the microcapsules is dependent on the shell thickness. According to Figure 5.4, thermoregulating materials containing unsaturated fatty acids can be produced by using the same PCM content, with the difference that the latent heat only dependent on the melting temperature. Finally, from Figure 5.4 it is possible to affirm that saturated acids can be microencapsulated following the same trend as paraffins, indicating that it is inert in this process contrary to unsaturated fatty acids. The results obtained during synthesis with paraffins show that the highest PCM content for Rubitherm®RT21 was 65.10%, which is very close to the 64.39% obtained when the PA was encapsulated. These results are higher than those reported by Sanchez et al. [15] encapsulating 10.92 of Rubitherm®RT20 and 28.69% Rubitherm®RT27 in a polystyrene shell by the same methodology. Bayes-Garcia et al. [13] developed microcapsules with Rubitherm®RT27 in Agar-Agar/Arabic Gum and Sterilized Gelatin/Arabic Gum shells with the PCM contents 48 and 49%. Accordingly, the technique of the synthesis, recipe and shell materials have an important influence on the final thermal properties of the products. Although it seems evident that the double bonds of the unsaturated fatty acids are interacting with monomers in the polymerization process, solid scientific evidences supported by the characterization techniques will be provided for the definitive confirmation. TGA thermograms of P(St-DVB) without PCM and microcapsules containing fatty acids are shown in Figure 5.5. Pure fatty acids evaporate at lower temperatures than the copolymer. The evaporation trend PA>LinA>OA>EA agrees with the expected volatility for the fatty acids with respect to their boiling temperatures 256.39, 268.83, 281.81 and 296.88 °C, respectively. It can be also seen that thermal degradation of the P(St-DVB) copolymer and pure fatty acids occurs in only one weight loss. 141 Chapter 5 Figure 5.5. TG and DTG curves for P(St-DVB), pure fatty acids and their microcapsules. 142 Synthesis of green microcapsules containing fatty acids as PCM For the microcapsules, the weight loss depends on the encapsulated fatty acid. Microcapsules containing PA and LinA exhibit two steps, whereas three steps are observed for microcapsules containing OA and EA. All the materials from fatty acids, except PA, present a wide, smooth and continuous weight loss in the temperature range. This is due to the evaporation of pure fatty acid and the degradation of P(St-DVB). This weight loss could be caused by the formation of by-products having a higher vapor pressure than pure fatty acids. Wilson et al. [36] reported the decreasing vapor pressure of some fatty acids with increasing molecular weight. In homologous organic compounds, vapor pressure at any temperature decreases with increasing molecular weight [37]. Kaya et al. [38] reported higher resist of oligomer against temperature than monomer due to long conjugated band systems. Finally, Yusof et al. [39] commented lower thermal stability of monomers compare to polymers caused by lower molecular weight of the monomers. This confirms the reactivity of unsaturated bonds in the radical polymerization process. The results observed by using OA do not agree with the results reported by Roberge et al., [27] where the fatty acid was not incorporated into final polymer product. However, OA had a significant effect on the polymerization kinetics, due to resonance stabilization of radicals from the oligomers which was beneficial for conjugated LinA initiation. Hence, fatty acids with double bonds can act as monomers in radical polymerization processes initiated by BPO. Accordingly, the suspension-like polymerization technique is not a suitable microencapsulation technique for production of microcapsules having a large latent heat and containing fatty acids with double bonds as core materials. Figure 5.6 shows the FT-IR spectra for pure fatty acids, P(St-DVB) and their microcapsules. Vibrations of alkene C-H stretching, aliphatic C-H, carbonyl group and C=C alkene bending can be observed in the bands 3010, 2850 – 2950, 1700 and 948 cm-1, respectively, for the pure fatty acids [27]. The IR spectrum of P(St-DVB) exhibited an absorption band at 3020 cm-1 attributed to aromatic C-H stretching vibration. In addition, peaks located between 1600 and 1440 cm-1 supported the presence of C=C phenyl stretching. The bands at 750 cm-1 and 695 cm-1 corresponds to aromatic C-H bending of benzene derivative [40,41]. In the spectrum of microcapsules, the vibrations for aliphatic C-H and carbonyl groups can be observed as well. Furthermore, the peak 143 Chapter 5 assigned for C=C alkene bending disappear in case of LinA and OA and decreases the intensity for EA, confirming the contribution in the shell formation. (a) (b) 144 Synthesis of green microcapsules containing fatty acids as PCM (c) Figure 5.6. Infrared spectra of fatty acids, P(St-DVB) and their microcapsules a) LinA; b) OA; c) EA. Reactivity of unsaturated fatty acids With the purpose of verifying the reactivity of LinA, OA and EA in presence of BPO, several bulk polymerization experiments were carried out. Initially, the homopolymerization of LinA, OA and EA initiated by using BPO for checking the formation of homopolymers (PLinA, POA and PEA) from the original fatty acids were performed. Table 5.4. Thermogravimetric parameters for unsaturated fatty acids and their homopolymers. Td (°C) Sample T onset (°C) ΔT (°C) Char (%) I II LinA 237.90 268.48 - 38.65 1.43 PLinA 237.24 274.92 448.65 233.49 0.29 OA 242.55 273.99 - 41.06 0.22 POA 244.10 274.79 440.91 215.05 0.12 EA 264.04 297.22 - 46.80 2.49 PEA 266.53 299.56 449.81 200.86 0.10 145 Chapter 5 100 3 I LinA 75 PLinA DTG LinA 2 50 DTG PLinA 1 25 II 0 0 C) 100 3 OA 75 I POA DTG OA 2 50 DTG POA 1 25 II Weight (%) Weight 0 0 100 3 (%/ Weight Deriv. EA 75 I PEA DTG EA 2 50 DTG PEA 1 25 II 0 0 0 100 200 300 400 500 600 Temperature (C) Figure 5.7. TGA and DTG thermograms for unsaturated fatty acids and their homopolymers. Figure 5.7 shows that the volatility trend of homopolymers is in agreement with the results obtained for pure fatty acids, where the lowest stability was exhibited by LinA, then OA and EA. Table 5.4 presents onset temperature, mean degradable temperature, difference between onset and endset temperature and char residue of unsaturated fatty acids and their homopolymers. As expected, two weight losses for all homopolymerization products were clearly noted. The first one corresponds with the evaporation of the pure fatty acids (peak I), while the second one is due to the degradation of oligomers or polymers produced from the unsaturated fatty acids (peak II), constituting 14.11, 15.71 and 46.02% for PEA, POA and PLinA, respectively. A shifting of the evaporation temperature of the pure fatty acids can be observed, indicating the possible absorption of those compounds in the liquid polymers (Td). ΔT increases in the case of polymer due to formation of products with wide range of molecular weights, indicating a large polydispersity of the material [39]. The formation of char is very low due to the absence of promotors like nitrogen and phosphor in the chemical structure of the fatty acids. Accordingly, these materials do not have fire retardant properties. 146 Synthesis of green microcapsules containing fatty acids as PCM PLinA 0.6 POA PEA 0.4 0.2 0.0 Heat flow (W/g) Heat -0.2 -0.4 -0.6 -40 -20 0 20 40 60 80 Temperature (C) Figure 5.8. DSC thermograms for PLinA, POA and PEA. The total amount of non-reactive fatty acids can be obtained by DSC (Figure 5.8), by examining the endothermic and exothermic values of the latent heat of the products, during the melting and crystallization processes, respectively. The conversion of fatty 푃 acid (휒퐹퐴) in the production of each polymer (Pi) was estimated by the following Equation 5.1. ∆HP Reactants 휒푃 (%) = (1 − 푖 × ) × 100% (5.1) 퐹퐴 ∆H FA FA푖 푖 where ∆퐻푃푖, ∆퐻퐹퐴푖 are the average latent heats from the endothermic and exothermic values of the product and the original fatty acid, respectively. The weight of fed raw materials is named Reactants and FAi is the weight of initial fatty acid added. A small supercooling effect was observed for the fatty acids, resulting in a shift of the peak positions between the heating and cooling curves [42]. From the experimental data, the conversions of EA, OA and LinA were 31.84, 53.49 and 67.47%, respectively. The conversion confirm the hypothesis that the most reactive fatty acid is LinA followed by OA. From DSC analyses it is possible to obtain the non-affected amount of fatty acids by the oligomerization process, which can be 147 Chapter 5 obtained from the conversion values. The amount of non-reactive fatty acids were 68.16, 46.51 and 32.53% for EA, OA and LinA, respectively. All of these values are lower than those obtained from the TGA (80.38, 79.32 and 63.32% for PEA, POA and PLinA, respectively), indicating that fatty acid absorption occurs in presence of the liquid homopolymer. Accordingly, both TGA and DSC thermal analyses confirmed the homopolymerization process of unsaturated fatty acids initiated by benzoyl peroxide. Once the reactivity of the unsaturated fatty acid was confirmed, the effect of the presence of styrene and DVB as vinyl monomers giving terpolymerization products P(LinA+St+DVB), P(OA+St+DVB) and P(EA+St+DVB) was also studied. Products from the bulk polymerization of OA and EA with vinyl monomers consisted of a large amount of solid material and a thin layer of liquid, whereas in the case of the LinA, the whole product was solid. Terpolymerization solid products are shown in Figure 5.9. 100 2.0 1.8 80 1.6 60 P(LinA+St+DVB) DTG P(LinA+St+DVB) 1.4 C) P(OA+St+DVB) 40 DTG P(OA+St+DVB) 1.2 P(EA+St+DVB) 1.0 20 DTG P(EA+St+DVB) 0.8 Weight (%) Weight 0 0.6 Deriv. weight (%/ weight Deriv. -20 0.4 0.2 -40 0.0 0 100 200 300 400 500 600 Temperature (C) Figure 5.9. TG and DTG thermograms of solid P(LinA+St+DVB), P(OA+St+DVB) and P(EA+St+DVB). Thermograms show that the thermal evaporation of the material increases in presence of vinyl monomers. The TGA of P(OA+St+DVB) and P(EA+St+DVB) copolymer exhibits weight losses compatible with the evaporation of pure fatty acids and 148 Synthesis of green microcapsules containing fatty acids as PCM the thermal degradation of the polymer. The decomposition of the P(LinA+St+DVB) was observed in a wide temperature range, as result of a high polydispersity of the final polymer and the high conversion of this fatty acid. The reactivity of LinA was previously reported by Roberge et al. [27,26] using styrene and butyl acrylate in presence of benzoyl peroxide as initiator at 80 °C. Due to the high conversion of the LinA, it was chosen for further studies of its reactivity with the pure monomers. Copolymers P(LinA+St) and P(LinA+DVB) were produced, with the same PCM/monomer mass ratio. Figure 5.10 shows the DSC results for all LinA products from the bulk polymerization process. 0.8 LinA 0.6 PLinA P(LinA+St) 0.4 P(LinA+DVB) P(LinA+St+DVB) 0.2 0.0 -0.2 Heat flow (W/g) Heat -0.4 -0.6 -0.8 -40 -20 0 20 40 60 80 Temperature (C) Figure 5.10. DSC thermograms LinA, PLinA, P(LinA+St), P(LinA+DVB) and P(LinA+St+DVB). The latent heats for the obtained polymers PLinA, P(LinA+St), P(LinA+DVB) and P(LinA+St+DVB) were 26.64, 12.34, 2.46 and 2.69 J g-1, which leads to LinA P conversions of 67.47, 69.87, 93.99 and 95.20%, respectively. As can be seen, the χLinA increases significantly with the use of vinyl monomers and mainly with DVB. The values of conversion are higher than reported by Roberge et al. [28] This differences in the conversion values could be due to the use of a higher mass ratio between initiator and 149 Chapter 5 monomers in this study compared to that used by Roberge et al. [28] Accordingly, the double bonds of the fatty acids are activated by the radical initiator benzoyl peroxide, causing a depletion in the thermal energy storage properties. The molecular weight distributions of the soluble fraction of polymers in THF from the above products were determined by GPC (Figure 5.11). The molecular weight of the soluble copolymers (P(LinA+St) and P(LinA+DVB)) and the soluble fraction of terpolymer (P(LinA+St+DVB)) exhibited high values. The PLinA exhibits a molecular weight very close to that of original LinA - retention volume 16.2 ml - but with a higher amount of oligomers. The peak of LinA disappears completely in the case of the soluble part of the terpolymer P(LinA+St+DVB). As noted above, LinA exhibited 95.20% conversion in the production of the terpolymer P(LinA+St+DVB). This can be explained by the presence of another reactive unsaturated fatty acid (OA) as the main impurity of the used LinA. Accordingly, the GPC analyses also confirm that the reactivity of the unsaturated acids in radical polymerization process can be enhanced by using vinyl monomers and BPO concentration of 7 wt.% with respect to the monomers. The obtained elution is not unimodal because the products have large polydispersity (Table 5.5). In Figure 5.7 we can observe the TG thermograms for the terpolymer containing LinA, the weight loss is broad which can indicate a material with high polydispersity. Table 5.5. Number- and Weight-Average Molecular Weight for LinA homopolymer, copolymers and terpolymer. -1 -1 Sample Mn (g mol ) Mw (g mol ) Mw/Mn PLinA 915 924 1.00 P(LinA+St) 1775 2058 1.16 P(LinA+DVB) 8682 15048 1.73 P(LinA+St+DVB) 8854 15407 1.74 150 Synthesis of green microcapsules containing fatty acids as PCM PLinA P(LinA+St) P(LinA+DVB) P(LinA+St+DVB) P(LinA) P(LinA+St) P(LinA+St+DVB) P(LinA+DVB) 7 8 9 10 11 12 13 14 15 16 17 Retention Volume (mL) 10 11 12 13 14 15 16 17 Retention Volume (mL) Figure 5.11. GPC of the synthesized polymers and pure LinA. 5.4. Conclusions Microcapsules containing PA or paraffins Rubitherm®RT6, Rubitherm®RT21 and Rubitherm®RT27 as core materials and P(St-DVB) as the polymeric shell were successfully produced by a suspension-like polymerization technique. SEM and PSD analysis confirmed the spherical shape and uniform particle size of MC(PA). It was found that the latent heat of microcapsules containing fatty acids as well as in the case of paraffins, increased with the melting temperature of the core material. However, the encapsulation of fatty acids by radical polymerization was strongly dependent on the presence and number of double bonds in the core material. Microcapsules containing LinA exhibited the lowest content of PCM and latent heat, indicating that this acid has a high reactivity in the polymerization medium due to the two double bonds in its chemical structure. DSC and TGA analyses confirmed that unsaturated fatty acids can be considered as a monomer in the free radical polymerization process initialized by benzoyl peroxide, which causes a decrease in the PCM efficiency. The conversions of fatty acids in the homopolymerization processes were 67.47, 53.49 and 31.84% for PLinA, POA and PEA, respectively. Fatty acids containing double bonds can play two different roles in 151 Chapter 5 processes that involve radical polymerization: as core materials and monomers. Accordingly, this technique was suitable to encapsulate saturated fatty acids and paraffins having a latent heat and a PCM content higher than 100 J g-1 and 64 wt.%, respectively. Nevertheless, it offered a poor performance for the encapsulation of unsaturated fatty acids. 152 Synthesis of green microcapsules containing fatty acids as PCM References 1. Kośny J (2015) PCM-enhanced building components: An application of phase change materials in building envelopes and internal structures. Springer. doi:10.1007/978-3-319- 14286-9 2. Konuklu Y, Unal M, Paksoy HO (2014) Microencapsulation of caprylic acid with different wall materials as phase change material for thermal energy storage. Solar Energy Materials and Solar Cells 120, Part B:536-542. doi:http://dx.doi.org/10.1016/j.solmat.2013.09.035 3. Alkan C, Sarı A, Karaipekli A, Uzun O (2009) Preparation, characterization, and thermal properties of microencapsulated phase change material for thermal energy storage. Solar Energy Materials and Solar Cells 93 (1):143-147. doi:10.1016/j.solmat.2008.09.009 4. Sari A, Kaygusuz K (2002) Thermal performance of palmitic acid as a phase change energy storage material. Energy Conversion and Management 43 (6):863-876. doi:http://dx.doi.org/10.1016/S0196-8904(01)00071-1 5. Hasan A, Sayigh AA (1994) Some fatty acids as phase-change thermal energy storage materials. Renewable Energy 4 (1):69-76. doi:http://dx.doi.org/10.1016/0960- 1481(94)90066-3 6. Sari A, Kaygusuz K (2002) Thermal performance of a eutectic mixture of lauric and stearic acids as PCM encapsulated in the annulus of two concentric pipes. Solar Energy 72 (6):493-504. doi:http://dx.doi.org/10.1016/S0038-092X(02)00026-9 7. Cao VD, Pilehvar S, Salas-Bringas C, Szczotok AM, Rodriguez JF, Carmona M, Al- Manasir N, Kjøniksen A-L (2017) Microencapsulated phase change materials for enhancing the thermal performance of Portland cement concrete and geopolymer concrete for passive building applications. Energy Conversion and Management 133:56-66. doi:https://doi.org/10.1016/j.enconman.2016.11.061 8. He B, Setterwall F (2002) Technical grade paraffin waxes as phase change materials for cool thermal storage and cool storage systems capital cost estimation. Energy Conversion and Management 43 (13):1709-1723. doi:http://dx.doi.org/10.1016/S0196- 8904(01)00005-X 153 Chapter 5 9. Shan XL, Wang JP, Zhang XX, Wang XC (2009) Formaldehyde-free and thermal resistant microcapsules containing n-octadecane. Thermochimica Acta 494 (1–2):104- 109. doi:http://dx.doi.org/10.1016/j.tca.2009.04.026 10. Yang R, Zhang Y, Wang X, Zhang Y, Zhang Q (2009) Preparation of n-tetradecane- containing microcapsules with different shell materials by phase separation method. Solar Energy Materials and Solar Cells 93 (10):1817-1822. doi:http://dx.doi.org/10.1016/j.solmat.2009.06.019 11. Tseng YH, Fang MH, Tsai PS, Yang YM (2005) Preparation of microencapsulated phase-change materials (MCPCMs) by means of interfacial polycondensation. Journal of Microencapsulation 22 (1):37-46. doi:10.1080/02652040400026558 12. Alay S, Alkan C, Göde F (2011) Synthesis and characterization of poly(methyl methacrylate)/n-hexadecane microcapsules using different cross-linkers and their application to some fabrics. Thermochimica Acta 518 (1–2):1-8. doi:http://dx.doi.org/10.1016/j.tca.2011.01.014 13. Bayés-García L, Ventolà L, Cordobilla R, Benages R, Calvet T, Cuevas-Diarte MA (2010) Phase Change Materials (PCM) microcapsules with different shell compositions: Preparation, characterization and thermal stability. Solar Energy Materials and Solar Cells 94 (7):1235-1240. doi:http://dx.doi.org/10.1016/j.solmat.2010.03.014 14. Sánchez-Silva L, Rodríguez JF, Romero A, Sánchez P (2012) Preparation of coated thermo-regulating textiles using Rubitherm-RT31 microcapsules. Journal of Applied Polymer Science 124 (6):4809-4818. doi:10.1002/app.35546 15. Sánchez L, Sánchez P, de Lucas A, Carmona M, Rodríguez JF (2007) Microencapsulation of PCMs with a polystyrene shell. Colloid and Polymer Science 285 (12):1377-1385. doi:10.1007/s00396-007-1696-7 16. Alkan C, Kaya K, Sarı A (2009) Preparation, Thermal Properties and Thermal Reliability of Form-Stable Paraffin/Polypropylene Composite for Thermal Energy Storage. Journal of Polymers and the Environment 17 (4):254. doi:10.1007/s10924-009- 0146-7 17. Wang Y, Xia TD, Feng HX, Zhang H (2011) Stearic acid/polymethylmethacrylate composite as form-stable phase change materials for latent heat thermal energy storage. Renewable Energy 36 (6):1814-1820. doi:http://dx.doi.org/10.1016/j.renene.2010.12.022 154 Synthesis of green microcapsules containing fatty acids as PCM 18. Alkan C, Sari A (2008) Fatty acid/poly(methyl methacrylate) (PMMA) blends as form-stable phase change materials for latent heat thermal energy storage. Solar Energy 82 (2):118-124. doi:http://dx.doi.org/10.1016/j.solener.2007.07.001 19. Konuklu Y, Paksoy HO, Unal M, Konuklu S (2014) Microencapsulation of a fatty acid with Poly(melamine–urea–formaldehyde). Energy Conversion and Management 80:382-390. doi:http://dx.doi.org/10.1016/j.enconman.2014.01.042 20. Konuklu Y (2014) Microencapsulation of phase change material with poly(ethylacrylate) shell for thermal energy storage. International Journal of Energy Research 38 (15):2019-2029. doi:10.1002/er.3216 21. Bao Y-H, Pan W, Wang T-W, Wang Z, Wei F-M, Xiao F (2011) Microencapsulation of Fatty Acid as Phase Change Material for Latent Heat Storage. Journal of Energy Engineering 137 (4):214-219. doi:doi:10.1061/(ASCE)EY.1943-7897.0000053 22. Özonur Y, Mazman M, Paksoy HÖ, Evliya H (2006) Microencapsulation of coco fatty acid mixture for thermal energy storage with phase change material. International Journal of Energy Research 30 (10):741-749. doi:10.1002/er.1177 23. Giro-Paloma J, Konuklu Y, Fernández AI (2015) Preparation and exhaustive characterization of paraffin or palmitic acid microcapsules as novel phase change material. Solar Energy 112:300-309. doi:http://dx.doi.org/10.1016/j.solener.2014.12.008 24. Sarı A (2004) Form-stable paraffin/high density polyethylene composites as solid– liquid phase change material for thermal energy storage: preparation and thermal properties. Energy Conversion and Management 45 (13):2033-2042. doi:https://doi.org/10.1016/j.enconman.2003.10.022 25. Kaushik P, Dowling K, Barrow CJ, Adhikari B (2015) Microencapsulation of omega- 3 fatty acids: A review of microencapsulation and characterization methods. Journal of Functional Foods 19, Part B:868-881. doi:https://doi.org/10.1016/j.jff.2014.06.029 26. Roberge S, Dubé MA (2015) Bulk Copolymerization of Conjugated Linoleic Acid With Styrene and Butyl Acrylate: Reactivity Ratio Estimation. Journal of Macromolecular Science, Part A 52 (12):961-970. doi:10.1080/10601325.2015.1095594 27. Roberge S, Dubé MA (2016) Infrared process monitoring of conjugated linoleic acid/styrene/butyl acrylate bulk and emulsion terpolymerization. Journal of Applied Polymer Science 133 (26):n/a-n/a. doi:10.1002/app.43574 155 Chapter 5 28. Roberge S, Dubé MA (2016) Bulk Terpolymerization of Conjugated Linoleic Acid with Styrene and Butyl Acrylate. ACS Sustainable Chemistry & Engineering 4 (1):264- 272. doi:10.1021/acssuschemeng.5b01106 29. Jeong KS, Tang J, Liu H, Kim J, Schaefer AW, Kemp K, Levina L, Wang X, Hoogland S, Debnath R, Brzozowski L, Sargent EH, Asbury JB (2012) Enhanced Mobility-Lifetime Products in PbS Colloidal Quantum Dot Photovoltaics. ACS Nano 6 (1):89-99. doi:10.1021/nn2039164 30. Laible RC (1958) Allyl Polymerizations. Chemical Reviews 58 (5):807-843. doi:10.1021/cr50023a001 31. Khan MU, Reddy KR, Snguanwongchai T, Haque E, Gomes VG (2016) Polymer brush synthesis on surface modified carbon nanotubes via in situ emulsion polymerization. Colloid and Polymer Science 294 (10):1599-1610. doi:10.1007/s00396- 016-3922-7 32. Yuan Y, Zhang N, Tao W, Cao X, He Y (2014) Fatty acids as phase change materials: A review. Renewable and Sustainable Energy Reviews 29:482-498. doi:http://dx.doi.org/10.1016/j.rser.2013.08.107 33. Jimenez M, García HS, Beristain CI (2004) Spray-drying microencapsulation and oxidative stability of conjugated linoleic acid. European Food Research and Technology 219 (6):588-592. doi:10.1007/s00217-004-0992-4 34. Edirisinghe EMRKB, Bamunuarachchi A, Jayaweera V, Samaradiwakara R (1997) Studies on the effect of some plant component extracts on the preservation of fish oils. In: Summary Report of and Papers Presented at the Tenth Session of the Working Party on Fish Technology and Marketing: Colombo, Sri Lanka, 4-7 June 1996. vol 10. Food and Agriculture Organization of the United Nations, pp 199-205 35. Johnsen H, Schmid RB (2007) Preparation of polyurethane nanocapsules by miniemulsion polyaddition. Journal of Microencapsulation 24 (8):731-742. doi:10.1080/02652040701585179 36. Wilson JA, Chickos JS (2013) Vapor Pressures and Vaporization, Sublimation, and Fusion Enthalpies of Some Fatty Acids. Journal of Chemical & Engineering Data 58 (2):322-333. doi:10.1021/je300902c 156 Synthesis of green microcapsules containing fatty acids as PCM 37. Gooch JW (2011) Vapor Pressure. In: Gooch JW (ed) Encyclopedic Dictionary of Polymers. Springer New York, New York, NY, pp 789-789. doi:10.1007/978-1-4419- 6247-8_12451 38. Kaya İ, Çulhaoğlu S (2008) SYNTHESIS, CHARACTERIZATION, THERMAL DEGRADATION AND ELECTRICAL CONDUCTIVITY OF OLIGO[2-(2- HYDROXYPHENYLIMINOMETHYL-BENZYLIDENE)AMINOPHENOL] AND OLIGOMER-METAL COMPLEXES. Chinese Journal of Polymer Science 26 (02):131- 143. doi:10.1142/s0256767908002765 39. Yusof N, Rahman S, Hussein M, Ibrahim N (2013) Preparation and Characterization of Molecularly Imprinted Polymer as SPE Sorbent for Melamine Isolation. Polymers 5 (4):1215 40. Bartholin M, Boissier G, Dubois J (1981) Styrene-divinylbenzene copolymers, 3. Revisited IR analysis. Die Makromolekulare Chemie 182 (7):2075-2085. doi:10.1002/macp.1981.021820719 41. Dubé MA, Li L (2010) In-Line Monitoring of SBR Emulsion Polymerization Using ATR-FTIR Spectroscopy. Polymer-Plastics Technology and Engineering 49 (7):648-656. doi:10.1080/03602551003664909 42. Rozanna D, Chuah T, Salmiah A, Choong TS, Sa'ari M (2005) Fatty acids as phase change materials (PCMs) for thermal energy storage: a review. International Journal of Green Energy 1 (4):495-513 157 6. Synthesis of thermoregulating microcapsules with flame retardant properties YNTHESIS OF THERMOREGULATING S CHAPTER 6 MICROCAPSULES WITH FLAME RETARDANT PROPERTIES Based on the article . Szczotok, A. M.; Carmona, M.; Serrano, A.; Kjøniksen, A. L.; Rodriguez, J. F., Development of thermoregulating microcapsules with cyclotriphosphazene as a flame retardant agent. IOP Conference Series: Materials Science and Engineering 2017, 251 (1), 012120. . Szczotok, A. M.; Garrido, I.; Carmona, M.; Kjøniksen, A.-L.; Rodriguez, J. F., Predicting microcapsules morphology and encapsulation efficiency by combining the spreading coefficient theory and surface polarity. Submitted to the Colloids and Surfaces A: Physicochemical and Engineering Aspects. Chapter 6 Abstract Thermoregulating microcapsules containing phase change material (Rubitherm®RT27) were produced by using the suspension-like polymerization technique with styrene (St), divinylbenzene (DVB) and hexa(methacryloylethylenedioxy) cyclotriphosphazene (PNC-HEMA) as co- monomers. It was found that the thermal energy storage (TES) capacity of the microcapsules increased in the presence of PNC-HEMA. However, the morphology of the microcapsules became irregular when the content of monomer with flame retardant properties was increased. Thermogravimetric analyses performed under atmospheric air confirmed that the PNC-HEMA raised the amount of residue after the burning process, proving the formation of thermally stable char layer. Micro-combustion calorimetry analyses were performed to evaluate the fire resistance. Total heat release and peak heat release were lower than those exhibited by microcapsules without PNC-HEMA. Thus, these materials could be considered as an important alternative to commonly used microcapsules containing phase change materials (PCMs), where a lower flammability is required for their application. 160 Synthesis of thermoregulating microcapsules with flame retardant properties 6.1. Background The drawbacks such as low thermal conductivity and flammability properties have limited the application of thermoregulating microcapsules, especially in the building industry. Two modes of action to prevent the flame spreading are known, this is physical and chemical action. The combustion process can be retarded by physical action in several ways. Some flame retardant additives contribute to the formation of a protective layer between the gaseous phase where combustion occurs and the solid phase where thermal degradation takes place. This is an effect of a low thermal conductivity of that layer which can reduce the heat transfer from the heat source to the material. This process decreases the degradation rate of the material and reduce the access to fuel, preventing feeding the flame with flammable molecules. The decomposition reactions of the additive may cause temperature decrease by heat consumption. The flame retardant additive can degrade endothermically and involve cooling of the substrate to a temperature below its combustion temperature. Finally, the flame retardants can decompose forming inert gases such as H2O, CO2, NH3, etc. and thereby dilute the combustible gas mixture [1,2]. The most important flame retardant chemical reactions prevent the combustion process from taking place in the condensed and gas phases. In the condensed phase two types of chemical reactions occur. Firstly, the rupture of the polymer can be accelerated by the flame retardant, causing the polymer to drip and thus moving away from the flame action zone. Secondly, the flame retardant causes the formation of ceramic-like or a glass layer of carbon on the polymer surface. This char creates a physical insulation layer between the gas phase and the condensed phase [1,2]. On the other hand, the free-radical mechanism of the combustion process can be avoided by the incorporation of flame retardant additives. These additives release specific radicals (e.g. Cl and Br) in the gas phase which are able to react with highly reactive species (such as H and OH) to form less reactive or even inert molecules. If the exothermic process is stopped the system cools down, leading to a reduction in the fuel production [1,2]. Several methods have been reported to reduce the flammability of building materials containing PCMs. One of them involve addition of the PCMs to the building block and surrounding on all sides with non-combustible concrete without changing the 161 Chapter 6 strength properties of the block [3]. Kośny et al. [4] developed microencapsulated fire- resistive PCM using fatty acids esters, which are less flammable than paraffins, and by covering the surface of the microcapsules with a flame retardant agent. The PCM was applied into two insulations systems: PCM-enhanced cellulose insulation [5] and a blend of PCM with blown fiberglass [6]. These insulations are designed to be used in residential walls and attics. Banu et al. [7] tested ordinary gypsum wallboard impregnated with approximately 24% organic PCM and discussed the possibilities to reduce its flammability. Studies regarding improving the fire-retardancy of paraffin with a melting point between 54 and 56 °C by preparing a form-stable PCM based on paraffin and HDPE with and without EVA were performed by Cai et al. [8-14]. Different types of fire retardants such as expandable graphite (EG), ammonium polyphosphate (APP), zinc borate (ZB), pentaerythritol (PER), brominated fire- retardant, melamine phosphate (MPP), antimony oxide (AO), melamine cyanurate (MCA), organophilic montmorillonite (OMT), magnesium hydroxide (MH) and red phosphorus microencapsulated with melamine formaldehyde (MRP) were used. Additionally, the studies on synergistic effect between those fire retardants were carried out. From TGA analyses it was concluded that the addition of those compounds improved the thermal stability of PCM resulting in a residue amount between 5 and 20 wt.% at 800 °C in comparison with no char present for form-stable PCM without fire- retardant. Furthermore, little influence on thermal energy storage was found in the presence of fire retardant compounds. Accordingly, improved fire resistant properties and decreased flammability of microencapsulated paraffins are an important issue for many applications. In this chapter, a monomer containing phosphorous and nitrogen atoms was selected. Monomers with heteroatomic bonds in their chemical structure have good fire resistant properties due to the larger bonding energy compared to homoatomic bonds. Compounds of phosphazene having a P-N bond linked with other organic groups, inorganic or semiinorganic compounds, exhibit high heat resistant properties [15]. The chemical structure in Figure 6.1 represents the full substitution of hexachlorophosphazene (PNC) by hydroxyethylmethacrylate (HEMA), giving PNC- HEMA and pyridine hydrochloride as the final products of the synthesis. 162 Synthesis of thermoregulating microcapsules with flame retardant properties CH2 H3C O H2C O O H3C O O O HCl Cl Cl O N N P P N N + 6 H2C OH N N Cl P P Cl O + 6 N O P P O H C Cl Cl 3 CH3 N O O O CH2 Pyridine hydrochloride PNC HEMA O O H2C O CH3 O O CH3 O CH3 O CH2 H2C PNC-HEMA Figure 6.1. Synthesis reaction of hexa(methacryloyl ethylenedioxy)-cyclotriphosphazene. In this chapter, a monomer exhibiting flame retardant properties, hexa(methacryloyl ethylenedioxy)cyclotriphosphazene (PNC-HEMA), was synthesized to improve the thermal stability and flammability of thermoregulating microcapsules from styrene and divinylbenzene containing Rubitherm®RT27. Microcapsules containing different amounts of PNC-HEMA and PCM were developed, and their effect on TES capacity, morphology, thermal stability and fire resistance were analyzed. 6.2. Materials Hexa(methacryloyl ethylenedioxy)cyclotriphosphazene (PNC-HEMA) was synthesized in our laboratory. A short chemical characterization is presented below. Infrared analyses were performed for PNC, HEMA and PNC-HEMA in order to confirm the chemical structure of PNC-HEMA. The spectra are shown in Figure 6.2 and characteristic absorbance groups are listed in Table 6.1. 163 Chapter 6 HEMA PNC -COO PNC-HEMA P-O-C C=C P-N C=C O-H P-N=P P-Cl Absorbance 4000 3500 3000 2500 2000 1500 1000 500 -1 Wavenumber (cm ) Figure 6.2. Infrared spectra for HEMA, PNC and PNC-HEMA. The ring stretching vibration and ring vibration of the P – N group in the PNC can be found at the wavenumber 1205 and 875 cm-1, respectively. Whereas, in PNC-HEMA those peaks are shifted to 1242 and 949 cm-1. One of the most significant bond is P – Cl, which confirm that the reaction occurred. The P-Cl band can be recognized at 600 and 520 cm-1 in PNC whereas in the case of PNC-HEMA is practically negligible. Furthermore, the sorption due to the C=C vinyl group and -COO which correspond to HEMA was observed at similar wavenumbers. Thus, C=C appeared at 1645 and 815 cm-1 and -COO occurred at 1730 and 1170 cm-1. In HEMA, the hydroxyl group was found at a wavenumber of 3450 cm-1 and a small peak can be observed in case of PNC-HEMA since the conversion of HEMA to PNC-HEMA is not 100%. Nevertheless, new peaks occurred at 1070 and 970 cm-1 which correspond to the P-O-C group, indicating and confirming the reaction between PNC and HEMA. It can be concluded that, for the PNC-HEMA product obtained, the incorporation of HEMA into the PNC takes place by substitution of its chlorides [15]. 164 Synthesis of thermoregulating microcapsules with flame retardant properties Table 6.1. Main absorbance peaks for PNC, HEMA and PNC-HEMA. Functional group Wavenumbers (cm-1) PNC HEMA PNC-HEMA P – N 1205 - 1240 P – N = P 875 - 940 P – Cl 600 - - 520 - - C = C - 1645 1645 - 815 815 -COO - 1730 1730 - 1170 1170 O – H - 3450 - P – O – C - - 1070 - - 970 Thermogravimetric analyses in nitrogen atmosphere were performed to determine the thermal stability of the PNC, HEMA and PNC-HEMA monomer and they are shown in Figure 6.3. As can be seen, PNC and HEMA compounds present a single stage of thermal degradation at 170 and 150 °C, respectively. It is also observed that the percentage of residue in both cases is negligible, since the compounds are practically degraded. However, an important change can be observed for the PNC-HEMA, presenting two stages of degradation. The first one at 150 °C corresponds to toluene, unreacted PNC and HEMA present in the monomer whereas the second occurs from 260°C due to the degradation of the PNC-HEMA compound and it is elongated up to 550°C. It is important to point out that this monomer has a 35 wt.% of residue, indicating that it polymerizes at high temperature and the product could work as flame retardant, being stable at high burning temperatures. These results show a greater stability of the PNC-HEMA compared to its initial compounds. Together with the infrared spectra in Figure 6.2, the chemical reaction 165 Chapter 6 between the PNC and the HEMA was confirmed whereby the PNC-HEMA shows synergistic effect with respect to its raw materials. 100 PNC HEMA PNC-HEMA 80 60 Weight (%) Weight 40 20 0 0 100 200 300 400 500 600 700 Temperature (C) Figure 6.3. TGA analysis for the PNC, HEMA and PNC-HEMA. In this chapter, microcapsules with 50, 32 and 0% of PCM were synthesized keeping the parameters optimized in Chapter 4 constant. A St/DVB mass ratio of 1, toluene/PCM is 1.92, the continuous/discontinuous mass ratio is 3.33, the agitation rate 800 rpm, using PVP as suspending agent at a concentration of 5.03%. Products with and without PNC-HEMA were examined. The percentage associated with PNC-HEMA in the nomenclature of microcapsules is related to the amount of PNC-HEMA used based on the total amount of monomers. In addition, the amount of PCM based on total amount of monomers and PCM used is indicated in the sample name. 6.3. Results Morphology and particle size of the products The morphology of the synthesized products was partially described in Chapter 4. The morphology was characterized by SEM and it is shown also in Figure 6.4. Increasing the amount of PNC-HEMA resulted in poorer shape and higher 166 Synthesis of thermoregulating microcapsules with flame retardant properties agglomeration of the microcapsules. Irregular shape and agglomeration were mainly observed for products having a content of PNC-HEMA higher than 10 wt.% (Figure 6.4d and 6.4e). This is contrary to the expected results in which higher crosslinker contents improve the appearance of the polymer surface [16], in this case a broken product was obtained. According to Anzai et al. [15,17], the PNC-HEMA polymer presents small polymerization shrinkage, large hardness and a small coefficient of thermal expansion. These properties are attributed to the inorganic P3N3 ring and its hetero bonding effects [18]. There are two possible factors that can influence the final morphology of the obtained microcapsules. Firstly, unsubstituted hydroxyl groups that remain in the PNC-HEMA, and do not react with the six active sites of PNC. Secondly, the compressive strength of the copolymer rises in the presence of PNC-HEMA, leading to a higher hardness of the shell. It is possible that during the reaction, the particles having high hardness are broken due to the vigorous stirring required in the suspension- like polymerization process. It can be concluded, based on the SEM analysis, that the PNC-HEMA employed as a monomer should be maximum 10 wt.%, which provide spherical morphology and not a damaged structure. (a) (b) (c) (d) 167 Chapter 6 (e) Figure 6.4. SEM images for microcapsules containing different amount of PNC-HEMA and particles (a) MC-P(St-DVB)68%PCM; (b) MC-P(St-DVB-5%PNC-HEMA)68%PCM; (c) MC- P(St-DVB-10%PNC-HEMA)68%PCM; (d) MC-P(St-DVB-20%PNC-HEMA)68%PCM; (e) MC-P(St-DVB-30%PNC-HEMA)68%PCM. Additionally, synthesis with 0, 32 and 50% of PCM were performed and the SEM images are presented in Figure 6.5. The morphology varies according to the mass ratio used in the synthesis. For a low paraffin/monomer mass ratio, the microcapsules have a spherical morphology with smoother surfaces than that one with a higher mass ratio, in which a rougher surface is observed. Furthermore, particles synthesized by using 10 wt.% of PNC-HEMA without PCM depict smooth and spherical surfaces. Similar results were observed by Luo et al. [19] who reported improved morphology with increasing monomer content. Accordingly, the morphology of the microcapsules is affected by the PCM content and also by the presence of the monomer with flame retardant properties. (a) (b) 168 Synthesis of thermoregulating microcapsules with flame retardant properties (c) (d) (e) (f) Figure 6.5. SEM images for microcapsules containing different amount of PNC-HEMA and particles (a) MC-P(St-DVB)50%PCM; (b) MC-P(St-DVB-10%PNC-HEMA)50%PCM; (c) MC-P(St-DVB)32%PCM, (d) MC-P(St-DVB-10%PNC-HEMA)32%PCM; (e) P(St- DVB)0%PCM, (f) P(St-DVB-10%PNC-HEMA)0%PCM. Volume and number particle size distributions of the microcapsules with different amount of PNC-HEMA and MC-P(St-DVB)68%PCM are shown in Figure 6.6. The volume particle size distributions are bimodal, except for microcapsules without PNC-HEMA monomer or PCM. However, the number average distributions are unimodal for all types of microcapsules. The volume particle size distribution decreases by increasing content of flame retardant monomer and follow this order: MC-P(St-DVB-20%PNC-HEMA)68%PCM < MC-P(St-DVB-30%PNC-HEMA)68%PCM < MC-P(St-DVB-10%PNC-HEMA)68%PCM < MC-P(St-DVB-5%PNCHEMA)68%PCM < MC-P(St-DVB)68%PCM, having average values of 57, 64, 112, 145 and 213 μm, respectively. This trend can be related to the irregular morphology and damage of microcapsules in the presence of PNC-HEMA. 169 Chapter 6 10 MC-P(St-DVB-30%PNC-HEMA)68%PCM 5 0 MC-P(St-DVB-20%PNC-HEMA)68%PCM 10 5 0 MC-P(St-DVB-10%PNC-HEMA)68%PCM 10 5 Volume (%) Volume 0 15 MC-P(St-DVB-5%PNC-HEMA)68%PCM 10 5 0 15 MC-P(St-DVB)68%PCM 10 5 0 0 100 200 300 400 500 1000 1500 2000 Size (m) Figure 6.6. Volume average particle size distribution for microcapsules containing different amount of PNC-HEMA and particle. Regarding the number average particle size distribution shown in Figure 6.7, the values are very close and within 7 and 14 μm. The differences between the number and volume distributions are due to the cubic influence of the particle size and on the volume distribution whereas the number distribution is only dependent on the number of particles per particle size [20]. Accordingly, the volume average is biased towards large particles. The huge difference between number and volume averages indicates that most of the particles are quite small (number average), but there are a few very large particles present (which will contribute strongly to the volume average). The large sizes can be attributed to agglomeration. 170 Synthesis of thermoregulating microcapsules with flame retardant properties MC-P(St-DVB-30%PNC-HEMA)68%PCM 10 5 0 15 MC-P(St-DVB-20%PNC-HEMA)68%PCM 10 5 0 15 MC-P(St-DVB-10%PNC-HEMA)68%PCM 10 5 Number (%) Number 0 20 MC-P(St-DVB-5%PNC-HEMA)68%PCM 15 10 5 0 30 MC-P(St-DVB)68%PCM 20 10 0 0 10 20 30 40 50 60 Size (m) Figure 6.7. Number average particle size distribution for microcapsules containing different amount of PNC-HEMA and particles. Thermal energy storage properties of the microcapsules Figure 6.8 shows the PCM content, encapsulation efficiency and microcapsule yield. The latent heat of the microcapsules with flame retardant monomer increases significantly compared to the ones without (101.8 J/g), being 107.8, 109.0, 111.2 and 110.4 J/g for microcapsules with 5, 10, 20 and 30% of PNC-HEMA, respectively. These values correspond to PCM contents ranged between 62.97 and 64.95%. Accordingly, the PCM content of the thermoregulatory materials is very similar to the theoretical value (68.5%). This indicate that employment of PNC-HEMA as a monomer in the microencapsulation process favors the PCM encapsulation. When using only St and DVB as monomers, the PCM content reaches 59.46%. Considering that the interfacial 171 Chapter 6 tension influence the encapsulation process [21], the higher hydrophilicity of water/PNC-HEMA than for water/styrene or water/divinylbenzene, favors the formation of core/shell morphologies using PNC-HEMA as a monomer [22]. This was discussed in detail in Chapter 4.2.8. It can be noticed that the PCM content becomes higher with increasing amount of PNC-HEMA, while the encapsulation efficiency and microcapsule yield decrease. 1.30 CPCMi/CPCM0 EEi/EE0 1.25 ri/ ro 1.20 1.15 1.10 1.05 1.00 5 10 15 20 25 30 PNC-HEMA amount Figure 6.8. Effect of PNC-HEMA amount on microcapsules yield (ηr), paraffin content (CPCM) and encapsulation efficiency (EE). The heating/cooling DSC thermograms for microcapsules synthesized with 32, 50 and 68% of PCM are presented in Figure 6.9. Obviously, the latent heat obtained from these recipes is significantly lower than in the previous set of experiments due to low PCM load. The latent heat for microcapsules with PNC-HEMA were 27.25 and 69.4 J/g, respectively and for microcapsules from P(St-DVB), 28.73 and 58.01 J/g, respectively. It is noteworthy that the detected melting point is also shifted from 18 to 25.75 °C because of a thicker shell and lower thermal conductivity when a higher amount of monomers were used. 172 Synthesis of thermoregulating microcapsules with flame retardant properties 1.0 1.0 MC-P(St-DVB-10%PNC-HEMA)68%PCM MC-P(St-DVB)68%PCM 0.8 MC-P(St-DVB-10%PNC-HEMA)50%PCM 0.8 MC-P(St-DVB)50%PCM MC-P(St-DVB-10%PNC-HEMA)32%PCM MC-P(St-DVB)32%PCM 0.6 0.6 0.4 ) -1 0.4 0.2 0.2 0.0 0.0 -0.2 Heat flow (W/g) Heat Heat flow (W g flow (W Heat -0.4 -0.2 -0.6 -0.4 -0.8 -0.6 -40 -20 0 20 40 60 80 -40 -20 0 20 40 60 80 Temperature (C) Temperature (C) Figure 6.9. Heating/cooling cycles for microcapsules with 32, 50 and 68% of PCM with 0 and 10% of PNC-HEMA. Thermal degradation study of products Figure 6.10 shows TG and DTG curves of particles (P(St-DVB) and P(St-DVB- 10% PNC-HEMA)) and neat PNC-HEMA under air atmosphere. Neat PNC-HEMA exhibits a weight loss up to 200 °C, which is related to toluene evaporation and thereafter a wide monomer combustion with a final residue of 13.39%. For the particles, two weight losses can be distinguished. The first one corresponds to the polymer combustion and the second one is the degradation of char layer formed at high temperatures. The temperature at which 50% of sample’s weight has been lost rises from 398°C to 482 °C in the presence of 10 wt.% of PNC-HEMA. Furthermore, the amount of char formed by P(St-DVB-10% PNC-HEMA) was 10% higher than for the product without flame retardant properties. 173 Chapter 6 100 (a) PNC-HEMA 90 P(St-DVB) P(St-DVB-10% PNC-HEMA) 80 70 60 50 Weight (%)Weight 40 30 20 10 0 (b) PNC-HEMA P(St-DVB) 0.8 P(St-DVB-10% PNC-HEMA) C) 0.6 0.4 Deriv. Weight (%/ 0.2 0.0 0 100 200 300 400 500 600 700 Temperature (C) Figure 6.10. Degradation curves of neat PNC-HEMA and particles under air atmosphere a) TG, b) DTG. As can be seen in Figure 6.11, all microcapsules containing PCMs present three degradation stages. The first one corresponds to the release of encapsulated PCM, and the lowest maximum degradation temperatures (Tmax1) occurs for microcapsules without PNC-HEMA and the highest one is exhibited by product with 10% PNC- HEMA. The second (Tmax2) and third (Tmax3) weight losses can be assigned to combustion of copolymer and degradation of char, respectively. Hence, the char layer formed after combustion of PNC-HEMA protects the internal part of the material and 174 Synthesis of thermoregulating microcapsules with flame retardant properties delays the degradation. Accordingly, the thermal stability of the microcapsules at high temperatures is improved. Finally, the microcapsules without PNC-HEMA were completely degraded at 700 °C whereas a char yield of 1.35, 3.84, 3.26 and 3.76% has been obtained for microcapsules with 5, 10, 20 and 30% of PNC-HEMA, respectively. 100 (a) MC-P(St-DVB)68% PCM 90 MC-P(St-DVB-5% PNC-HEMA)68% PCM MC-P(St-DVB-10% PNC-HEMA)68% PCM 80 MC-P(St-DVB-20% PNC-HEMA)68% PCM MC-P(St-DVB-30% PNC-HEMA)68% PCM 70 60 50 Weight (%)Weight 40 30 20 10 0 1.8 (b) MC-P(St-DVB)68% PCM MC-P(St-DVB-5% PNC-HEMA)68% PCM 1.6 MC-P(St-DVB-10% PNC-HEMA)68% PCM MC-P(St-DVB-20% PNC-HEMA)68% PCM 1.4 MC-P(St-DVB-30% PNC-HEMA)68% PCM C) 1.2 1.0 0.8 Deriv. Weight (%/ 0.6 0.4 0.2 0.0 0 100 200 300 400 500 600 700 Temperature (C) Figure 6.11. Degradation curves of microcapsules under air atmosphere a) TG, b) DTG. Microcapsules with PNC-HEMA concentrations higher than 10% did not show significant additional change in the thermal properties. This can be explained by a small amount of unreacted PNC-HEMA, leading to a lower particle yield. The results are in agreement with those reported by Qu et al. [23], in which epoxy composites with poly(phosphazene-co-bisphenol A)-coated boron nitride were prepared and analyzed by 175 Chapter 6 TGA in nitrogen and air atmospheres. It was reported that the presence of phosphazene and boron nitride improved thermal stability and thermal conductivity of the composites, as well as decreased flammability, yielding a residue around 5%. Ash determination regarding ISO 3451-1 were performed in order to analyze ash obtained after calcination. Digital photos and SEM images of microcapsules without and with 10, 20 and 30% of PNC-HEMA are shown in Figure 6.13. For microcapsules without PNC-HEMA, no char occurred and the residual mass was very little after 1 h in the temperature of 600 °C. The morphology of the residue obtained for the other products maintained the original shape. Additionally, for MC-P(St-DVB-10%PNC- HEMA)68%PCM a spherical morphology of the char can be observed, showing a good dimensional stability during the whole experiment. The decomposition of cyclotriphosphazene present in PNC-HEMA not only increased the weight-loss temperature of the products in the higher region but also raised the char yield. For microcapsules synthesized with addition of PNC-HEMA, the residual mass increased with increasing amount of PNC-HEMA (Figure 6.12). 1.6 1.4 1.2 1.0 0.8 Residue (%) Residue 0.6 0.4 0.2 0.0 0 10 20 30 PNC-HEMA amount (%) Figure 6.12. Residue amount obtained in the calcination. EDAX analyses were performed in order to evaluate the elemental composition of the residue. Phosphorous constituted 16.91, 21.37, 28.11 and 31.12% of the char for microcapsules with 5, 10, 20 and 30% of PNC-HEMA, respectively. This indicates successful incorporation of this monomer in the microcapsules shell as well as improved resistant to high temperatures. 176 Synthesis of thermoregulating microcapsules with flame retardant properties (a) (b) (c) (d) Figure 6.13. Digital photographs and SEM images of the ashes after burning in mufle furnace at 600 °C (a) MC-P(St-DVB)68%PCM; (b) MC-P(St-DVB-10%PNC-HEMA)68%PCM; (c) MC- P(St-DVB-20%PNC-HEMA)68%PCM; (d) MC-P(St-DVB-30%PNC-HEMA)68%PCM. 177 Chapter 6 Regarding the thermal degradation results, MC-P(St-DVB-10%PNC- HEMA)68%PCM seems to be the most suitable product for the production of thermoregulating microcapsules with flame retardant properties. It can be concluded that the presence of PNC-HEMA promotes char formation and enhances the thermal stability and the flame retardant properties of the microcapsules. Flame retardant properties of the microcapsules Micro combustion calorimeter (MCC) is a useful technique to investigate effect of the polymer chemical structure on combustion behavior based on oxygen consumption theory. Figure 6.14 reveals the heat release rate (HRR) curves of different synthesized microcapsules. 600 MC-P(St-DVB) MC-P(St-DVB-5% PNC-HEMA) MC-P(St-DVB-10% PNC-HEMA) 500 MC-P(St-DVB-20% PNC-HEMA) I MC-P(St-DVB-30% PNC-HEMA) ) 400 -1 300 HRR (W g 200 II 100 0 0 100 200 300 400 500 600 Temperature (C) Figure 6.14. HRR curves of the microcapsules without and with different amount of PNC-HEMA and 68% of PCM. The HRR curves for microcapsules with paraffin encapsulated present two peaks. The first one corresponds to the heat released during the paraffin pyrolysis (I) whereas the second one is attributed the polymer pyrolysis (II). For MC-P(St-DVB)68%PCM the peak heat release rate (PHRR) values were 459.3 and 164.9 W g-1 for paraffin and polymer shell, respectively. The heat released in the pyrolysis of the polymer depends strongly on the polymer composition and is lower for the microcapsules with the highest PNC-HEMA content. This can be related with the higher thermal stability of the phosphorous compounds 178 Synthesis of thermoregulating microcapsules with flame retardant properties at this pyrolysis temperature, and due to the char formed by the presence of nitrogen and phosphorous elements in the PNC-HEMA which confirms its activity as a flame retardant. Table 6.3 summarizes the average values of 3 measurements for the parameters obtained from MCC tests. The table consist of two parts, the first one include results for microcapsules with increasing amount of PNC-HEMA and product obtained from P(St-DVB) copolymer. The second part combine the results for microcapsules with different amounts of PCM and 10% of PNC-HEMA and corresponding microcapsules without flame retardant monomer. The table lists peak heat release rate (PHRR), the temperature of peak heat release rate (TPHRR), total heat release rate (THR) and heat release capacity (HRC). Table 6.3. Parameters obtained from MCC test for the microcapsules. PHRR I TPHRR I PHRR II TPHRR II THR HRC Sample (W g-1) (°C) (W g-1) (°C) (kJ g-1) (J g-1K-1) Different amounts of PNC-HEMA MC-P(St-DVB)68%PCM 459.3±6.7 251.1±0.1 164.9±1.9 443.8±0.8 40.2±0.1 645.5±12.5 MC-P(St-DVB-5% PNC- 448.3±2.6 247.4±1.4 147.9±0.1 450.2±2.5 36.6±0.2 606.5±3.5 HEMA)68%PCM MC-P(St-DVB-10% PNC- 408.0±7.9 269.9±1.7 148.7±0.9 448.8±0.5 38.2±0.1 574.7±8.4 HEMA)68%PCM MC-P(St-DVB-20% PNC- 520.5±7.1 250.6±0.5 148.3±0.5 452.4±0.9 35.5±0.6 678.0±9.4 HEMA)68%PCM MC-P(St-DVB-30% PNC- 490.6±10.2 250.9±0.3 120.6±2.7 452.1±2.3 34.7±0.8 582.5±11.5 HEMA) 68%PCM Different amounts of PCM MC-P(St-DVB)68%PCM 459.3±6.7 251.1±0.1 164.9±1.9 443.8±0.8 40.2±0.1 645.5±12.5 MC-P(St-DVB-10% PNC- 408.0±7.9 269.9±1.7 148.7±0.9 448.8±0.5 38.2±0.1 574.7±8.4 HEMA)68%PCM MC-P(St-DVB)50%PCM 263.3±2.3 271.3±1.3 222.1±0.7 450.2±0.7 35.8±0.4 492.7±2.6 MC-P(St-DVB-10% PNC- 257.9±2.0 272.0±1.2 223.5±1.2 458.0±0.2 34.7±0.3 489.0±1.5 HEMA)50%PCM MC-P(St-DVB)32%PCM 181.7±4.3 258.7±2.0 275.5±2.7 450.5±0.8 33.5±0.3 469.7±4.1 MC-P(St-DVB-10% PNC- 175.7±3.7 266.2±1.8 260.6±3.0 457.6±0.6 33.0±0.2 448.5±6.8 HEMA)32%PCM MP-P(St-DVB) - - 466.2±3.4 447.0±0.4 33.1±1.1 476.3±6.2 MP-P(St-DVB-10%PNC- - - 434.1±2.2 459.9±0.7 28.9±0.8 443.0±7.4 HEMA) 179 Chapter 6 In order to examine the details of the effect of the presence of PNC-HEMA on the PHRR and THRR, these variables were plotted in Figures 6.15. 600 180 170 550 160 ) ) 500 -1 -1 150 450 140 130 PHRR I (W g PHRR 400 II (W PHRR g 120 350 110 300 100 0 5 10 15 20 25 30 0 5 10 15 20 25 30 280 480 470 260 460 240 450 220 440 200 430 TPHRR I (°C) TPHRR TPHRR II (°C) TPHRR 180 420 410 160 - - - - TPHRR for microcapsules without PNC-HEMA - - - - TPHRR for microcapsules without PNC-HEMA 400 5 10 15 20 25 30 5 10 15 20 25 30 PNC-HEMA amount (%) PNC-HEMA amount (%) Figure 6.15. Peak heat release rate and corresponding temperature for microcapsules with 68% of PCM and different amount of PNC-HEMA. It could be observed that the PHRR of PCM release decreases until 10% of PNC- HEMA was incorporated. However, in the samples containing 20 and 30% of flame retardant monomer, the peak increased with a maximum of 520.5 W g-1 for the microcapsules with 20 wt.% of PNC-HEMA. This behavior can be attributed to a thinner shell and the highest amount of encapsulated PCM (latent heat of 111.2 J g-1). Nevertheless, the PHRR of the polymer shell gradually decreased with increasing PNC-HEMA content. The phosphazene in the PNC- HEMA located at the microcapsules surface could promote the rate of polymer decomposed into char. This char layer could inhibit heat transfer to the internal part of the product and diminish evaporation of the combustible gases. This improve the flame retardant properties of the microcapsules [23]. The temperature of PHRR which correspond to paraffin release is basically stable, except for MC-P(St-DVB-10% PNC-HEMA)68%PCM which increased 180 Synthesis of thermoregulating microcapsules with flame retardant properties around 18 °C. A higher heat release temperature is expected in the presence of PNC- HEMA. Accordingly, the optimal concentration of PNC-HEMA is the MC-P(St-DVB- 30% PNC-HEMA) due to the lower heat release of the polymer shell, indicating a high tendency to form char. 500 550 0% PNC-HEMA 0% PNC-HEMA 450 500 10% PNC-HEMA 10% PNC-HEMA 400 450 400 350 ) ) -1 -1 350 300 300 250 250 200 200 PHRR I (W g PHRR PHRR II (W PHRR g 150 150 100 100 50 50 0 0 68% 50% 32% 68% 50% 32% 0% 300 500 0% PNC-HEMA 0% PNC-HEMA 490 290 10% PNC-HEMA 10% PNC-HEMA 480 280 470 C) C) 270 460 260 450 250 440 TPHRR I ( TPHRR TPHRR II ( TPHRR 430 240 420 230 410 220 400 68% 50% 32% 68% 50% 32% 0% PCM amount (%) PCM amount (%) Figure 6.16. Peak heat release rate and corresponding temperature for microcapsules with 10% of PNC-HEMA and different amount of PCM. Figure 6.16 illustrates the comparison between microcapsules with 0 or 10% of PNC-HEMA and various amounts of PCM. The PHRR and TPHRR decreased and increased, respectively, confirming the improved thermal properties of the products. Another important parameter to evaluate the fire safety included in Table 6.3 is THR. The THR is the amount of heat released throughout the decomposition and can be indicative of the total amount of fuel generated [24]. The THR of microcapsules without PNC-HEMA was always higher than these of microcapsules containing the flame retardant monomer. The heat release capacity (HRC) correspond to the maximum heat release rate divided by the heating rate of the MCC test. This value can be used as a reliable indicator of a polymer’s flammability [24]. The presence of PNC-HEMA as co-monomer in microcapsules lead to a decrease of HRC values. Moreover, the temperature of heat release peaks were sifted to higher 181 Chapter 6 values for microcapsules with PNC-HEMA than those without the monomer. The above parameters of all microcapsules with PNC-HEMA were lower than the corresponding values of microcapsules without PNC-HEMA, proving that the addition of PNC-HEMA could improve the flame retardancy of the microcapsules. The thermal decomposition is an endothermic process in which the obtained products such as phosphate, metaphosphate formed a nonvolatile protective film on the surface of the polymer at the same time isolating it from the air. In addition, the non-flammable gases released (e.g. CO2, NH3 and N2) cut off the supply of oxygen resulting in synergistic flame retardancy. 6.4. Conclusions Thermoregulating microcapsules from PNC-HEMA, styrene and divinylbenzene as co-monomers were manufactured. It was observed that a high content of PNC-HEMA is deleterious for the morphology of the microcapsules. Nevertheless, all the thermal properties of the microcapsules such as TES capacity, thermal stability and fire resistance were improved with the content of PNC-HEMA. According to the physical and thermal results the use of 10 wt.% of PNC-HEMA respect to the other monomers was the optimal concentration for obtaining microcapsules with the best morphology, the highest TES capacity and thermal stability. MC-P(St-DVB-10% PNC- HEMA)68%PCM presented an spherical shape with relatively small particle size (112 μm), a latent heat of 109.0 J/g and a charred residue of 3.84 wt.%. Additionally, it was confirmed that the phosphazene compound was able to create a dense and stable char at high temperature improving the flame retardant properties of microcapsules. Therefore, it is possible to improve the fire-resistive of the materials by incorporating flame retardant monomers as shell constituents. 182 Synthesis of thermoregulating microcapsules with flame retardant properties References 1. Laoutid F, Bonnaud L, Alexandre M, Lopez-Cuesta JM, Dubois P (2009) New prospects in flame retardant polymer materials: From fundamentals to nanocomposites. Materials Science and Engineering: R: Reports 63 (3):100-125. doi:https://doi.org/10.1016/j.mser.2008.09.002 2. Bourbigot S, Duquesne S (2007) Fire retardant polymers: recent developments and opportunities. Journal of Materials Chemistry 17 (22):2283-2300. doi:10.1039/B702511D 3. Salyer IO (1998) Building products incorporating phase change materials and method of making same. 4. Kośny J, Yarbrough DW, Riazzi T, Leuthold D, Smith JB, Bianchi M (2009) Development and testing of ignition resistant microencapsulated phase change material. Proceedings of the Effstock 5. Kosny J, Yarbrough DW, Miller WA, Petrie T, Childs PW, Syed AM (2008) 2006/07 Field Testing of Cellulose Fiber Insulation Enhanced with Phase Change Material. Oak Ridge National Laboratory (ORNL); Building Technologies Research and Integration Center, 6. Miller WA, Kosny J Next generation roofs and attics for homes. In: ACEEE summer conference, Pacific Grove, CA, 2008. 7. Banu D, Feldman D, Haghighat F, Paris J, Hawes D (1998) Energy-storing wallboard: flammability tests. Journal of Materials in civil Engineering 10 (2):98-105 8. Cai Y, Hu Y, Song L, Tang Y, Yang R, Zhang Y, Chen Z, Fan W (2006) Flammability and thermal properties of high density polyethylene/paraffin hybrid as a form-stable phase change material. Journal of Applied Polymer Science 99 (4):1320-1327. doi:10.1002/app.22065 9. Cai YB, Wei QF, Shao DF, Hu Y, Song L, Gao WD (2009) Magnesium hydroxide and microencapsulated red phosphorus synergistic flame retardant form stable phase change materials based on HDPE/EVA/OMT nanocomposites/paraffin compounds. Journal of the Energy Institute 82 (1):28-36. doi:10.1179/014426008X370988 10. Cai Y, Hu Y, Song L, Lu H, Chen Z, Fan W (2006) Preparation and characterizations of HDPE–EVA alloy/OMT nanocomposites/paraffin compounds as a shape stabilized 183 Chapter 6 phase change thermal energy storage material. Thermochimica Acta 451 (1):44-51. doi:https://doi.org/10.1016/j.tca.2006.08.015 11. Cai Y, Hu Y, Song L, Kong Q, Yang R, Zhang Y, Chen Z, Fan W (2007) Preparation and flammability of high density polyethylene/paraffin/organophilic montmorillonite hybrids as a form stable phase change material. Energy Conversion and Management 48 (2):462-469. doi:https://doi.org/10.1016/j.enconman.2006.06.021 12. Cai Y, Wei Q, Huang F, Gao W (2008) Preparation and properties studies of halogen- free flame retardant form-stable phase change materials based on paraffin/high density polyethylene composites. Applied Energy 85 (8):765-775. doi:https://doi.org/10.1016/j.apenergy.2007.10.017 13. Cai Y, Song L, He Q, Yang D, Hu Y (2008) Preparation, thermal and flammability properties of a novel form-stable phase change materials based on high density polyethylene/poly(ethylene-co-vinyl acetate)/organophilic montmorillonite nanocomposites/paraffin compounds. Energy Conversion and Management 49 (8):2055- 2062. doi:https://doi.org/10.1016/j.enconman.2008.02.013 14. Cai Y, Wei Q, Huang F, Lin S, Chen F, Gao W (2009) Thermal stability, latent heat and flame retardant properties of the thermal energy storage phase change materials based on paraffin/high density polyethylene composites. Renewable Energy 34 (10):2117-2123. doi:https://doi.org/10.1016/j.renene.2009.01.017 15. Anzai M, Ohashi M (1984) Studies on the reaction product of hexachlorocyclotriphosphazene and 2-hydroxyethyl methacrylate and on the physical properties of its polymer. The Journal of Nihon University School of Dentistry 26 (2):109-118 16. Kangwansupamonkon W, Damronglerd S, Kiatkamjornwong S (2002) Effects of the crosslinking agent and diluents on bead properties of styrene–divinylbenzene copolymers. Journal of Applied Polymer Science 85 (3):654-669. doi:10.1002/app.10620 17. Anzai M, Ohashi M (1984) Studies on Composite Resins with Hexa (methacryloylethylene-dioxy) Cyclotriphosphazene Used as a Monomer Fillers and Physical Properties. The Journal of Nihon University School of Dentistry 26 (3):238-242. doi:10.2334/josnusd1959.26.238 18. Antonucci JM (1986) Resin Based Dental Composites — An Overview. In: Chiellini E, Giusti P, Migliaresi C, Nicolais L (eds) Polymers in Medicine II: Biomedical and 184 Synthesis of thermoregulating microcapsules with flame retardant properties Pharmaceutical Applications. Springer US, Boston, MA, pp 277-303. doi:10.1007/978- 1-4613-1809-5_22 19. Luo Y, Zhou X (2004) Nanoencapsulation of a hydrophobic compound by a miniemulsion polymerization process. Journal of Polymer Science Part A: Polymer Chemistry 42 (9):2145-2154 20. Alcázar Á, Carmona M, Borreguero AM, de Lucas A, Rodríguez JF (2015) Synthesis of microcapsules containing different extractant agents. Journal of Microencapsulation 32 (7):642-649 21. Torza S, Mason SG (1970) Three-phase interactions in shear and electrical fields. Journal of Colloid and Interface Science 33 (1):67-83. doi:http://dx.doi.org/10.1016/0021-9797(70)90073-1 22. Sánchez-Silva L, Rodríguez JF, Romero A, Borreguero AM, Carmona M, Sánchez P (2010) Microencapsulation of PCMs with a styrene-methyl methacrylate copolymer shell by suspension-like polymerisation. Chemical Engineering Journal 157 (1):216-222. doi:10.1016/j.cej.2009.12.013 23. Qu T, Yang N, Hou J, Li G, Yao Y, Zhang Q, He L, Wu D, Qu X (2017) Flame retarding epoxy composites with poly(phosphazene-co-bisphenol A)-coated boron nitride to improve thermal conductivity and thermal stability. RSC Advances 7 (10):6140-6151. doi:10.1039/C6RA27062J 24. Tretsiakova-McNally S, Joseph P (2015) Pyrolysis Combustion Flow Calorimetry Studies on Some Reactively Modified Polymers. Polymers 7 (3):453 185 7. Thermal and morphological stability of microcapsules THERMAL AND MORPHOLOGICAL CHAPTER 7 STABILITY OF MICROCAPSULES Based on the article: . . Szczotok, A. M.; Kjøniksen, A.-L.; Rodriguez, J. F.; M.; Carmona , Estimation of the vapor pressure and enthalpy of vaporization of phase change materials by thermogravimetric analyses, Journal of… (Draft prepared). Chapter 7 Abstract Thermal and morphological stability of microcapsules were analyzed. It was found that the microcapsules kept an intact structure but the PCM content decreased to 0 J g-1 at 180 °C. This means that the PCM is released from the microcapsule by diffusion through the shell and evaporation. Three different approaches to estimate the vapor pressure by thermogravimetric analyses of commercial phase change materials (Rubitherm®RT31, Rubitherm®RT27, Rubitherm®RT21 and Rubitherm®RT6) have been compared. Langmuir vaporization relation and two different models in stepwise and non-isothermal thermogravimetric modes based on mass transfer and estimating the binary diffusion coefficients in the gas phase as temperature function by Fuller equation were evaluated. The Antoine equation was used to estimate the variation of vapor pressure with temperature and the three unknown variables were obtained by a non-linear regression fitting utilizing the Marquardt algorithm. A good agreement between the experimental data and theoretical models has been found. The non-isothermal method exhibited the highest accuracy for obtaining the vapor pressure. n-Hexadecane and n-decane have been also used as reference compounds to estimate the deviation of the experimental boiling temperature of PCMs with respect to those predicted from this model. If was observed that the boiling point and heat of vaporization can be predicted with good accurately by this model. Accordingly, thermogravimetric analyses using the fastest ramp mode can be utilized to obtain the vapor pressure of pure compounds and the boiling point and the heat of vaporization can be also predicted. Finally, the diffusion of PCM from microcapsules was modeled considering the microcapsule stability and the diffusion of the liquid PCM into the solid matrix, vaporization from external surface of microcapsule and the mass transfer in the gas phase. Diffusion coefficients for microcapsules with Rubitherm®RT27, Rubitherm®RT21 and Rubitherm®RT6 were found. The diffusion coefficients were significantly lower than for pure PCM, confirming the expected influence of microencapsulation. 188 Thermal and morphological stability of microcapsules 7.1. Background For an application of PCM microcapsules in building materials, they are required to have good thermal stability and mechanical strength. These properties must be ensured in order to maintain an intact structure during the manufacturing processes of building materials and after that during its entire life cycle. In the production processes of rigid polyurethane foam and concrete, exothermic reactions are observed, promoting significant changes in the temperature of the raw materials. In production process of rigid polyurethane foams, the maximum temperature can increase up to 155 °C within first 200 s of reaction [1]. During concrete preparation, the factors that influence the temperature rise are the initial temperature of the materials, ambient temperature, cement content, section size and use of admixtures [2]. There are many complex computer programs available to predict the temperature development in concrete. One of the approaches define a typical value of temperature rise of 11.15 ºC per 100 kg cement/m3. Stanley [2] reported that a peak temperature of 65 °C was reached 1.6 days after casting, whereas most of the heat gain took place over 0.6 of a day. Nine days after casting, the concrete core temperature was still 20 ℃ above the ambient temperature. It is therefore important to study the thermal and morphological resistance of microencapsulated PCM. The thermal behavior of a PCM microcapsules are determined by the two phase changes that are produced with increasing temperature. The melting phase change can produce a leakage or diffusion of PCM from the microcapsules that is avoided by the shell. If the temperature continues to increase, the PCM can evaporate and the shell cannot retain this vapor. Accordingly, the determination of the vapor pressure of the PCMs will be the key to determine the release rate and the degradation thermal profile of the microencapsulated PCMs. The determination of the vapor pressure of a liquid is necessary for a wide variety of materials and their applications. It is especially important to know the rate of evaporation of a compound at a specific temperature when the loss of additives in polymers leads to an undesired reduction of their efficiency and changes in the material properties which is important in terms of storage and long term stability [3]. Several methods to determine the vapor pressure are reported such as ebulliometry, gas saturation and manometry. However, these methods require rigorous vacuum, very accurate pressure measurements and are time consuming [4]. 189 Chapter 7 Thermogravimetric analysis is considered a simple, easily available and rapid technique which requires only a small amount of sample with sufficient accuracy for engineering purposes [5-7]. Currently, two different methods are considered for determination of vapor pressures from thermogravimetric analysis. The first procedure is based on the Langmuir equation of evaporation [8]. In fact, this approach is broadly applied due to the simple procedure where only the molecular weight is necessary for the vapor pressure calculations. Nowadays, this method is utilized for determination of vapor pressure of UV absorbers [3], explosives [4,9] and pharmaceuticals [10-12]. The second method merge Fick’s law of diffusion with mass transfer for the evaporation in a stagnant gas [13]. However there is lack of information about long term stability for commercially used phase change materials. Behzadi and Farid [14] performed studies regarding the thermal stability and reported vapor pressure considering the future application of propyl ester and n-alkanes. To the best of our knowledge, no systematic studies have been carried out for the determination of vapor pressure, Antoine constants and related thermodynamic properties of commercial phase change materials. In this chapter, the use of thermogravimetric analysis for the estimation of vapor pressure, enthalpy of vaporization, and normal boiling point of different PCMs such as Rubitherm®RT31, Rubitherm®RT27, Rubitherm®RT21, and Rubitherm®RT6 are considered. Stepwise and non-isothermal studies were performed. The Langmuir equation of vaporization and mass transfer combined with Fick’s law of diffusion were used for the estimation of thermodynamic properties of the PCMs. The morphological stability is tested by SEM, increasing the temperature of the environment from room temperature to 240 °C. Finally, a model which considers the microcapsule stability and the diffusion of the liquid PCM into the solid matrix, vaporization on the microcapsule external surface and the mass transfer in the gas phase was considered and the diffusion coefficients of the PCMs into the solid matrix were determined. 7.2. Materials For heat-treating, MC-P(St-DVB)68%PCM and MC-P(St-DVB-10% PNC- HEMA)68%PCM were used. For vapor pressure model, Rubitherm®RT31, 190 Thermal and morphological stability of microcapsules Rubitherm®RT27, Rubitherm®RT21, and Rubitherm®RT6 as neat PCMs were used. After that, the microcapsules containing respective PCM as core material were evaluated, here denoted shortly as MC(RT27), MC(RT21), MC(RT6), respectively. MC(RT27) is the same as MC-P(St-DVB)68%PCM, but in order to differ between other PCMs, short name of PCM was used in parenthesis. 7.3. Results Thermal stability Thermal decomposition curves of microcapsules stored at room temperature and after heating to 200°C are given in Figure 7.1. Microcapsules weight loss profile occurred in two steps before heat treating which correspond to evaporation of PCM and decomposition of polymer. The onset and ending evaporation temperature of PCM are 144.1 and 232.8 °C, respectively. On the other hand, the decomposition of P(St-DVB) occurred from 363.7 to 462.2 °C. After treating the microcapsules at 200 °C, only one decomposition step was observed, which corresponds to the polymeric shell. This indicates that the PCM was released from the microcapsules. In order to analyze this mechanism, the vapor pressure of pure PCMs was estimated, the morphological stability at high temperature was analyzed and finally, a model of diffusion of PCM thorough the microcapsules was developed. Figure 7.1. Thermal decomposition of microcapsules before and after heat treating. 191 Chapter 7 Study on neat PCMs a) Mathematical model for estimation of vapor pressure for neat PCM Stepwise study Price et al. [3,15] reported the application of thermogravimetric analysis for determination of the vapor pressure of low volatility substances based on Langmuir equation 7.1 by using experiments under vacuum conditions, knowing the weight loss at a constant temperature (stepwise mode). 푑푚 푀 퐴 = 푃0푆훼√ 퐴 (7.1) 푑푡 퐴 2휋푅푇 −1 0 where dm/dt corresponds to the rate of mass loss (g·min ), 푃퐴 is the vapor pressure (Pa) of species A at absolute temperature T (K), S is the vaporization surface area (m2), α is −1 the evaporation coefficient, MA is the molecular weight of A (g·mol ) of the evaporating substance, R is the gas constant (m3·Pa·K−1·mol−1). However some uncertainty appears since the Langmuir theory was designed for calculation of saturated vapor pressure in vacuum and the evaporation coefficient is equal to one [8]. When substances evaporate in a flowing gas stream at finite pressure, the evaporation coefficient cannot be assumed as unity [3]. Thus, it is necessary to determine it through experiments by utilizing substances with well-known vapor pressure. The values can then be utilized for unknown samples, since it has been shown that it is independent on the substance and temperature but dependent on the apparatus [3,16,17]. Consequently, Equation 7.2 can be written as follows: 0 푃퐴 = 푘푣 (7.2) where k is a calibration constant and v is material dependent √2휋푅푔 푘 = (7.3) 훼푆 192 Thermal and morphological stability of microcapsules 푑푚 푇 푣 = 퐴 √ (7.4) 푑푡 푀퐴 The model based on the Langmuir equation is hereafter going to be denoted as the Langmuir method. In this model the diffusion of the pure component into the stagnant gas is not included. Nevertheless, Pieterse and Focke [13] stated that diffusion cannot be neglected when the measurements are performed at finite pressure. For that reason, a model based on Fick’s law considering that diffusion controls the mass transfer in the gas phase must be analyzed. Then, the molar flux of the species A in a binary system (A, B) is expressed by: 푑푦 푁 = −푐퐷푔 퐴 + 푦 (푁 + 푁 ) (7.5) 퐴,푧 퐴퐵 푑푧 퐴 퐴,푧 퐵,푧 푑푦 where 푐퐷푔 퐴 is the diffusional flux and 푦 (푁 + 푁 ) is the bulk flow term for the 퐴퐵 푑푧 퐴 퐴,푧 퐵,푧 푔 species A, c is the overall vapor contribution, 퐷퐴퐵 is the diffusion coefficient of A in B, yA is the molar fraction of A in the vapor phase, NA,z and NB,z are the net fluxes of A and B, respectively and z is the vapor-liquid interphase height. If B is a stagnant gas, the net flux B (NB,z) is zero through the diffusion path and Equation 7.5 yields 푔 (7.6) 푐퐷퐴퐵 푑푦퐴 푁퐴,푧 = − (1 − 푦퐴) 푑푧 where NA,Z is the vaporization rate of A at each temperature, flowing through the container surface (S) and being constant at a stepwise mode. 푑푚 푁 = /(푀 푆) (7.7) 퐴,푧 푑푡 퐴 Then, yA can be integrated respect to z in Equation 7.6, obtaining the corresponding net flux of A at a constant temperature (Equation 7.8) 193 Chapter 7 푐퐷푔 1 − 푦 (7.8) 푁 = 퐴퐵 푙푛 퐴 퐴,푧 (푧 − 푧 ) 1 − 푦 0 퐴,푧0 where z0 is zero and it is defined at the top of the pan at which the vapor concentration of A is practically negligible. When the log-mean average concentration of gas B is used and yA+yB=1, we have 푦퐵2 − 푦퐵1 푦퐵,푙푚 = 푦 ln ( 퐵2) (7.9) 푦퐵1 1 − 푦퐴 푦퐴,푧0 − 푦퐴 ln = (7.10) 1 − 푦퐴,푧0 푦퐵,푙푚 Finally, the molar flux is 푔 푐퐷퐴퐵 푦퐴,푧0 − 푦퐴 푁퐴,푧 = (7.11) 푧 푦퐵,푙푚 Replacing the Equation 7.7 in 7.11 and given to the full gas the consideration as ideal gas, c can be replaced by PT/RT and return: 푑푚 푀 푆푃 퐷푔 (푦 − 푦 ) (7.12) = 퐴 푇 퐴퐵 퐴,푧0 퐴 푑푡 푅푇푧 푦퐵,푙푚 where PT is the total pressure. Thus, if the gas phase is considered as ideal, the vapor phase follows the Dalton’s law, where the total pressure is the sum of the partial pressures of the individual gases (P=PA+PB).Taking into account that the liquid is constituted only for pure A, Raoult’s law can be used for the liquid phase in which the partial pressure of A is equal to the 0 vapor pressure (PA≅ 푃퐴 ). 푑푚 푀 푆푃 퐷푔 푃0 = 퐴 푇 퐴퐵 (− 퐴 ) (7.13) 푑푡 푅푇(푧) 푦퐵,푙푚푃푇 194 Thermal and morphological stability of microcapsules 0 Vapor pressure 푃퐴 can also be determined using the Antoine equation as function of temperature in the form: 퐵 푙푛푃0 = 퐴 − (7.14) 퐴 퐶 + 푇 where A, B, and C are the Antoine constants. These constants are only tabulated for a limited numbers of compounds [18]. These constants are therefore going to be determined from experimental values of dm/dt by using the Marquardt algorithm. In this way, the weighted sum of squared residuals is defined as: 푇 푑푚푒푥푝 푑푚푡ℎ 푑푚푒푥푝 푑푚푡ℎ Φ = ( − ) ( − ) (7.15) 푑푡 (푖) 푑푡 (푖) 푑푡 (푖) 푑푡 (푖) where is the weighted sum of squared residuals for the weight loss of PCM during the 푑푚푒푥푝 푑푚푡ℎ TGA analysis, and are the experimental and theoretical weight losses of the 푑푡 푑푡 PCM at each specific temperature, respectively and i is a counter which indicates the weight loss at a measured temperature. This model based on Fick’s law and developed for a diffusional flux at constant temperature is called as Stepwise method. Non-isothermal study Mass transfer with Fick’s law model can be also used to predict the theoretical weight loss obtained in non-isothermal study. In this case, a pseudo-steady state assumption may be considered and thus, the Equation 7.11 can be applied. In the thermogravimetric analysis, the temperature at each time can be obtained by using the heating rate (dT/dt): 푑푇 푇 = 푇 + 푡 (7.16) 0 푑푡 where T0 is the initial temperature at which the non-isothermal analysis has started. 195 Chapter 7 Mass change as function of time can be also expressed in terms of the change in the pan level z with the time (Equation 7.17). 1 푑푚 휌 푆 푑푧 = 푁퐴,푧 = (7.17) 푀퐴 푑푡 푀퐴 푑푡 Combining the Equation 7.11 and 7.16, we obtain 푑푧 푀 푐퐷푔 (푦 − 푦 ) = 퐴 퐴퐵 퐴1 퐴2 (7.18) 푑푡 휌 푆 푧 푦퐵,푙푚 The numerical integration Fourth Order Runge-Kutta method must be applied to solve this differential equation, obtaining z as function of time. The molar fractions of A in the vapor phase can be obtained by means of its vapor pressure, defined by the Antoine equation and in terms of the three unknown parameters A, B and C. Once z is quantified, 푡ℎ the theoretical weight loss 푤푙표푠푠 at each time can be estimated. 푡ℎ 휌푆(ℎ0 − 푧) 푤푙표푠푠 = 1 − (7.19) 푚0 where h0 and m0 are the initial values of the liquid level and mass weight, respectively. The values of the three unknown parameters were also determined by a multiple non- linear regression procedure by means of the Marquardt algorithm. The non-isothermal model based on Fick’s law of diffusion combined with mass transfer equation is hereafter denoted as the Ramp method. b) Results of vapor pressure, boiling point and heat of vaporization of PCMs Four phase change materials supplied by Rubitherm with different melting points and n-hexadecane as reference were thermogravimetrically analyzed in order to determine their vapor pressure by mean of three methods (Langmuir, Stepwise and Ramp). Physical properties of investigated substances are gathered in Table 7.1. 196 Thermal and morphological stability of microcapsules Table 7.1. Melting temperature, molecular weight and density in liquid state of analyzed substances. Melting point Molecular weight Density liquid Substance (°C) (g mol-1) (g cm-3) a Rubitherm®RT31 (RT31) 32 268b 0.76 Rubitherm®RT27 (RT27) 26 258c 0.76 Rubitherm®RT21 (RT21) 21 244c 0.77 Rubitherm®RT6 (RT6) 7 222 0.77 n-Decane (C10) -30 142a 0.73 n-Hexadecane (C16) 18 226a 0.77 a Data from supplier, b ref. [19], c ref. [20]. In order to apply Fick’s law, binary diffusion coefficients (DAB) in the purge gas for all the studied PCMs must be determined. They can be experimentally determined or predicted by using theoretical models proposed by Sutherland, Jeans, Chapman and Cowling or Fuller, Schettler and Giddings [21]. In this study, the method developed by Fuller, Schettler, and Giddings was used due to its high accuracy to predict this coefficient with deviations about 4% (Equation 7.20) [21-23]. 1 1 10−3푇1.75( + )1/2 푀퐴 푀퐵 (7.20) 퐷퐴퐵 = 2 ∑ 1/3 ∑ 1/3 푃푇[( 푣)퐴 + ( 푣)퐵 ] ∑ ∑ where ( 푣)퐴 and ( 푣)퐵 are sum of atomic diffusion volumes of A and B. The diffusion volumes for A was calculated from the elemental composition and multiplied by recommended atomic diffusion volumes which for carbon atom is 16.5 and hydrogen atom is 1.98. The diffusion volume of the nitrogen molecule is 17.9 [21]. The elemental composition of Rubitherm products was determined by elemental analysis which allowed to obtain average fractions of carbon and hydrogen atoms of each studied paraffin. The elements present in their structure are shown in Table 7.2. 197 Chapter 7 Table 7.2. Chemical composition of studied phase change materials. Chemical composition Substance Catoms Hatoms RT31 18.73 40.53 RT27 18.43 39.67 RT21 17.70 37.46 RT6 15.61 33.91 C16 15.85 33.64 In Figure 7.2 the two used methodologies to estimate the weight loss of the PCM with temperature are shown. 300 Ramp method (ramp 10 K min-1) Stepwise method 250 200 150 Time (min) Time 100 50 0 300 350 400 450 500 550 Temperature (K) Figure 7.2. Stepwise and non-isothermal methodologies to estimate the weight loss of the PCM with temperature. As can be seen, in the stepwise a finite number of experimental data in which the temperature is hold constant are obtained whereas, in the ramp an infinity or large number of experimental data can be used in the quantification. The evaporation rate at the temperature range 323.15 to 433.15 K was measured for n-hexadecane and Rubitherm compounds under the stepwise conditions and from room temperature to 550 K under non-isothermal conditions. The mass loss rates for stepwise and weight losses versus temperature for non-isothermal analyses are shown in Figure 7.3. 198 Thermal and morphological stability of microcapsules 100 0.00 (b) (a) -0.05 80 ) -1 -0.10 60 -0.15 40 RT31 (%) loss Weight RT6 dm/dt (mg min (mg dm/dt -0.20 RT27 RT21 RT21 RT27 20 -0.25 RT6 RT31 hexadecane hexadecane -0.30 0 320 340 360 380 400 420 440 300 350 400 450 500 550 Temperature (K) Temperature (K) Figure 7.3. (a) Experimental weight loss rate at a stepwise condition and (b) the weight loss at non-isothermal condition with heating rate 10 K min-1. As can be seen, from these two methodologies the paraffin volatility order RT6 > n-hexadecane > RT21 > RT27 > RT31 can be obtained and it agrees perfectly with the melting temperature and molecular weight of the PCM indicated in Table 7.1. As expected, the lower the number of carbon in the paraffin, the lower its melting point and the lighter the chemical compound. It is also important to point out that by using the non- isothermal methodology it is possible to study the vaporization of these compounds at temperatures closer to the respective boiling point. In the case of the stepwise method, the weight loss rate is too high at high temperatures, making it impossible to reach a high enough number of data in only one analysis. Using Langmuir model For the Langmuir method, the n-hexadecane was selected as a reference compound because its vapor pressure has been widely studied in the literature [5,24-27]. The calibration constant found for this compound could be used to determine the vapor pressure of the other investigated substances. The Antoine constants of n-hexadecane 0 were taken from the literature [27]. For that purpose, the correlation of vapor pressure 푃퐴 vs. mass loss rate v for n-hexadecane was plotted (Figure 7.4) exhibiting a linear trend with intercept set as 0. Thus, k could be estimated from Equation 7.2. A k value of - 2.23x106 Pa1/2m−1/2K−1/2mol−1/2 was found, which is very similar to those reported by Shahbaz et al. [17] for glycerol and ethylene glycol. 199 Chapter 7 1000 900 800 700 600 500 p (Pa) 400 300 200 100 0 -4.0x10-4 -3.0x10-4 -2.0x10-4 -1.0x10-4 0.0 0.5 -1 0.5 0.5 v (g min mol K ) Figure 7.4. Calibration curve using n-hexadecane. For the investigated PCMs, the three Antoine constants were obtained by fitting the vapor pressure values with temperature to Antoine equation, using the solver excel tool in a non-linear regression homemade procedure. The Antoine coefficient are reported in Table 7.3. Table 7.3. Antoine correlations for analyzed PCMs obtained from Langmuir model. Antoine coefficient PCM A B C RT6 12.96 1310.05 -191.08 RT21 10.93 915.52 -236.12 RT27 17.43 3089.19 -157.47 RT31 11.21 953.32 -248.02 In Figure 7.5, the experimental and fitted mass loss rates are shown for investigated PCMs. These plots (scatter and line type) show good correlation with each other. 200 Thermal and morphological stability of microcapsules 0.0E+00 -1.0E-06 -2.0E-06 -3.0E-06 dm/dt (g/s) dm/dt Experimental Theoretical -4.0E-06 RT31 RT31 RT27 RT27 RT21 RT21 -5.0E-06 RT6 RT6 320 340 360 380 400 420 440 Temperature (K) Langmuir Figure 7.5. Experimental and fitted to Langmuir method mass loss rates due to evaporation in stepwise measurement. Using the stepwise model In order to determine the vapor pressure based on the Stepwise method, Equation 7.13 was applied. The Antoine coefficients were found by the Marquardt algorithm with their respective confidence interval using a confidence level of 99.9%. The parameters A, B and C are listed in Table 7.4 with their corresponding confident intervals. The intervals are very close; in the range of 10-12 to 10-15. Table 7.4. Antoine correlations for analyzed phase change materials obtained by Stepwise model. Antoine coefficient PCM A B C C16 22.12±3·10-12 5005.19±4·10-14 -80.84±9·10-14 RT6 19.75 ±3·10-13 4572.76 ±2·10-15 -39.81 ±5·10-14 RT21 17.55 ±3·10-13 4488.38 ±1·10-15 -25.46 ±3·10-14 RT27 20.44 ±3·10-12 5005.05 ±2·10-12 -75.36 ±1·10-11 RT31 19.35 ±2·10-12 4557.96 ±2·10-14 -80.64 ±1·10-14 The experimental data of mass loss rates are fitted with high accuracy to the Stepwise model. This can be observed for the investigated compounds and n-hexadecane in Figure 7.6. 201 Chapter 7 0.0 -1.0x10-6 -2.0x10-6 -3.0x10-6 dm/dt (g/s) dm/dt Experimental Theoretical -4.0x10-6 RT31 RT31 RT27 RT27 RT21 RT21 RT6 RT6 -5.0x10-6 C16 C16 320 340 360 380 400 420 440 Temperature (K) Stepwise Figure 7.6. Experimental and fitted to Stepwise method mass loss rates due to evaporation in stepwise measurement. Using the Ramp method The Ramp method was used to predict and reproduce the evaporation weight loss of the analyzed phase change materials. The numerical integration by Runge-Kutta and non-linear regression method of Marquardt were applied together in order to find the right values of the Antoine constants with confidence interval using a confidence level of 99.9%. The parameters A, B and C are listed in Table 7.5 with their corresponding confident interval. Table 7.5. Antoine correlations for analyzed PCMs obtained by Ramp model. Antoine coefficient PCM A B C C10 22.95 ±0.04 4782.52 ±0.002 -27.17 ±0.16 C16 21.74 ±0.28 5004.76 ±0.002 -71.80 ±0.12 RT6 21.50 ±0.20 4997.01 ±0.001 -43.26 ±0.11 RT21 18.91 ±0.19 4183.80 ±0.001 -64.29 ±0.11 RT27 20.55 ±0.44 4925.00 ±0.003 -87.95 ±0.11 RT31 20.51 ±0.50 5005.39 ±0.003 -85.07 ±0.12 Figure 7.7 shows the comparison between experimental weight losses measured by TGA with a ramp of 10 K min-1 compared to the estimated values. In order to extend this method, n-decane and n-hexadecane were also studied. As can be seen in Figure 7.7, 202 Thermal and morphological stability of microcapsules a good fitting of the experimental data is observed when using this model. An important advantage of the non-isothermal methodology is that it cover the complete temperature range of the evaporation. 1.0 0.8 0.6 Weight loss Weight 0.4 Experimental Theoretical RT31 RT31 RT27 RT27 0.2 RT21 RT21 RT6 RT6 C16 C16 C10 C10 0.0 200 250 300 350 400 450 500 550 Temperature (K) Figure 7.7. Weight loss vs. time for experimental analyses and fitted curves of the Ramp method for the investigated PCMs. In Figure 7.8, the correlation of the vapor pressure data from three considered models are shown. The symbols illustrates the experimentally calculated values whereas the lines correspond to the values obtained from the Antoine equation. 203 Chapter 7 2500 2500 RT6 experimental Langmuir RT21 experimental Langmuir RT6 fitted Langmuir RT21 fitted Langmuir 2000 RT6 experimental Stepwise 2000 RT21 experimental Stepwise RT6 fitted Stepwise RT21 fitting Stepwise RT6 Ramp method RT21 Ramp method 1500 1500 1000 1000 Vapor pressure (Pa) pressure Vapor Vapor pressure (Pa) pressure Vapor 500 500 0 0 320 340 360 380 400 420 440 320 340 360 380 400 420 440 Temperature (K) Temperature (K) 1000 1000 RT27 experimental Langmuir RT31 experimental Langmuir RT27 fitted Langmuir RT31 fitted Langmuir 800 RT27 experimental Stepwise 800 RT31 experimental Stepwise RT27 fitting Stepwise RT31 fitting Stepwise RT27 Ramp method RT31 Ramp method 600 600 400 400 Vapor pressure (Pa) pressure Vapor Vapor pressure (Pa) pressure Vapor 200 200 0 0 320 340 360 380 400 420 440 320 340 360 380 400 420 440 Temperature (K) Temperature (K) 100000 C16 fitted Literature C16 experimental Stepwise C16 fitted Stepwise 80000 C16 Ramp method 2500 60000 2000 1500 40000 1000 Vapor pressure (Pa) pressure Vapor Vapor pressure (Pa) pressure Vapor 500 20000 0 320 340 360 380 400 420 440 Temperature (K) 0 350 400 450 500 550 600 Temperature (K) Figure 7.8. Vapor pressure as function of temperature for the Langmuir, Stepwise and Ramp models. As expected, the vapor pressure increased with increasing temperature. Furthermore, the vapor pressure for each compound was raised with an increasing melting point. Interestingly, the vapor pressure values from the Stepwise method are slightly higher than those obtained by the Langmuir model. These differences can be explained by the reflectance of the temperature dependent diffusion phenomena in the Langmuir model. 204 Thermal and morphological stability of microcapsules In the approach, including mass transfer combined with Fick’s law of diffusivity, α cannot be longer consider as constant, since it is dependent on molecular weight, temperature, vapor-liquid interphase height and the diffusion coefficient of the substance [13]: 퐷 2휋푀 훼 = 퐴퐵 √ 퐴 (7.21) 푧 푅푇 On the other hand, the Ramp model predicts a pressure significantly higher than the semi-isothermal models in the case of RT6 and RT21, and similar to the other models in the case of RT27 and RT31. As far as we know, neither experimental nor theoretical data regarding vapor pressure and diffusion coefficients for these compounds are available in the literature. In order to compare these three models, the boiling temperatures of the compounds at normal atmospheric pressure were estimated. This can be obtained by extrapolating the curve of the vapor pressure as function of temperature to 101325 Pa [3]. This estimation did not give reasonable results in the case of the Langmuir model and a high deviation for the Stepwise model. This can be attributed to the narrow temperature range in which the measurements were performed and a final studied temperature notably lower than the expected boiling point. Furthermore, in the stepwise method errors from the integration of the isothermal step might affect the final boiling point prediction. Isothermal models are suitable within the analyzed temperature range but cannot be applied in prediction of the boiling point. Only the Ramp method allowed us to determine the boiling temperature of the investigated PCMs. The boiling temperature of the control compounds, n-decane and n-hexadecane were reported in the literature as 447.3 and 560.3 K [5], respectively whereas according to the Antoine parameters obtained in this study the expected boiling temperatures were 447.0 and 562.15 K. These results confirmed the suitability of this method to calculate the vapor pressure and for predicting the boiling point. Boiling temperature obtained for the other PCMs are tabulated in Table 7.6. As expected, the boiling points increase with the average chain length and melting point of the PCMs. 205 Chapter 7 The dependence of vapor pressure on the temperature can be also described by the Clausius – Clapeyron equation including the enthalpy of vaporization/sublimation (ΔH) [18,28]: 0 푑(푙푛푃퐴 ) ∆퐻 = (7.22) 푑푇 푅푇2 The results of the vaporization enthalpy were obtained from Equation 7.22 by plotting lnp (estimated from the vapor pressure obtained by Ramp method) and the reciprocal temperature (1/T). Enthalpy of vaporization was found from the slope of the linear trend line from a weight loss of 0.9 to 0.1 and the values are shown Table 7.6. For n-hexadecane, the enthalpy of vaporization at 482 K was reported as 61.7 kJ mol-1 which is very close to our value 58.87 kJ mol-1 of in the temperature range from 425 to 482 K [29]. The deviation between the experimental and literature values is -2.83 kJ mol-1. The other PCM exhibited increasing heat of vaporization, nevertheless the heat of vaporization of RT21 is slightly underestimated according to the trends of the PCM boiling points. Lower heat of vaporization in the studied temperature range can be caused by the mixture of alkanes with wide amount of hydrocarbons. Additionally, vacuum distillation of Rubitherm compounds and n-hexadecane were performed in order to experimentally estimate the boiling point. The nomograph diagram was applied to recalculate the boiling temperature at reduced pressure for normal boiling point. Since the Rubitherm compounds consist in a mixture of different paraffins, a wide evaporation temperature range was recorded. The initial temperature is recorded when the first drop of vapor appeared. For Rubitherm®RT6 the boiling started at 437 K, afterwards the main fraction evaporated from 541 to 561 K, which is in agreement with the calculated boiling temperature. A similar protocol was used for the other compounds. It can be observed for RT21 and RT27 that the estimated boiling temperatures are slightly higher than those obtained during distillation. The deviation can be due to the assumptions used in the model such as behavior of the vapor as ideal gas and constant liquid density or errors related with the predicted DAB. In the case of RT31, it was not possible to determine the boiling point by vacuum distillation due to decomposition of the material with temperature. This was caused by the long time and high temperature required before the distillation began [30,31]. Subsequently, only the boiling temperature for the lower molecular weight compounds could be recorded, giving a false result. 206 Thermal and morphological stability of microcapsules Table 7.6. Heat of vaporization, calculated boiling point and experimental boiling point for studied phase change materials. T-range ΔHvap Tb Tb vacuum distillation Substance (K) (kJ mol-1) (K) (K) C16 413-482 58.85 562 560 RT31 445-514 61.72 642 563-579 RT27 445-514 61.63 634 563-606 RT21 413-483 47.51 631 503-602 RT6 390-457 51.61 545 437-561 According to the results presented above, the proposed ramp model can be used to determine the vapor pressure of pure PCMs and also to quantify the boiling point and the vaporization enthalpy. These results also confirm that these materials are mainly vaporized in the TGA analyses and the decomposition of these compounds does not take place under a nitrogen atmosphere. Once the vapor pressure was determined, it is important to know the possible changes in the morphology of the microcapsules promoted by the high temperature. Heat treated microcapsules The SEM micrographs of heat-treated MC-P(St-DVB)68%PCM and MC-P(St- DVB-10%PNC-HEMA)68%PCM at the temperature range from 20 to 240°C are presented in Figure 7.9 and 7.10, respectively. The morphology of the microcapsules were practically the same as that of the neat sample, no broken microcapsules were observed similarly to previously reported findings [32]. Furthermore, no significant change in the particle size was noticed. However, this is contrary to results obtained by Sanchez et al. [33]. They reported that the microcapsules kept thermal stability until 91 °C and above glass transition temperature of polymer (73 °C), the microcapsules became more sticky and rubbery, converting finally into a liquid drop at 172 °C. This different behavior with temperature indicates that the crosslinked structure (copolymer and terpolymer) provides the shell structure with a significantly higher thermal and physical stability than a non- crosslinked polystyrene shell. The copolymer of St, DVB and PNC-HEMA is also more stable than a melamine-formaldehyde shell in which microcapsules were broken by the expansion of the core material [34]. 207 Chapter 7 20 °C 40 °C 60 °C 80 °C 100 °C 120 °C 140 °C 160 °C 180 °C 200 °C 220 °C 240 °C Figure 7.9. SEM micrographs of heat-treated microcapsules MC-P(St-DVB)68%PCM in temperature range 20 – 240 °C. 20 °C 40 °C 60 °C 80 °C 100 °C 120 °C 140 °C 160 °C 180 °C 200 °C 220 °C 240 °C Figure 7.10. SEM micrographs of heat-treated microcapsules MC-P(St-DVB-10%PNC- HEMA)68%PCM in temperature range 20 – 240 °C. 208 Thermal and morphological stability of microcapsules The latent heat of the heat-treated microcapsules in the temperature range from 80 to 200 °C were analyzed by DSC. The enthalpy decreased with increasing temperature. After treatment at 160 °C, the latent heat was reduced by 50% and finally reached zero enthalpy at 180 °C. This indicates that although the shell was intact, the PCM escapes from the shell at high temperatures. The seal tightness offered by the macro-reticulated matrix was not high enough to prevent the loss of the PCM [32]. Table 7.7. Latent heat of heat treated microcapsules. Temperature MC-P(St-DVB-10%PNC- MC-P(St-DVB)68%PCM (°C) HEMA)68%PCM Latent heat (J g-1) 80 101.4 104.8 100 100.5 103.3 120 99.53 100.53 140 95.47 98.63 160 51.73 52.63 180 0 0 200 0 0 A desorption model that allows for the influence of the vapor pressure of PCM on the diffusion of the PCM inside the polymeric matrix is going to be developed. 209 Chapter 7 Estimation of the diffusion coefficient of PCM from microcapsules a) Mathematical model R c r r+Δr Figure 7.11. Spherical particle representation. Considerations: The internal diffusion of PCM through the polymeric matrix to the external microparticle surface. Instantaneous vaporization of the liquid PCM that reaches the external surface. Mass transfer of the vaporized PCM from the external surface of the particle to the gas phase. An unsteady-state mass balance of species A in spherical microcapsules from an inner radius r to an outer radius r + Δr is considered. The radial flux is based on the total area (voids and solid) to diffusion transport. The mass balance over the path thickness Δr is equal to the molar loss observed in TGA analysis. 휕푛 퐴 푁 │ − 퐴 푁 │ = − (7.23) 푟 퐴 푟 푟 퐴 푟+∆푟 휕푡 2 where Ar is the total area of microcapsule (4휋푟 ), NA is the molar flux of the species A 휕푛 and is the molar loss in TGA analysis. The above equation can be transformed in terms 휕푡 of concentration as follow, 휕푐 4휋푟2푁 │ − 4휋푟2푁 │ = − 퐴 (4휋푟2∆푟) (7.24) 퐴 푟 퐴 푟+∆푟 휕푡 3 where cA is the concentration of A in mol/cm . 210 Thermal and morphological stability of microcapsules The flux equation in terms of concentration gradient is given by 휕푦 휕푐 푁푟 = −푐퐷 퐴 = −퐷 퐴 (7.25) 퐴 퐴퐵 휕푟 퐴퐵 휕푟 푑푦퐴 where 푐퐷 is the diffusional flux, c is the overall vapor concentration, DAB is the 퐴퐵 푑푟 diffusion coefficient of A in B, yA is the molar fraction of A. After substituting equation 7.25 into equation 7.24, the following differential equation is obtained, 1 휕푟2 휕푐 휕푐 (퐷 퐴) = 퐴 (7.26) 푟2 휕푟 퐴퐵 휕푟 휕푡 Orthogonal collocation method can be used to solve the above partial differential equation considering the symmetry of sphere and in terms of the dimensionless variables R and C. 푟 푐퐴 = 푅 and = 푐 (7.28) 푟0 푐퐴0 where r0 is the radii of the particle and CA0 is the initial concentration of PCM inside the microcapsules. 푁+1 1 휕 휕퐶 푑퐶𝑖 푅2퐷푔 │ = ∑ 퐵 퐶 = (7.29) 푅2 휕푅 퐴퐵 휕푅 푅푖 푗푖 푖 푑푡 푖=1 The above partial differential equation is extended to the N number of nodes of orthogonal collocations (6) as shown in Figure 7.12. N 1 2 3 4 5 6 N+1 Figure 7.12. Microcapsule with displayed N number of nodes for orthogonal collocation. 211 Chapter 7 The nodes 0 and N+1 are considered as the boundary conditions and 0 is not required in orthogonal collocation, being the concentration at N+1 obtained by the boundary condition. In order to solve the partial differential equation is required the boundary condition and in this case, is the external flux which was defined in the above section for a vapor that diffuses into a stagnant gas, this is (eq 7.30) 푔 1 푑푦퐴 푁퐴 = −푐퐷퐴퐵 (7.30) 1 − 푦퐴 푑푧 Contrary to what happened in the case of liquid, z is constant for the particles because the particle size of the microcapsules does not change during the heating process and thus, the above gradient of dy/dz can be transformed to, 푑푦 ∆푦 푦 − 0 퐴 = = 퐴 (7.31) 푑푧 ∆푧 푧 − 0 푦퐴 = 푦퐴 푓표푟 푧 = 푧 where z is the distance of the gas phase path (cm), 푦퐴 = 0 푓표푟 푧 = 푧0 = 0 z0 is the upper part of the pan, 푝̅̅퐴̅ 푦퐴 = (7.32) 푝푇 With this equation is possible to correlate the total amount PCM that evaporates through the surface of the microcapsules with the net flux of PCM that scape by the gas phase. 푑푐 푁 │ · 푆 = 퐷푠 퐴 │ · 퐴 (7.33) 퐴 푝푎푛 푎푟푒푎 퐴퐵 푑푅 푝푎푟푡푖푐푙푒푠 푎푟푒푎 The total area of the microcapsules can be obtained by a relationship stablished between the area and volume of a single particle and the total weight of the microcapsules as follow, 3 푤 푠 푑푐퐴 푆 푁퐴 = 퐷퐴퐵 (7.34) 푅푃 휌 푑푅푃 212 Thermal and morphological stability of microcapsules Considering the PCM as an ideal gas, c can be replaced by PT/RT. 푔 푤푀퐶 푠 푑푐퐴 푝푇 퐷퐴퐵푦퐴 3 퐷퐴퐵 = 푆 (7.35) 휌푀퐶푅0 푑푟 푅푔푎푠푇 푧(1−푦퐴) Changing for dimensionless conditions 푐퐴 푠 푑 푤푀퐶 퐷퐴퐵 푐퐴0 푝푇 푔 푦퐴 푐퐴03 푟 = 퐷퐴퐵 푆 (7.36) 휌푀퐶푅0 푅0 푑 푅푔푎푠푇 푧 (1 − 푦퐴) 푅0 Finally, 푝푇 푅 푇 푔 푑푐 1 푔푎푠 휌푀퐶 퐷퐴퐵 푦퐴 2 (7.37) = − 푠 푅0 푆 푑푅 3 푐퐴0 푤푀퐶 퐷퐴퐵 푧 (1 − 푦퐴) The molar fraction can be defined according Dalton equation and considering that in the case of a solid material the total surface is not completely covered by the liquid, then the partial pressure can be defined as a fraction (k) of the vapor pressure. 0 0 푝 푝̅ = 푘푝 ∴ 푦퐴 = 푘 (7.38) 푝푇 푝0 푔 푑푐 1 푝 휌 퐷 (푘 푝 ) = (− ) 푇 푅2 푆 푀퐶 퐴퐵 푇 (7.39) 푑푅 3 푅 푇푐 0 푤 퐷푠 푝0 푔푎푠 퐴0 푀퐶 퐴퐵 푧 (1 − 푘 ) 푝푇 where CA0 is, 휌푤푖,푃퐶푀 (7.40) 퐶퐴0 = 푀퐴,푃퐶푀 This boundary condition must be also solved by using orthogonal collocation and using 푝0 the pi in the dimensionless = 푝푖, 푝푇 푁+1 푑푐 = ∑ 퐴 푦 (7.41) 푑푟 1,푖 푖 푖=1 Thus, the set of N differential equations is solved by numerical methods and satisfying the boundary condition (7.41). 213 Chapter 7 푠 In order to find the both unknown parameters 퐷퐴퐵 and k, the average theoretical concentration of PCM into the microcapsules at each time (푐̅푡) must be found. 푟 푟 =1 =1 ∫푅0 4푐 휋푟2푑푟 ∫푅0 4푐 휋푟2푑푟 1 0 푖 푖 0 푖 푖 2 푐̅푡 = = = 3 ∫ 푐푖푟푖 푑푟 (7.42) 푉푇 4 3 0 3 휋푅푖 This average concentration is perfectly correlated with the experimental weight loss that microcapsules present at each time and quantified by the TGA analyses. 퐶푃퐶푀(푇퐺퐴) ∙ 푤푒𝑖푔ℎ푡 푙표푠푠 푃퐶푀│푡 1 − 퐶푃퐶푀(푇퐺퐴) 푐̅ = 푤푒𝑖푔푡ℎ 푙표푠푠 = 푠표푙𝑖푑 = (7.43) 푡 퐶 푃퐶푀│푖푛푖푡푖푎푙 푃퐶푀(푇퐺퐴) 푠표푙𝑖푑 1 − 퐶푃퐶푀(푇퐺퐴) where PCM is the mass of PCM in microcapsules at the time 0 or t, solid correspond to fraction of polymer in microcapsules, and 퐶푃퐶푀(푇퐺퐴) is a PCM content in TGA analysis. Thus, the unknown parameters can be found by fitting the experimental data with the predicted values by means of the Marquardt algorithm described before. b) Results Microcapsules with encapsulated paraffin Rubitherm®RT6, RT21 and RT27 have been analyzed in stepwise mode by thermogravimetric analysis in order to estimate the diffusion coefficient of PCM through microcapsules. Density measured by helium pycnometer and radius of the microcapsules required in the model are listed in Table 7.8. Table 7.8. Properties of the investigated microcapsules. Sample Density (g cm-3) R (cm) MC(RT6) 0.9153 0.0073 MC(RT21) 0.9024 0.0068 MC(RT27) 0.8942 0.0119 Initially, the average concentration of PCM in the microcapsules at each time (푐̅푡) was found from TGA analysis. The experimental and fitted data are plotted in Figure 7.13. Three temperatures were selected for analyzing the diffusion of the PCM inside the 214 Thermal and morphological stability of microcapsules polymeric matrix. For MC(RT6) it was 323 K and for MC(RT21) and MC(RT27) a temperature of 353 K was chosen. Thus, the volatility trend was not maintained because MC(RT6) was analyzed at a lower temperature. Good correlation between experimental and theoretical data can be observed. 1.000 0.999 0.998 0.997 t C 0.996 0.995 Experimental Theoretical MC(RT27) MC(RT27) 0.994 MC(RT21) MC(RT21) MC(RT6) MC(RT6) 0.993 0 1 2 3 4 5 6 7 Time (min) Figure 7.13. Experimental and theoretical data for the average concentration of PCM into the microcapsules at each time. Examples of the internal profiles for the microcapsules at seven different distances changed with the time can be seen in Figure 7.14. Distances correspond to the radius dimensionless 0.23046, 0.44849, 0.64235, 0.80158, 0.9176, 0.98418, and 1. 215 Chapter 7 (a) (b) (c) Figure 7.14. Distribution of the concentrations along the microcapsule radius versus time (a) MC(RT6) at 323 K, (b) MC(RT21) at 353 K and (c) MC(RT27) at 353 K. A good stability can be observed for the method for solving the differential equations. As expected, the concentration decreased from the external part of the particle to the center, being the lowest at the surface of the microcapsules. The changes observed in the concentration at the surface were 0.0057, 0.0079, and 0.0024 for MC(RT6), MC(RT21) and MC(RT27), respectively. Although the decrease is little, it might imply that the diffusion of the PCM takes place at low temperatures although the matrix of the microcapsules has a high thermal stability as observed before. The PCM loss can be attributed to the matrix porosity which is promoted by the presence of the porogen, toluene in the synthesis process. This porosity enhances the loss of the PCM at high temperatures by diffusion through the polymeric matrix of the microcapsules. This results in a decrease of the PCM content and thus in the thermal capacity. 216 Thermal and morphological stability of microcapsules The diffusion coefficients of PCM through the microcapsules were fitted to experimental values by non-linear fitting using the Marquardt algorithm. The diffusion coefficients and k values are listed in Table 7.9 together with results obtained from the Fuller equation for binary diffusion coefficients of PCM – N2 determined in previous section (7.3.2). Lower k value was observed for MC(RT21) than for MC(RT27) and MC(RT6). This can be attributed to a slightly higher porosity of these microcapsules (2.17%), whereas for MC(RT27) and MC(RT6) values of 1.41 and 0.53% were obtained, respectively. Although some diffusion of the PCM from microcapsules is observed, the value is about 108 times lower than for neat PCM. This indicate that despite the porosity, microencapsulation was able to accomplish the purpose and decrease leakage and diffusion of the core material. In agreement with ramp TGA and DSC analyses, the core release at low temperatures is relatively low, nevertheless at 120 °C, this change would be important, consequently limiting the application of the microcapsules to temperatures below 120 °C as mentioned previously. Table 7.9. Diffusion coefficients for analyzed PCMs and microcapsules. Diffusion coefficient from Binary diffusion coefficient PCM k 2 -1 2 -1 microcapsules (cm s ) of PCM-N2 (cm s ) RT27 @353 K 0.005 8.12·10-10 5.87·10-2 RT21 @353 K 0.004 1.04·10-9 6.01·10-2 RT6 @323 K 0.005 8.35·10-10 6.40·10-2 The proposed model fits perfectly the experimental data obtained by TGA and it also allows to know the role that the vapor pressure of the PCM presents on its scape from the polymeric matrix by vaporization at the microcapsule surface. This is, the higher the vapor pressure, the higher the vaporization in the gas phase and the diffusion through the polymeric matrix is enhanced. 7.4. Conclusions In TGA, microcapsules were observed to start releasing the PCM at 120 °C. For that reason initially, the vapor pressures of a series of Rubitherm phase change materials have been estimated using different models based on a Langmuir vaporization relation 217 Chapter 7 and the mass transfer combined with Fick’s law of diffusivity. Different thermogravimetric measurements in the stepwise and non-isothermal modes were performed. Antoine parameters for vapor pressure were determined by fitting the experimental data to the proposed models, and a good accuracy was observed in all cases. Nevertheless, when the stepwise method is used, the final vaporization temperature was lower than those observed in non-isothermal analysis, thus the temperature range applicability of the fitting parameters was narrower. The non-isothermal method was extended to two additional known materials, observing that it was able to predict the boiling point and also the heat of vaporization by applying the Clausius - Clapeyron equation. Finally, the accuracy of the boiling points estimated with this method for PCMs were compared with those obtained a commonly used technique – vacuum distillation, and a good agreement was found although some differences can be observed because they are not pure components. Hence, the low cost and fast thermogravimetric analysis in the most commonly used non-isothermal mode can be used to provide information about vapor pressure, boiling point and heat of vaporization which are important thermal properties in the engineering industries. After that, microcapsules morphology was analyzed in situ by SEM from room temperature to 240 °C, showing good dimensional stability due to crosslinking process. However, in increasing temperature the latent heat decreased by 50% at 160 °C and being 0 J g-1 at 180 °C. Diffusion of PCM from microcapsules was studied and diffusion coefficient was fitted to the model. Diffusion coefficient of PCM from solid microcapsules was lower than pure PCM, thus microencapsulation accomplished requirements. 218 Thermal and morphological stability of microcapsules References 1. Malewska E, Prociak A (2017) Porous polyurethane-polystyrene composites produced in a co-expansion process. Arabian Journal of Chemistry. doi:https://doi.org/10.1016/j.arabjc.2017.01.014 2. Stanley C How hot is your concrete & does it really matter? Heat evolution & its effects on concrete cast on construction sites in Asia. In: Proceedings of the 30th International Conference on Our World in Concrete and Structures (OWICs05), 2005. pp 23-24 3. Price DM (2001) Vapor pressure determination by thermogravimetry. Thermochimica Acta 367 (Supplement C):253-262. doi:https://doi.org/10.1016/S0040-6031(00)00676-6 4. Cuddy MF, Poda AR, Chappell MA (2014) Estimations of Vapor Pressures by Thermogravimetric Analysis of the Insensitive Munitions IMX-101, IMX-104, and Individual Components. Propellants, Explosives, Pyrotechnics 39 (2):236-242. doi:10.1002/prep.201300069 5. Järvik O, Rannaveski R, Roo E, Oja V (2014) Evaluation of vapor pressures of 5- Methylresorcinol derivatives by thermogravimetric analysis. Thermochimica Acta 590:198-205. doi:https://doi.org/10.1016/j.tca.2014.07.001 6. Barontini F, Cozzani V (2007) Thermogravimetry as a screening tool for the estimation of the vapor pressures of pure compounds. Journal of Thermal Analysis and Calorimetry 89 (1):309-314 7. Bassi M (2011) Estimation of the vapor pressure of PFPEs by TGA. Thermochimica Acta 521 (1):197-201 8. Langmuir I (1913) The Vapor Pressure of Metallic Tungsten. Physical Review 2 (5):329-342 9. Mbah J, Knott D, Steward S (2014) Thermogravimetric study of vapor pressure of TATP synthesized without recrystallization. Talanta 129 (Supplement C):586-593. doi:https://doi.org/10.1016/j.talanta.2014.06.031 10. Chatterjee K, Hazra A, Dollimore D, Alexander KS (2002) Estimating vapor pressure curves by thermogravimetry: a rapid and convenient method for characterization of pharmaceuticals. European Journal of Pharmaceutics and Biopharmaceutics 54 (2):171- 180. doi:https://doi.org/10.1016/S0939-6411(02)00079-6 219 Chapter 7 11. Oliveira CELd, Cremasco MA (2014) Determination of the vapor pressure of Lippia gracilis Schum essential oil by thermogravimetric analysis. Thermochimica Acta 577 (Supplement C):1-4. doi:https://doi.org/10.1016/j.tca.2013.11.023 12. Ruz V, González M, Winant D, Rodríguez Z, Van den Mooter G (2015) Characterization of the Sublimation and Vapor Pressure of 2-(2-Nitrovinyl) Furan (G-0) Using Thermogravimetric Analysis: Effects of Complexation with Cyclodextrins. Molecules 20 (8):15175 13. Pieterse N, Focke WW (2003) Diffusion-controlled evaporation through a stagnant gas: estimating low vapour pressures from thermogravimetric data. Thermochimica Acta 406 (1):191-198. doi:https://doi.org/10.1016/S0040-6031(03)00256-9 14. Behzadi S, Farid MM (2014) Long term thermal stability of organic PCMs. Applied Energy 122 (Supplement C):11-16. doi:https://doi.org/10.1016/j.apenergy.2014.01.032 15. Price DM, Hawkins M (1998) Calorimetry of two disperse dyes using thermogravimetry. Thermochimica Acta 315 (1):19-24 16. Verevkin SP, Ralys RV, Zaitsau DH, Emel’yanenko VN, Schick C (2012) Express thermo-gravimetric method for the vaporization enthalpies appraisal for very low volatile molecular and ionic compounds. Thermochimica Acta 538:55-62 17. Shahbaz K, Mjalli FS, Vakili-Nezhaad G, AlNashef IM, Asadov A, Farid MM (2016) Thermogravimetric measurement of deep eutectic solvents vapor pressure. Journal of Molecular Liquids 222:61-66 18. Wright SF, Dollimore D, Dunn JG, Alexander K (2004) Determination of the vapor pressure curves of adipic acid and triethanolamine using thermogravimetric analysis. Thermochimica Acta 421 (1):25-30. doi:https://doi.org/10.1016/j.tca.2004.02.021 19. Sánchez-Silva L, Rodríguez JF, Sánchez P (2011) Influence of different suspension stabilizers on the preparation of Rubitherm RT31 microcapsules. Colloids and Surfaces A: Physicochemical and Engineering Aspects 390 (1-3):62-66. doi:10.1016/j.colsurfa.2011.09.004 20. Sánchez L, Sánchez P, Lucas A, Carmona M, Rodríguez JF (2007) Microencapsulation of PCMs with a polystyrene shell. Colloid and Polymer Science 285 (12):1377-1385. doi:10.1007/s00396-007-1696-7 21. Welty JR, Wicks CE, Rorrer G, Wilson RE (2009) Fundamentals of momentum, heat, and mass transfer. John Wiley & Sons, 220 Thermal and morphological stability of microcapsules 22. Fuller EN, Schettler PD, Giddings JC (1966) New method for prediction of binary gas-phase diffusion coefficients. Industrial & Engineering Chemistry 58 (5):18-27. doi:10.1021/ie50677a007 23. Poling BE, Prausnitz JM, O'Connell JP (2001) The properties of gases and liquids. McGraw-Hill, 24. Camin DL, Forziati AF, Rossini FD (1954) Physical properties of n-hexadecane, n- decylcyclopentane, n-decylcyclohexane, 1-hexadecene and n-decylbenzene. The Journal of Physical Chemistry 58 (5):440-442 25. Morgan DL, Kobayashi R (1994) Direct vapor pressure measurements of ten n- alkanes m the 10-C28 range. Fluid Phase Equilibria 97:211-242 26. Stull DR (1947) Vapor pressure of pure substances. Organic and inorganic compounds. Industrial & Engineering Chemistry 39 (4):517-540 27. Tevault DE, Buettner LC, Crouse KL (2014) Vapor Pressure of Methyl Salicylate and n-Hexadecane. ARMY EDGEWOOD CHEMICAL BIOLOGICAL CENTER APG MD, 28. Gupta PK, Ganesan K, Gutch PK, Manral L, Dubey DK (2008) Vapor pressure and enthalpy of vaporization of fentanyl. Journal of Chemical & Engineering Data 53 (3):841- 845 29. Linstrom PJ, Mallard WG NIST Chemistry WebBook; NIST Standard Reference Database No. 69. National Institute of Standards and Technology. http://webbook.nist.gov/. 30. Jaw K-S, Hsu C-K, Lee J-S (2001) The thermal decomposition behaviors of stearic acid, paraffin wax and polyvinyl butyral. Thermochimica Acta 367-368 (Supplement C):165-168. doi:https://doi.org/10.1016/S0040-6031(00)00680-8 31. Hu S, Wu G, Hua Y, Rashid N, Hu H (2015) Study on Thermal Degradation Characteristics and Regression Rate Measurement of Paraffin-Based Fuel. Energies 8 (9):10058 32. Li W, Song G, Tang G, Chu X, Ma S, Liu C (2011) Morphology, structure and thermal stability of microencapsulated phase change material with copolymer shell. Energy 36 (2):785-791. doi:https://doi.org/10.1016/j.energy.2010.12.041 33. Sánchez -Silva L, Rodríguez JF, Carmona M, Romero A, Sánchez P (2011) Thermal and morphological stability of polystyrene microcapsules containing phase-change materials. Journal of Applied Polymer Science 120 (1):291-297. doi:10.1002/app.33112 221 Chapter 7 34. Zhang X-x, Tao X-m, Yick K-l, Wang X-c (2004) Structure and thermal stability of microencapsulated phase-change materials. Colloid and Polymer Science 282 (4):330- 336. doi:10.1007/s00396-003-0925-y 222 8. Incorporation of microcapsules into concrete INCORPORATION OF MICROCAPSULES CHAPTER 8 INTO CONCRETE Chapter 8 Abstract The effect of microencapsulated phase change material (MPCM) on the mechanical and thermal properties of Portland cement concrete (PCC) were investigated. Portland cement concrete containing different amounts of MPCM (0, 10, 20 and 40%) were prepared and cured at both 20 °C and 40 °C which correspond to the temperatures below and above melting point of phase change material. It was found that the compressive strength of PCC decreased with the addition of MPCM at both tested temperatures. Thermal properties such as thermal conductivity, specific heat capacity, power consumption and power reduction were determined. The thermal conductivity decreased with increasing amount of MPCM whereas the specific heat capacity increased. The power consumption was 0.14 kWh m-2 lower for concrete with 40% of microcapsules than for concrete without MPCM. In that way, the power reduction of 20% was obtained. X-ray tomography imaging was utilized to examine the dispersibility of MPCM in the concrete sample. SEM imaging disclosed air gaps formed between the microcapsules and the concrete matrix. Cone calorimetry was used in order to evaluate the flammability and the negative effect of incorporation of MPCM into concrete, however at a heat flux of 50 kW m-2 ignition was not observed. 224 Incorporation of microcapsules into concrete 8.1. Background Concrete is widely used as a construction material in many applications from sidewalks to bridges, buildings or skyscrapers, as well as concrete pavement. Concrete is a composite material made by combining aggregate (gravel and sand), Portland cement and water. From a practical point of view, the water hydrates the cement and holds all the aggregates together. After mixing, the concrete can be cast in almost any shape desired, and once hardened, it can become a structural element [1]. Phase change materials (PCMs) can be integrated into building materials such as concrete, which has high volume and surface area in contact with indoor environment and solar radiation. In that way, a smart materials suitable for passive house construction are created. PCM incorporated into concrete may improve the thermal energy storage capacity, and reduce the energy consumption required for heating and cooling systems, or shift the heat peak hours during the day [2]. However, some limitations for utilizing bulk quantities of PCM have been discovered. Direct incorporation of the PCM significantly reduces the thermal conductivity of the concrete and may cause interactions between PCM and the matrix material [3-5]. These problems can be overcome by microencapsulation of PCM [6,7]. As mentioned in Chapter 2, microencapsulation ensures a high heat transfer area, prevents interactions between PCM and the building structure, and withstands volume change during phase change [8-14]. Microencapsulated PCMs (MPCM) are promising for reducing the temperature fluctuations inside buildings [15,16]. Nevertheless, microcapsules have been found to reduce the compressive strength of building materials [17-20]. Self-compacting concrete with incorporated microencapsulated PCMs (Micronal DS 5008 X) composed of a paraffin wax in poly(methyl methacrylate) shell exhibited significant improvement of the thermal performance of the material together with a decrease of the compressive strength due to the higher porosity [19]. Utilizing two real size concrete models with 5 wt. % PCM incorporated in the concrete have been shown to shift the maximum temperature in the wall to 2 hours later than without PCM [16]. In addition, the compressive strength was over 25 MPa and the tensile splitting strength was over 6 MPa after 28 days. Portland cement concrete with MPCM (Micronal DS 5001 X) [21] exhibited improved heat storage capacity, loss in compressive strength, while the 225 Chapter 8 thermal conductivity was not affected by MPCM addition. Incorporation of MPCM (MPCM 28-D, Microtek Laboratories, Inc.) in geopolymer mortar [22] leads to a slight decrease of the compressive strength keeping it sufficiently high for applications in buildings. During a 6.5 h temperature rising period, the air temperature inside cubicles with 10% and 20% PCM geopolymer mortar slabs were 4.4 and 5.5 °C lower than that inside a corresponding cubicle without PCM, respectively. The reduced rising of the inside air temperature over a long period of time leads to a substantial reduction of energy consumption for cooling of the building. Several reasons have been reported to affect the compressive strength reduction. It has been suggested that the decreased compressive strength might be caused by the rupture of microcapsules and leakage of PCM during the mixing process [19]. In addition, interface gaps between microcapsules shell and concrete matrix have been observed [20]. Furthermore, microcapsules exhibit lower stiffness and strength compare to sand it replaces [23]. Thermoregulating microcapsules with shell composed of styrene-divinylbenzene copolymer present high mechanical resistance and are inert in the concrete media. In this chapter, the effect of incorporation of MPCM into PCC on the mechanical and thermal properties of PCC are investigated. Special focus will be on the amount of sand replaced by MPCM as well as the curing temperature of concrete below (20 °C) and above (40 °C) the melting temperature of PCM). In addition, the thermal properties such as the thermal conductivity and the heat storage capacity are investigated. Moreover, fire resistant tests utilizing cone calorimetry are performed in order to evaluate the influence of flammable MPCM. 8.2. Materials Microcapsules obtained at pilot plant scale in Chapter 4 with a PVP concentration of 5.03% and theoretically 68% PCM were selected to analyze the effect of incorporated MPCM in Portland cement concrete on thermal and mechanical properties. They are denoted as MC-P(St-DVB). Furthermore, the microcapsules with 50% PCM and 0% PNC-HEMA (denoted as MC-P(St-DVB)50) as well as 50% PCM and 10% PNC-HEMA (denoted as MC-P(St-DVB-10%PNC-HEMA)50) were selected for fire resistant studies. 226 Incorporation of microcapsules into concrete The properties of the raw materials used in the concrete preparation are shown in Table 8.1. Table 8.1. Properties of raw materials used in the Portland cement concrete preparation. Compound Blaine fineness (cm2 g-1) Density (g cm-3) Portland cement II mixed with fly ash 4500 3.0 Dynamon SR-N Admixture - 1.1 Sand - 2.7 Gravel - 2.6 MPCM - 0.9 The mixture design of Portland cement concrete (PCC) without and with microencapsulated phase change material is displayed in Table 8.2. The percentages given in the sample names indicate the volume of sand replaced by microcapsules. For that purpose, the required amount of MPCM in the mixture was calculated by its volume percentage and replaced a certain volume percentage of sand [23]. In addition, concrete with incorporated 10% of microcapsules MC-P(St-DVB-10%PNC-HEMA)50 are denoted as PCC10FR. Table 8.2. Composition of Portland cement concrete (PCC) mixtures. MPCM Cement Water Admixture Sand Gravel MPCM Sample (vol.%) (g) (g) (g) (g) (g) (g) PCC 0 434 191.8 5.6 1057 705 0 PCC10 10 434 191.8 5.6 951.3 705 36 PCC20 20 434 191.8 5.6 845.6 705 72 PCC40 40 434 191.8 5.6 634.2 705 140 227 Chapter 8 8.3. Results Compressive strength The effect of microcapsules MC-P(St-DVB) on the mechanical properties of Portland cement concrete (PCC) was investigated by measuring compressive strength. The testes were performed with varied amounts of microcapsules (0, 10 and 20%) as well as at different temperatures (20 and 40 °C). The results are shown in Figure 8.1. It should be pointed out that both curing and measurements were performed at indicated temperatures, which can affect the final mechanical properties. 70 (a) 70 (b) 60 60 50 50 40 40 30 30 20 20 Compressive strength Compressive (MPa) strength Compressive strength Compressive (MPa) strength PCC10 @20 C PCC @20 C 10 10 PCC10 @40 C PCC @40 C 0 0 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Curing time (days) Curing time (days) 70 (c) 70 (d) 60 60 50 50 40 40 30 30 20 20 Day 28 @20 C Compressive strength Compressive (MPa) strength PCC20 @20 C Compressive (MPa) strength 10 PCC20 @40 C 10 Day 28 @40 C 0 0 0 5 10 15 20 25 30 0 10 20 Curing time (days) MPCM (%) Figure 8.1. Compressive strength of PCC cured at 20 and 40 °C versus curing time for (a) 0% MPCM, (b) 10% MPCM, and (c) 20% MPCM and (d) versus percentage of sand replaced by MPCM after curing for 28 days. Line ( ) at 20 MPa indicate the minimum compressive strenght of structural concrete according to spanish legislation [24]. 228 Incorporation of microcapsules into concrete 50 (a) 50 (b) 40 40 30 30 20 20 Strength reduction (%) reduction Strength Strength reduction (%) reduction Strength 10 10 PCC10 @20C PCC10 @40C PCC20 @20C PCC20 @40C 0 0 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Curing time (days) Curing time (days) 50 (c) 40 30 20 Strength reduction (%) reduction Strength 10 @20 C @40 C 0 0 10 20 MPCM (%) Figure 8.2. The compressive strength reduction (percentage compared to samples without MPCM at the same curing time) for PCC as function of curing time at (a) 20 °C and (b) 40 °C, and after 28 days as a function of MPCM concentration (c). The effect of MPCM addition in the concrete was studied. As can be seen in Figure 8.1, the compressive strength of PCC decreased with increasing amount of MPCM at both, curing times and temperatures. The decrease in compressive strength can be caused by lower stiffness and shear stress resistance of the microcapsules compared to the sand it replaces. In that way, microcapsules could be damaged or deformed during mixing of the concrete or during the compressive strength test [22]. Moreover, the strength might decrease due to voids and air bubbles induced by the microcapsules [25]. The hydrophobic character of MPCM results in formation of air gaps between the MPCM and the concrete matrix (Figure 8.4), which will affect the mechanical properties [20]. The percentage strength reduction of PCC compared to samples without MPCM are plotted in Figure 8.2. The strength reduction at both temperatures decreased with increasing curing time, and after 7 days it was stabilized at 20 and 40% for PCC10 and 229 Chapter 8 PCC20, respectively. It is important to note that despite the compressive strength reduction, the mechanical properties are still sufficient for building applications, where the minimum required compressive strength for structural concrete is 20 MPa [24]. The influence of curing temperature on compressive strength of concrete is shown in Figure 8.1a, b and c. As can be seen the compressive strength is higher at 40 °C than at 20 °C. This is due to a faster hydration rate at higher temperatures, what is in agreement with previous findings [26,27]. The curing and tested temperatures have direct influence on the encapsulated PCM. At 20 and 40 °C, the PCM core material of the microcapsules is in a solid and liquid state, respectively (the melting point of the PCM is 24 °C). The percentage compressive strength reduction can be used to examine the effect of solid and liquid state of the PCM. The relation of strength reduction with respect to the strength of a sample without MPCM cured at the same temperature and time helps to diminish the influence of curing conditions on the final result. The percentage strength reduction after 28 days was taken under consideration since the compressive strength is stabilized after long curing times, as can be observed in Figure 8.1. From Figure 8.2, it can be seen that there is a small difference of maximum 2% in strength reduction for PCC with 10 and 20% MPCM between curing temperatures. A liquid state of PCM can result in softer microcapsules and consequently in a weaker concrete matrix, especially with high amounts of microcapsules. However, in this study neither solid nor liquid state have significant effect on the mechanical properties of PCC. This is probably due to matrix structure and crosslinking of microcapsules which ensures dimensional stability even at high temperature as proved in Chapter 7. Nevertheless, both curing and measurement temperature affect the final compressive strength, which is higher for the samples cured at high temperatures. X-ray micro-tomography In order to evaluate the internal structure of the concrete sample, distribution of microcapsules and air voids, 2D X-ray micro-tomography were performed. 2D X-ray micro-tomography cross-sectional slices obtained from PCC without and with 20% MPCM, are shown in Figure 8.3. Dark colors correspond to the air bubbles and MPCM 230 Incorporation of microcapsules into concrete due to lack of absorption of X-rays whereas bright colors are related to sand and gravel which can easily adsorb high amounts of X-rays. (a) (b) Air bubble Air bubble MPCM PCC PCC20 Figure 8.3. X-ray-tomography images of samples (a) PCC without MPCM, (b) PCC with 20% MPCM. Taking into account the low level of X-ray attenuation of organic materials constituting MPCM, it is difficult to distinguish the microcapsules from air bubbles based on gray scale values. Nevertheless, some estimation based on the shape can be carried out since air voids should exhibit a spherical shape due to interfacial tension whereas MPCM may have more irregular shape due to agglomeration. It can be observed that microcapsules are mostly well dispersed in the central part of the concrete matrix with some agglomeration. SEM analysis The PCC samples with 20% MPCM after 28 days curing at 20 and 40 °C were chosen for SEM analysis and compared with a sample without MPCM cured at 20 °C. SEM images are displayed in Figure 8.4. It can be noted that air gaps from 8 to 16 μm between the microcapsules and the concrete matrix were observed. These gaps indicate weak bonding between the MPCM and concrete matrix as well as that air voids in the concrete samples might be caused by 231 Chapter 8 the presence of MPCM. Microcapsules did not bind or associate well with the concrete matrix due to the hydrophobic nature of the MPCM shell and hydrophilic character of the concrete. This is contrary to compatibility observed by Aguayo et al. [28]. However, in that study microcapsules with a hydrophilic shell (poly(methyl methacrylate) and melamine-formaldehyde copolymer) were used. Some broken microcapsules can be observed e.g. in Figure 8.4d. It is important to point out that the microcapsules have a matrix-like configuration as discussed in Chapter 4. In MPCM with a matrix-like internal structure, PCM is less susceptible to leak out than in microcapsules with core-shell structure. It is not certain whether the damaged microcapsules are caused by the mixing process during concrete preparation or by the sample preparation for the SEM analysis. Figure 8.4e illustrates the homogeneous concrete matrix for a sample without microcapsules. (a) (b) 8 μm 103 μm 16 μm (c) (d(d)) 16 μm 103 μm 10.8 μm 232 8 μm Incorporation of microcapsules into concrete (e) Figure 8.4. SEM images of fracture surface of 20% MPCM incorporated in (a) PCC at 20 °C, (b) dimensioning of MPCM and air gaps in PCC at 20 °C, (c) PCC at 40 °C, (d) dimensioning of MPCM and air gaps in PCC at 40 °C, (e) PCC without MPCM at 20 °C. Thermal properties The thermal conductivity of PCC containing microcapsules at temperatures below and above the melting point of PCM is shown in Figure 8.5. There is an obvious reduction of the thermal conductivity with increasing amount of MPCM. The decrease of thermal conductivity of PCC with MPCM might be caused by the lower thermal conductivity of the microcapsules respect to that of sand. The thermal conductivity of sand has an average value between 1.80-2.50 W m-1 °C-1 (information provided by supplier) while the PCM Rubitherm®RT27 has a thermal conductivity of 0.2 W m-1 °C -1 [29] and poly(styrene-co- divinylbenzene) of 0.14-0.18 W m-1 °C -1 [29,30], which is approximately 10 times lower than sand. In addition, a higher porosity and air voids formed between MPCM and the concrete matrix also diminish the thermal conductivity of the samples. The thermal conductivity of concrete when PCM is in a liquid state is somewhat lower than in a solid state because of differences in the thermal conductivity of neat PCM in solid and liquid states [31,32]. 233 Chapter 8 1.6 PCM in solid state PCM in liquid state ) -1 1.4 C -1 1.2 1.0 0.8 0.6 0.4 Thermal conductivity (W m (W conductivity Thermal 0.2 0.0 PCC PCC10 PCC20 PCC40 Figure 8.5. The thermal conductivity of concrete samples with different amounts of MPCM. Specific heat capacities of PCC in liquid and solid were measured in the temperature range 10-15 °C and 35-40 °C, respectively. From Figure 8.6 it can be seen that the difference in specific heat capacity of PCC vary about 8% between solid and liquid state of PCM, which is slightly higher than the 2% previously reported by Joulin et al. [33]. The average specific heat capacity increased from 939 to 990 J Kg-1 °C -1 for PCC without microcapsules and with 20% of MPCM, respectively. This can be explained by a higher specific heat capacity of MPCM compared to that of concrete. However, amount of MPCM incorporated into concrete is very small, thus the difference is also little. Similar observations can be found in a previous study [34]. 234 Incorporation of microcapsules into concrete 1200 PCM in solid state ) PCM in liquid state -1 1000 C -1 800 600 400 200 Specific heat capacity (J Kg (J capacity Specific heat 0 PCC PCC10 PCC20 PCC40 Figure 8.6. The specific heat capacity as a function of MPCM concentration. The latent heat of the concrete samples was analyzed within the temperature range from 10 to 35 ºC and the results are presented in Figure 8.7. The latent heat of PCC increases linearly with the increasing microcapsule concentration from 0 to 4.95 J g-1 for 0% of MPCM and 40% of MPCM, respectively. 5 4 ) -1 3 2 Latent heat (J g (J heat Latent 1 0 PCC PCC10 PCC20 PCC40 Figure 8.7. The latent heat of PCC as the function of MPCM concentration. 235 Chapter 8 Energy saving aspects In order to evaluate the thermal impact of PCC with MPCM measurements of indoor surface temperature as well as calculation of power consumption and power reduction were carried out. 30 PCC PCC20 C) 28 PCC40 26 24 22 Indoor surface temperature ( temperature surface Indoor 20 18 0 12 24 36 48 60 72 Time (h) Figure 8.8. Comparison of the indoor surface temperature of PCC without MPCM and with 20 and 40% of MPCM. Figure 8.8 presents the comparison of the indoor surface temperature of PCC without MPCM and with 20 and 40% of MPCM. Experimental data show that the variation of the indoor surface temperature of PCC containing MPCM is smaller than that of PCC without MPCM. The lower variation can be explained by higher heat storage capacity and lower thermal conductivity of concrete with incorporated MPCM. Therefore, the addition of thermoregulating microcapsules to concrete minimize the effect of the outdoor temperature on the indoor temperature. This can help in reducing the energy consumption for heating and cooling. The heat flux on the indoor side of concrete samples were collected and they are shown in Figure 8.9. The transition zones can be easily observed, and they are highlighted on the figure. 236 Incorporation of microcapsules into concrete 60 PCC ) PCC20 -2 PCC40 40 20 0 -20 Indoor surface heat flux (W m flux heat surface Indoor Transition zone -40 0 12 24 36 48 60 72 Time (h) Figure 8.9. Comparison of measurements for the indoor surface heat flux variations with time of PCC without MPCM and with 20 and 40% of MPCM. In order to estimate the heat gain/loss toward the indoor environment, the total heat transfer at the indoor surface can be used. The heat gain/loss must be compensated by heating or cooling to provide a constant indoor temperature. Therefore, it can be considered as the energy consumption of the heating/cooling system. The total energy consumption of the heating and cooling system is the sum of the power consumed for heating when the indoor surface temperature is lower than assumed room temperature and power consumed for cooling when the indoor surface temperature is higher than assumed room temperature. Figure 8.10 presents the total calculated power consumption for the heating/cooling system to keep the indoor temperature at 23 ºC and the power reduction, both as a function of MPCM concentration. In Figure 8.10a can be observed that the power consumption for the heating/cooling system to keep the indoor temperature at 23 ºC decreases when the MPCM concentration increases. The experimental data shown in Figure 8.10b illustrates that the system can save up to 6.5 and 20.9% of power consumption after adding 20 and 40% of microcapsules, respectively. This confirm the promising impact of concrete with MPCM on the thermal energy storage and also reduction of CO2 emission. The reduction of power consumption is related to the higher 237 Chapter 8 heat storage capacity and the better insulation properties of concrete after adding microcapsules [35]. 1.0 25 0.9 (a) (b) ) -2 0.8 20 0.7 15 0.6 0.5 10 0.4 0.3 Power reduction (%) reduction Power 5 0.2 Power consumption m (kWh Power 0.1 0 0.0 PCC PCC20 PCC40 PCC PCC20 PCC40 Figure 8.10. The power consumption (a) and the power reduction (b) of concrete as a function of microcapsule concentration. Fire reaction properties In order to analyze the influence of MPCM on flammability of PCC, a cone calorimeter tests according to ISO 5660-1:2002 were performed for PCC without MPCM, 10% MC-P(St-DVB)50 and 10% MC-P(St-DVB-10%PNC-HEMA)50. Measurements were repeated 3 times for each type of PCC sample. Nevertheless, none of the specimens were ignited during the test. Within 15 min, higher amounts of smoke release could be observed for PCC with microcapsules. When test was stopped after 1 hour, mass losses of 6.08, 7.22 and 7.04% for PCC, PCC with 10% of MC-P(St-DVB)50 and PCC with 10% of MC-P(St-DVB-10%PNC-HEMA)50, respectively were observed. These weight losses correspond mainly to water evaporation. Some cracks in PCC with 10% MC-P(St- DVB)50 could be observed because of the thermal gradient the specimens were subjected to [36]. Figure 8.11 shows photographs of the specimens before and after the flammability test. Accordingly, with addition of 10% MPCM to PCC any significant influence on the fire properties was observed. 238 Incorporation of microcapsules into concrete (a) (b) (c) (d) Crack (f) (e) Figure 8.11. Surface of specimens from composition (a) PCC before testing, (b) PCC after testing, (c) PCC 10% MPCM before test, (d) PCC 10% MPCM after test, (e) PCC 10% MPCM- FR before test, (f) PCC 10% MPCM-FR after test. 239 Chapter 8 8.4. Conclusions In this chapter, the effect of microencapsulated phase change material (MPCM) at temperatures below and above the melting point of PCM on the physical, mechanical and thermal properties of Portland cement concrete (PCC) was investigated. It was found that the compressive strength of PCC decreased with increasing amount of MPCM at both 20 and 40 °C. Uniform dispersion of MPCM in PCC was observed by X-ray tomography. SEM images revealed air voids between the MPCM and the concrete matrix, which will contribute to the strength reduction. The strength reduction was similar at both curing temperature, which proves that the mechanical properties are not affected by the solid or liquid state of PCM. Although, MPCM has significant influence on the compressive strength of concrete, the obtained values are still sufficiently high for structural applications (i.e. between 20 and 40 MPa). On the other hand, incorporation of MPCM into concrete reduced the thermal conductivity and improved the thermal storage capacity of PCC. This enhances the energy efficiency of the building envelope. The thermal conductivity was reduced by 20% when adding 20% of MPCM to PCC. The specific heat capacity did not vary significantly with the curing temperature and increased from 939 to 990 J Kg-1 °C-1 for PCC without MPCM and with 20% of MPCM, respectively. In addition, it was possible to reduce the energy consumption by 6.5 and 21% by addition of 20 and 40% of microcapsules to concrete, respectively. In spite of the high flammability of paraffin, 10% of MPCM in PCC did not present any significant fire risk, since the concrete samples were not ignited in the cone calorimetry test. Only small cracks were observed when 10% of MPCM without fire retardant properties were used. 240 Incorporation of microcapsules into concrete References 1. Kurdowski W (2014) Cement and concrete chemistry. Springer Science & Business, 2. Navarro L, de Gracia A, Niall D, Castell A, Browne M, McCormack SJ, Griffiths P, Cabeza LF (2016) Thermal energy storage in building integrated thermal systems: A review. Part 2. Integration as passive system. Renewable Energy 85 (Supplement C):1334-1356. doi:https://doi.org/10.1016/j.renene.2015.06.064 3. Demirbas MF (2006) Thermal Energy Storage and Phase Change Materials: An Overview. Energy Sources, Part B: Economics, Planning, and Policy 1:85-95 4. Gracia Ad, Cabeza LF (2015) Phase change materials and thermal energy storage for buildings. Energy and Buildings 103:414-419 5. Sharma A, Tyagi VV, Chen CR, Buddhi D (2009) Review on thermal energy storage with phase change materials and applications. Renewable and Sustainable Energy Reviews 13:318-345 6. A.Jamekhorshid, S.M.Sadrameli, M.Farid (2014) A review of microencapsulation methods of phase change materials (PCMs) as a thermal energy storage (TES) medium. Renewable and Sustainable Energy Reviews 31:531-542 7. Tyagi VV, Kaushik SC, Tyagi SK, Akiyama T (2011) Development of phase change materials based microencapsulated technology for buildings: A review. Renewable and Sustainable Energy Reviews 15 (2):1373-1391. doi:https://doi.org/10.1016/j.rser.2010.10.006 8. Buscombe A, Wu M (2013) A Review of PCM’s Thermal Performance Within Lightweight Construction. International Journal of Engineering Practical Research (IJEPR):174-177 9. Wang X, Niu J, Yi Li XW, Chen B, Zeng R, Song Q, Zhang Y (2007) Flow and heat transfer behaviors of phase change material slurries in a horizontal circular tube. International Journal of Heat and Mass Transfer 50:2480-2491 10. Chen L, Wang T, Zhao Y, Zhang X-R (2014) Characterization of thermal and hydrodynamic properties for microencapsulated phase change slurry (MPCS). Energy Conversion and Management 79:317-333 11. Zhang GH, Zhao CY (2011) Thermal and rheological properties of microencapsulated phase change materials. Renewable Energy 36:2959-2966 241 Chapter 8 12. Alvarado JL, Marsh C, Sohn C, Phetteplace G, Newell T (2007) Thermal performance of microencapsulated phase change material slurry in turbulent flow under constant heat flux. International Journal of Heat and Mass Transfer 50:1938-1952 13. Yamagishi Y, Takeuchi H, Pyatenko AT, Kayukawa N (1999) Characteristics of Microencapsulated PCM Slurry as a Heat-Transfer Fluid. AIChE 45 (4):696-707 14. Pollerberg C, Dotsch C (2006) Phase Changing Slurries in cooling and cold supply networks. Paper presented at the 10th International Symposium on Disctrict Heating and Cooling, Hannover (Germany), 15. Evola G, Marletta L, Sicurella F (2013) A methodology for investigating the effectiveness of PCM wallboards for summer thermal comfort in buildings. Building and Environment 59:517-527. doi:https://doi.org/10.1016/j.buildenv.2012.09.021 16. Cabeza LF, Castellón C, Nogués M, Medrano M, Leppers R, Zubillaga O (2007) Use of microencapsulated PCM in concrete walls for energy savings. Energy and Buildings 39 (2):113-119. doi:https://doi.org/10.1016/j.enbuild.2006.03.030 17. Norvell C, Sailor DJ, Dusicka P (2013) The effect of microencapsulated phase-change material on the compressive strength of structural concrete. Journal of Green Building 8 (3):116-124. doi:10.3992/jgb.8.3.116 18. Figueiredo A, Lapa J, Vicente R, Cardoso C (2016) Mechanical and thermal characterization of concrete with incorporation of microencapsulated PCM for applications in thermally activated slabs. Construction and Building Materials 112:639- 647. doi:https://doi.org/10.1016/j.conbuildmat.2016.02.225 19. Hunger M, Entrop AG, Mandilaras I, Brouwers HJH, Founti M (2009) The behavior of self-compacting concrete containing micro-encapsulated Phase Change Materials. Cement and Concrete Composites 31 (10):731-743. doi:https://doi.org/10.1016/j.cemconcomp.2009.08.002 20. Cui H, Liao W, Mi X, Lo TY, Chen D (2015) Study on functional and mechanical properties of cement mortar with graphite-modified microencapsulated phase-change materials. Energy and Buildings 105:273-284. doi:https://doi.org/10.1016/j.enbuild.2015.07.043 21. Eddhahak-Ouni A, Drissi S, Colin J, Neji J, Care S (2014) Experimental and multi- scale analysis of the thermal properties of Portland cement concretes embedded with 242 Incorporation of microcapsules into concrete microencapsulated Phase Change Materials (PCMs). Applied Thermal Engineering 64 (1):32-39. doi:https://doi.org/10.1016/j.applthermaleng.2013.11.050 22. Shadnia R, Zhang L, Li P (2015) Experimental study of geopolymer mortar with incorporated PCM. Construction and Building Materials 84:95-102. doi:https://doi.org/10.1016/j.conbuildmat.2015.03.066 23. Pania M, Yunping X Effect of Phase-Change Materials on Properties of Concrete. Materials Journal 109 (1). doi:10.14359/51683572 24. del Hormigón Estructural I (2008) EHE-08. Madrid, Ministerio de Fomento, Secretaría General Técnica 25. Lecompte T, Le Bideau P, Glouannec P, Nortershauser D, Le Masson S (2015) Mechanical and thermo-physical behaviour of concretes and mortars containing phase change material. Energy and Buildings 94 (Supplement C):52-60. doi:https://doi.org/10.1016/j.enbuild.2015.02.044 26. Allahverdi A, Pilehvar S, Mahinroosta M (2016) Influence of curing conditions on the mechanical and physical properties of chemically-activated phosphorous slag cement. Powder Technology 288:132-139. doi:https://doi.org/10.1016/j.powtec.2015.10.053 27. Lura P, van Breugel K, Maruyama I (2001) Effect of curing temperature and type of cement on early-age shrinkage of high-performance concrete. Cement and Concrete Research 31 (12):1867-1872. doi:https://doi.org/10.1016/S0008-8846(01)00601-9 28. Aguayo M, Das S, Maroli A, Kabay N, Mertens JCE, Rajan SD, Sant G, Chawla N, Neithalath N (2016) The influence of microencapsulated phase change material (PCM) characteristics on the microstructure and strength of cementitious composites: Experiments and finite element simulations. Cement and Concrete Composites 73:29-41. doi:https://doi.org/10.1016/j.cemconcomp.2016.06.018 29. Borreguero AM, Garrido I, Valverdea JL, Rodríguez JF, Carmona M (2014) Development of smart gypsum composites by incorporatingthermoregulating microcapsules. Energy and Buildings 76:631-639 30. Yamamoto O, Kambe H (1971) Thermal Conductivity of Cross-linked Polymers. A Comparison between Measured and Calculated Thermal Conductivities. Polymer Journal 2 (5):623-628 243 Chapter 8 31. Cui H, Liao W, Mi X, Lo TY, Chen D (2015) Study on functional and mechanical properties of cement mortar withgraphite-modified microencapsulated phase-change materials. Energy and Buildings 105:273-284 32. N. Ukrainczyk SK, and J. Šipušiæ (2010) Thermophysical Comparison of FiveCommercial ParaffinWaxes as Latent Heat StorageMaterials. Chem Biochem Eng Q 24:129-137 33. Joulin A, Zalewski L, Lassue S, Naji H (2014) Experimental investigation of thermal characteristics of a mortar with or without a micro-encapsulated phase change material. Applied Thermal Engineering 66 (1):171-180. doi:https://doi.org/10.1016/j.applthermaleng.2014.01.027 34. Joulin A, Zalewski L, Lassue S, Naji H (2014) Experimental investigation of thermal characteristics of a mortar with or without a micro-encapsulated phase change material. Applied Thermal Engineering 66:171-180 35. M. Hunger AGE, I. Mandilaras, H.J.H. Brouwers, M. Founti (2009) The behavior of self-compacting concrete containing micro-encapsulated Phase Change Materials. Cement & Concrete Composites 31:731-743 36. Correia J, Marques A, Pereira C, De Brito J (2012) Fire reaction properties of concrete made with recycled rubber aggregate. Fire and Materials 36 (2):139-152 244 9. Conclusions and future work CHAPTER 9 CONCLUSIONS AND FUTURE WORK Chapter 9 9.1. Conclusions An improved class of microcapsules to be applied in concrete applications has been developed. The introduction into the recipe of highly level of crosslinking through the addition of divinylbenzene as crosslinking agent allows to obtain particles with enhanced physical resistance and thermal stability. The incorporation of a flame retardant co-monomer (PNC-HEMA) into the polymer network not only improve the fire and thermal resistance of the particle but also produces a better-defined core-shell structure of microcapsules. The incorporation of microcapsules into concrete reduces the thermal conductivity and provides thermal storage capacity with a low impact on its mechanical resistance. From the present study of the development of microcapsules with thermal energy storage capacity for concrete applications, the following specific conclusions can be drawn: . Suspension-like polymerization is a feasible technique for the encapsulation of saturated phase change materials within P(St-DVB) shells. Its implementation in semi-industrial scale gave similar results to the lab scale. . PVP at the concentration of 5.03% was chosen as the most suitable suspending agent. This parameter should be combined with an agitation rate of 800 rpm, a mass ratio between the continuous/discontinuous phases (3.24), toluene/PCM mass ratio of 1.92 and a PCM percentage of 68.5% with respect to the total monomers and paraffin amount to operate at optimal synthesis conditions. . The type and amount of suspending agent have strong influence on the particle size and morphology of microcapsules but not on the internal structure. The properties of the microcapsules depend on the chemical characteristic of the phase change materials and polymer. In order to modify the microcapsules from matrix to core- shell structure, a polymer and PCM with higher polarity difference should be used. . The incorporation of a surfactant agent into the microcapsules was studied. The PVP distribution data were fitted by the Langmuir model obtaining a value of the 246 Conclusions and future work maximum retention capacity and the equilibrium constant as 192.9 g kg-1 and 0.18 m3 kg-1, respectively. . Free radical polymerization for the encapsulation of unsaturated fatty acids (as TES material) produces the co-polymerization of them with the monomers intended to form the shell. As a consequence of this side polymerization, a decrease of the thermal capacity in the microcapsules with unsaturated fatty acids with respect to those with saturated is observed. . Incorporation of a flame retardant monomer (PNC-HEMA) into the polymer shell enhanced the thermal and fire resistance of the microcapsules, resulting in lower total heat released and higher amount of char produced. Furthermore, the PCM content was higher due to core-shell structure of microcapsules. Microcapsules with 10 wt.% of PNC-HEMA were the optimal concentration for obtaining microcapsules with a spherical morphology and the highest TES capacity. . Microcapsules with a crosslinking agent revealed dimensional stability at high temperatures. In spite of exhibiting an intact the polymer structure, zero enthalpy was observed at 180 °C, because the PCM had been completely released by vaporization. The PCM release from the microcapsules was mathematical modeled on the basis of an internal diffusion model. . It was found that the compressive strength of Portland cement concrete decreased with increasing amounts of incorporated microcapsules. However, it was still within the acceptable level for structural concrete (higher than 20 MPa). Incorporation of microcapsules into concrete reduced the thermal conductivity and improved the thermal storage capacity of PCC. This enhance the energy efficiency of the building envelope. 247 Chapter 9 9.2. Future work The following recommendations are interesting for future research: . Improving the recipe for the synthesis of microcapsules with fire retardant properties in order to obtain a smooth surface and a continues shell. . Completing the study of the physical and mechanical properties of the microcapsules with different polymeric shells by nanoindentation. . Developing nanosized microcapsules with enhanced TES capacity and lower impact into the mechanical properties of concrete. . Optimizing the pilot plant operational conditions based on economic criteria and market trends. . Extending the study about the microcapsules incorporation on the properties of other building materials, such as the geopolymer concrete, mortar, gypsum, and bricks. . Performing an economical evaluation and life cycle assessment of the production of microcapsules in industrial scale and incorporation to the concrete. 248 Nomenclature Nomenclature Abbreviations AA Acrylic acid AG Arabic gum AIBN Azobisisobutyronitrile AMA Allyl methacrylate APP Ammonium polyphosphate ATR Attenuated Total Reflectance BDDA 1,4-Butylene glycol diacrylate BMA n-Butyl methacrylate BPO Benzoyl peroxide BSED Backscattered Electron Detector CA Caprylic acid DA Decanoic acid DSC Differential Scanning Calorimetry DVB Divinylbenzene EA Erucic acid EDAX Energy dispersive x-ray spectrometer EG Expanded graphite FS Form stable materials FT-IR Fourier transform infrared spectroscopy GC-MS Gas chromatrography with mass spectroscopy GPC Gel Permeation Chromatography HDPE High density polyethylene HEMA Monomer 2-hydroxyethyl methacrylate HRR Heat release rate HVAC Heating, ventilation and air conditioning IFR Intumescent flame retardant system LALLS Low Angel Laser Light Scattering LauA Lauric acid LFD Large Field Detector LHTES Latent heat Thermal energy storage LinA Linoleic acid MA Myristic acid MAA Methacrylic acid MCC Micro-combustion calorimetry MMA Methyl methacrylate MMT Treated montmorillonite MPCM Microencapsulated phase change material MPCM-FR Microcapsules with flame retardant NGD Neopentyl glycol diacrylate Nomenclature OA Oleic acid OMMT Organomontmorillonite PA Palmitic acid PCC Portland cement concrete PCM Phase change material PEDMA Poly(ethylene dimethacrylate) PER Pentaerythritol PETRA Pentaerythritol tetraacrylate PMMA Poly(methyl methacrylate) PNC Phosphonitrile chloride trimer PNC-HEMA Hexa(methacryloylethylenedioxy)cyclotriphosphazene PSD Particle size distribution PSt Polystyrene PVP Polyvinylpyrrolidone RPUF Rigid polyurethane foams RT21 Rubitherm®RT21 RT27 Rubitherm®RT27 RT31 Rubitherm®RT31 RT6 Rubitherm®RT6 SA Stearic acid SBS Styrene–butadiene–styrene copolymer SDS Sodium dodecylsulfate SEM Scanning Electron Microscope SHTES Sensible heat thermal energy storage SMA Sodium salt of styrene-maleic anhydride polymer SSPCM Shape stabilized phase change materials St Styrene TBHP tert-Butyl hydroperoxide TES Thermal energy storage TEVS Triethoxyvinylsilane TGA Thermogravimetric Analysis THF Tetrahydrofuran TMPTA Trimethylolpropanetriacrylate XRD X-ray micro-tomography Symbols ar Extent of conversion Ar Area of microcapsules c Overall concentration cA Concentration of A Cb Concentration of bulk solution 250 Nomenclature Cp Specific heat of the material CPCM PCM content DAB Diffusion coefficient dm/dt Mass loss rate dn/dt Molar loss rate Ea Activation energy EE Encapsulation efficiency f Fraction of suspending agent g Gravitational constant h0 Liquid level K Equilibrium constant of the system kd Rate of decomposition kL Thermal conductivity of liquid kS Thermal conductivity of solid m Mass of material MA Molecular weight NA,z Molar flux P Power P(St-DVB)MC Mass of polymer 0 PA Vapor pressure PCMfeed Weights of the PCM fed to the reactor PCMMC Mass of PCM in microcapsules PMC Product mass Pr Power reduction PT Total pressure Q Quantity of stored heat R Universal gas constant R0 Radius r0 Radius of the microcapsules S Surface area SA Total amount of the suspending agent (St-DVB)feed Weights of the monomers fed to the reactor T Temperature t1/2 Half-life time V Total volume of the bulk solution Vpore Volume of pores X Conversion of monomers yA Molar fraction z Vapor-liquid interphase height 251 Nomenclature ΔHMC Enthalpy of microcapsules ΔHPCM Enthalpy of pure PCM Δhr Endothermic heat of reaction α Evaporation coefficient β Shape factor Г Retention capacity Гmax Maximum retention capacity γop Oil/polymer interfacial tension γow Water/oil interfacial tension γp Surface free energy γpw Water/polymer interfacial tension Σν Sum of atomic diffusion volumes ηr Microcapsules yield Weighted sum of squared residuals Contact angle ρ Density φave Average heat fluxes φindoor Heat flux on the indoor side 252 Publications and conferences ANNEX I PUBLICATIONS AND CONFERENCES Annex I Publications 1. A.M. Szczotok, I. Garrido, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez, Predicting microcapsules morphology and encapsulation efficiency by combining the spreading coefficient theory and surface polarity, Colloid and Surfaces A: Physicochemical and Engineering Aspects (Submitted). 2. A.M. Szczotok, A.-L. Kjøniksen, J.F. Rodriguez, M. Carmona, Estimation of the vapor pressure and enthalpy of vaporization of phase change materials by thermogravimetric analyses, Journal of… (Draft prepared). 3. A.M. Szczotok, D. Madsen, A. Serrano, P. van Hees, A.-L. Kjøniksen, J.F. Rodriguez, M. Carmona, Fire properties of newly developed microcapsules for thermal energy storage, Journal of… (Draft prepared). 4. V.D. Cao, S. Pilehvar, C. Salas-Bringas, A.M. Szczotok, L. Valentini, M. Carmona, J.F. Rodriguez, A.-L. Kjøniksen, Influence of microcapsule size and shell polarity on thermal and mechanical properties of thermoregulating geopolymer concrete for passive building applications, Energy Conversion and Management 164:198-209. 5. A.M. Szczotok, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez, The role of radical polymerization in the production of thermoregulating microcapsules or polymers from saturated and unsaturated fatty acids, Journal of Applied Polymer Science, 135 (2018) 45970-n/a. 6. A.M. Szczotok, M. Carmona, A. Serrano, A.L. Kjøniksen, J.F. Rodriguez, Development of thermoregulating microcapsules with cyclotriphosphazene as a flame retardant agent, IOP Conference Series: Materials Science and Engineering, 251 (2017) 012120. 7. S. Pilehvar, V.D. Cao, A.M. Szczotok, L. Valentini, D. Salvioni, M. Magistri, R. Pamies, A.-L. Kjøniksen, Mechanical properties and microscale changes of geopolymer concrete and Portland cement concrete containing micro-encapsulated phase change materials, Cement and Concrete Research, 100 (2017) 341-349. 8. V.D. Cao, S. Pilehvar, C. Salas-Bringas, A.M. Szczotok, J.F. Rodriguez, M. Carmona, N. Al-Manasir, A.-L. Kjøniksen, Microencapsulated phase change materials for enhancing the thermal performance of Portland cement concrete and geopolymer 254 Publications and conferences concrete for passive building applications, Energy Conversion and Management, 133 (2017) 56-66. 9. A.M. Szczotok, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez, Equilibrium adsorption of polyvinylpyrrolidone and its role on thermoregulating microcapsules synthesis process, Colloid and Polymer Science, 295 (2017) 783-792. 10. A.M. Szczotok, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez Using fatty acids for developing thermal energy storage materials. 11th Conference on Advanced Building Skins 2016, Conference Proceedings. 255 Annex I Conferences 1. A.M. Szczotok, A. Serrano, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez Development of thermoregulating microcapsules having flame retardant properties for building applications. 3rd International Conference on “Innovative Materials, Structures and Technologies” IMST 2017. Riga, Latvia (2017). Oral presentation. 2. V.D. Cao, S. Pilehvar, C. Salas-Bringas, A.M. Szczotok, A.-L. Kjøniksen, Smart Geopolymer Concrete Containing Microencapsulated Phase Change Materials for Passive Building Applications, 3rd International Conference on “Innovative Materials, Structures and Technologies” IMST 2017. Riga, Latvia (2017). Poster. 3. V.D. Cao, S. Pilehvar, C. Salas-Bringas, A.M. Szczotok, A.-L. Kjøniksen, The Effect of Microcapsulated Phase Change Materials on Thermal and Mechanical properties of Geopolymer Concrete, 12th Conference on Advanced Building Skins. Bern, Switzerland (2017). Oral presentation. 4. S. Pilehvar, A.-L. Kjøniksen, A.M. Szczotok, R. Pamies, The Effect of Micro- encapsulated Phase Change Materials on the Properties of Geopolymer and Portland Cement Concrete. 3rd International Conference on “Innovative Materials, Structures and Technologies” IMST 2017. Riga, Latvia (2017). Poster. 5. S. Sanfèlix, I. Santacruz, A.M. Szczotok, L.M. Ordonez, A. De la Torre, A.-L. Kjøniksen, Rheological performance of cement composites with Microencapsulated Phase Change Materials. 3rd International Conference „Innovative Materials, Structures and Technologies” IMST 2017. Riga, Latvia (2017). Poster. 6. A.M. Szczotok, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez The effect of suspending agent type on thermal physical properties of thermoregulating microcapsules. E2KW 2016 Energy and Environment Knowledge Week. Paris, France (2016). Oral presentation. 7. A.M. Szczotok, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez Using fatty acids for money and energy savings. VI Jornadas Doctorales de la Universidad de Castilla-La Mancha. Toledo, Spain (2016). Poster. 256 Publications and conferences 8. A.M. Szczotok, M. Carmona, A.-L. Kjøniksen, J.F. Rodriguez Using fatty acids for developing thermal energy storage materials. 11th Conference on Advanced Building Skins. Bern, Switzerland (2016). Oral presentation. 9. A. Serrano, J. C. de Haro, J. A. Martín del Campo, I. Izarra, A.M. Szczotok, C. Gutiérrez, M. Carmona, The use of prototypes to avoid chemophobia in society. 6th EuCheMS Chemistry Congress. Seville, Spain (2016). Oral presentation. 257