Applying an Experimental Design to Improve the Characteristics of

(6) prepared microcapsules containing eicosane with 0.1−10 μm for ... polystyrene-co-methylmethacrylate,(14) and polyurethane(15) microspheres have...
0 downloads 0 Views 2MB Size
Ind. Eng. Chem. Res. 2008, 47, 9783–9790

9783

Applying an Experimental Design to Improve the Characteristics of Microcapsules Containing Phase Change Materials for Fabric Uses Luz Sa´nchez, Engracia Lacasa, Manuel Carmona, Juan F. Rodrı´guez, and Paula Sa´nchez* Department of Chemical Engineering, UniVersity of CastillasLa Mancha, AVda. Camilo Jose´ Cela s/n 13004 Ciudad Real, Spain

Microcapsules with a high amount of PRS paraffin wax encapsulated and narrow size distribution were prepared by a Shirasu porous glass (SPG) emulsification technique and a subsequent suspensionlike polymerization process. An experimental design approach, based on a central composite design, was used to determine quantitatively the effect of RepSol-YPF Paraffin (PRS) paraffin wax/styrene mass ratio (PRS/St), percentage of polyvinylpyrrolidone/styrene mass ratio (% PVP/St), and water/styrene mass ratio (H2O/St) on the phase change material (PCM) microcapsules properties. The % PVP/St mass ratio was the most important parameter affecting the particle size distribution. The thermal energy storage of microcapsules increases with the PRS/ St mass ratio used. The following synthesis conditions, mass ratios PRS/St of 1.02, % PVP/St of 9.43, and H2O/St of 8.23, allowed the the proper main particle size in number (4.80 µm) for fabrics applications containing the maximum phase change material encapsulated with a latent heat of 102.42 J/g to be achieved. 1. Introduction Phase change materials (PCMs) are a series of functional materials with storing and realizing energy properties when experiencing a phase transition. The PCM material confinement by microencapsulation facilitates their incorporation into a wide variety of applications such as in fibers, fabrics, coatings, physiotherapy devices, insulation panels, and walls.1 The textile industry has been slow to react to the possibilities of microencapsulation, although in the early 1980s phase change materials were used by the US National Aeronautics and Space Administration (NASA) with the aim of managing the thermal barrier properties of space suits.2 The number of commercial applications of microencapsulation of PCMs in the textile industry continues to grow into textiles with new properties and added value, for instance, medical textiles and technical textiles. Textiles containing PCMs help to counteract cold and overheating; in general, this effect can be described as thermoregulation. For use in textile materials, an appropriate particle size should range from 0.5 to 100 µm.3 Colvin and Bryant4 used microcapsules containing PCMs of 30-100 µm for textile fibers, composites, and foams. Pause5 made PCM microcapsules of 1-60 µm for nonwoven protective garments with thermoregulating properties. Shin et al.6 prepared microcapsules containing eicosane with 0.1-10 µm for development of thermoregulating textile materials. For addition of microcapsules to textile materials, the particle size and size uniformity are critical parameters.7 Paraffin waxes compared to other PCMs have high heatstorage capacities (145 and 240 J/g), are easily available, and not expensive.8,9 Shin et al.6 prepared microcapsules with a heat storage capacity of 134.3 J/g for thermoregulating textile materials. In a previous work, PRS paraffin wax was successfully encapsulated by a polymer cover (polystyrene) by means of a suspensionlike polymerization technique.10 In this study, an experimental setup was designed in order to obtain microcapsules with narrow size distribution by means of a microporous membrane. By using this method, monodis* To whom correspondence should be addressed. Phone: +34 926 295300 ext. 3418. Fax: +34 926 295242. E-mail: Paula.Sanchez@ uclm.es.

persed polystyrene,11 polylactide,12 polyimide,13 polystyreneco-methylmethacrylate,14 and polyurethane15 microspheres have been successfully prepared. A 23 factorial design was initially performed to study the relative importance of three synthesis variables: PRS paraffin wax/styrene mass ratio (PRS/St), percentage of polyvinylpyrrolidone/styrene mass ratio (% PVP/St), and water/styrene mass ratio (H2O/St) in the particle size and latent heat of microcapsules. According to Palasota16 and Barnabas,17 this factorial design was followed with a central composite design to demonstrate that the recipe can be optimized to obtain microcapsules appropriate to be incorporated into textile. The PRS/St mass ratio was studied in order to improve the thermal storage/release capacity of the PCM microcapsules. The studied interval of this variable ranged from 0.25 to 1.11 based on previous studies carried out by the authors.18 On the other hand, % PVP/St mass ratio and H2O/St mass ratio were selected due to have an important influence on particles size and size distribution.19,20 In the literature, the use of a large amount of suspension agent to produce small and uniform polymeric particles has been widely reported.19-21 H2O/St mass ratio of 10:1 has been used in the literature to decrease the droplet collision frequency.22 Keeping this in mind and with the aim of obtaining a minimum particle size, both % PVP/St and H2O/ St mass ratios ranging from 4.09 to 11.31 were studied. The aim of this work was to obtain the proper conditions for the synthesis that allows PCM microcapsules with the minimum main particle size in number and the maximum thermal storage capacity to be created. For this, an empirical equation relating the synthesis variables of a given experiment (PRS/St mass ratio, % PVP/St mass ratio, and H2O/St mass ratio) with the heat capacity of microcapsules and the particle size distribution (PSD) was proposed. Experimental design methodology was used and the experimental results of the central composite design (CCD) were fitted with a second-order polynomial equation. 2. Experimental Details 2.1. Materials. Styrene (99 wt %) was of reagent grade (Merck Chemical). Styrene was washed with sodium hydroxide to remove the inhibitor and calcium chloride as desiccant.

10.1021/ie801107e CCC: $40.75  2008 American Chemical Society Published on Web 11/08/2008

9784 Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008

Figure 1. Schematic diagram of the microencapsulation system. Table 1. Factor Levels and Experimental Design variable

low (-)

high (+)

center (0)

axial (-R)

axial (+R)

PRS/St % PVP/St H2O/St

0.36 5.03 5.03

1.00 10.37 10.37

0.68 7.70 7.70

0.25 4.09 4.09

1.11 11.31 11.31

experiment central 23 factorial design + composite 3 central points design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

PRS/ % St PVP/St

H2O/ St

0.36 0.68 1.00 0.68 0.36 0.25 0.68 1.00 0.68 0.68 0.68 1.00 0.68 0.36 1.11 1.00 0.36

10.37 11.31 5.03 7.70 10.37 7.70 7.70 5.03 7.70 4.09 7.70 10.37 7.70 5.03 7.70 10.37 5.03

10.37 7.70 5.03 4.09 5.03 7.70 7.70 10.37 7.70 7.70 7.70 5.03 11.31 10.37 7.70 10.37 5.03

Benzoyl peroxide (97 wt %) was used as initiator (Fluka Chemical). PRS paraffin wax (Mw 478 g/mol) was of commercial grade (Repsol YPF) and was used as core material. Polyvinylpyrrolidone (K30, Mw 40 000 g/mol) of reagent grade (Fluka Chemical) was used as stabilizer and methanol to pour the samples. All these reagents were used as received. Water was purified by distillation followed by deionization using ionexchange resins. Nitrogen was of high-purity grade (99.999%). A tubular type Shirasu porous glass (SPG) membrane with pores sizes of 5.5 µm was used to produce a narrow microcapsules size distribution. At the end of every experiment, the used membrane has been retrieved by a treatment with sodium dioctyl sulfosuccinate (AOT) (Fluka Chemical, 96 wt %), ethanol (Panreac Chemical), and chloride acid (Prolabo Chemical) reagents. 2.2. Microcapsule Synthesis. Suspensionlike polymerization reactions were performed in a 1-L double-jacketed glass reactor equipped with digital control of stirring rate and temperature, a reflux condenser, and a nitrogen gas inlet tube. The experimental setup is shown schematically in Figure 1.

Figure 2. Second-order polynomial fitting. (a) Average particle size in number (µm). (b) Latent heat of microcapsules (J/g).

The synthesis process involves two phases: a continuous phase containing water and polyvinylpyrrolidone and a discontinuous phase containing styrene, PRS paraffin wax, and benzoyl peroxide. The continuous phase was transferred to a glass reactor with mild agitation during 10 min. After that, continuous phase was circulated during 3 min through the SPG membrane because the hydrophilic membrane used must be prewetted with the aqueous phase. As the initiator was already mixed with

Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008 9785 Table 2. Mathematic Treatment of the Experimental Data Obtained by Applying a Second-Order Polynomial Fitting second-order polynomial fitting equation ) β0 + β1(PRS/St) + β2(% PVP/St) + β3(H2O/St) + β11(PRS/ St)2 + β22(% PVP/St)2 + β33(H2O/St)2 + β12(PRS/St)(% PVP/ St) + β13(PRS/St)(H2O/St) + β23(% PVP/St)(H2O/St) parameters

dpn0.5 (µm)

latent heat of fusion (J/g)

β0 β1 β2 β3 β11 β22 β33 β12 β13 β23

-8.654 19.000 -1.481 2.933 1.619 0.229 0.084 -0.715 -1.514 -0.312

-77.053 80.024 7.126 25.558 -60.534 0.035 -2.055 -5.072 11.060 -0.318

monomer and paraffin wax, these products were added into dispersion phase storage tank. Next, the discontinuous phase was permeated by nitrogen gas through the uniform pores of the SPG membrane into the continuous phase to form uniform droplets in situ at the membrane/continuous phase interface. In order to ensure a regular droplet detachment from the pore outlets, shear stress is generated at the membrane/continuous phase interface by recirculating the continuous phase during 10 min using a low shear pump. The rate of mixing should be high enough to provide the required tangential shear on the membrane surface, but not too excessive to induce further droplet breakup.23 After every experiment, the used membrane was recovered by means of the following procedure. First of all, it was soaked in 2 wt % solution of AOT in ethanol for 2 h. After that, it was rinsed and soaked in pure water. Next, it was immersed for 1 h in water solution of 2 M HCl at 70 °C, and finally, it was rinsed and soaked in pure water for 30 min.24 This procedure fully recovers the initial hydrophilic state of the SPG membrane, which can be used in subsequent experiments. The suspensionlike polymerization reactions were maintained under vigorous agitation (1600 rpm), the reaction media was bubbled with nitrogen, and the temperature was set at 110 °C in the thermostat bath. Temperature and agitation were maintained in a fixed value during the experiments. The polymerization process was carried out for 6 h under a nitrogen atmosphere. Next to the polymerization process, 200 g of methanol were added into the glass reactor by mild agitation when its temperature was 70 °C. Methanol is used to improve the decantation procedure. Finally, the polymerized microcapsules were filtrated by pressure to remove paraffin wax that was not encapsulated and latex particles. The purified microcapsules were dried at room temperature. 2.3. Environmental Scan Electron Microscopy (ESEM) Observation. The surface features of the microcapsules after polymerization were observed by using a XL30 (LFD) ESEM. 2.4. Calculation of Number-Average Diameter and VolumeAverage Diameter. Particle size and particle size distribution were determined on a Malvern Mastersizer Hydro 2000 SM light scattering apparatus in a diluted dispersion of the particles in methanol. 2.5. Differential Scanning Calorimetry (DSC). Measurements of melting point and melting heats of different materials employed and obtained were performed in a differential scanning calorimeter model DSC Q100 of TA Instruments equipped with a refrigerated cooling system and nitrogen as the purge gas.

These measurements were done varying the temperature in the range from -30 to 80 °C with a heating rate of 10 °C/min. Each sample was analyzed at least twice and the average value was recorded. Furthermore, the content of paraffin wax in the microcapsule can be estimated according to the measured melting enthalpy of different standards mixtures of polystyrene and paraffin wax. The experimental values obtained were fitted to a straight line with the following form: % paraffin wax content ) (DeltaHm-7.2004)/1.921

(1)

where DeltaHm is the enthalpy of analyzed microcapsules (J/g). 2.6. Experimental Design Approach. The factor design requires fewer measurements than the classical one-at-time experiment to give the same precision. At the same time, it detects and estimates any interaction between factors, which the classical experiment cannot do. According to Deming and Palasota,16 for chemical systems that have more than one factor, the application of properly designed experiments and adequate multifactor models, such as a central composite design, is necessary. A central composite design consists of a two-level full factorial design superimposed on a star design. This design allows the estimation of the intercept, slope, curvature, and interaction terms, and it can be used to describe the system with the significance of each term being characterized by the statistical beta coefficients. Central composite design is fairly efficient for a small number of factors. In a coded factor space, the star points are usually located at a distance R R2 ) √2k-2(2k + 2k + C) - 2k-1

(2)

from the center, where k is the number of factors and C is the repeat number of the center point. The design can be broken into three groups of design points, a two-level factorial design (coded (1), star points (coded (R), and the center point (coded 0). The star points allow estimation of the curvature in the model. The center point is repeated C times to give an estimate of the experimental reproducibility. The results from a central composite design can be fitted to a linear model or response surface, which is adequate for describing a wide variety of multifactor chemical systems16 of the following form: k

Yi ) β0 +

k

k

k

∑βx +∑β x +∑∑β xx 2

i i

i)1

ii i

i)1

ij i j

(3)

i)1 j)1

In this case, Yi are the responses; β0 is the intercept which is the value of the fitted response at the center of the design; βi are slopes with respect to each of the three factors; βii are curvature terms; βij are the interaction terms; and xi are the studied factors. Parameters are found by means of the solver tool in an in-house Excel spreadsheet and using the Newton nonlinear fitting method.25 The objective cell is the sum of each square difference between experimental and calculated values of the studied response variable and is minimized using the solver tool in Excel. The model accuracy was evaluated by calculating the medium error using the following equation: nexp

medium error )

i)1

- Y* Yexp × 100 n

exp

∑Y

(4)

where Yexp are the experimental results of response variables, Y* are the predicted values, and n is the number of experimental data points considered.

9786 Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008 Table 3. First and Second Derivative of Second-Order Polynomial Fitting First Derivative of Second-Order Polynomial Fitting

∂/[∂(PRS/St)] ∂/[∂(% PVP/St)] ∂/[∂(H2O/St)]

average particle size in number (µm)

latent heat of fusion (J/g)

19 + 3.24PRS/St - 0.71% PVP/St - 1.51H2O/St -1.48 - 0.71PRS/St + 0.46% PVP/St - 0.31H2O/St +2.93 - 1.51PRS/St - 0.31% PVP/St + 0.17H2O/St

80.02 - 121.07PRS/St - 5.07% PVP/St + 11.06H2O/St -7.13 + 5.07PRS/St - 0.07% PVP/St + 0.32H2O/St 25.56 + 11.06PRS/St - 0.32% PVP/St - 4.11H2O/St

Second Derivative of Second-Order Polynomial Fitting

2

2

∂ /[∂(PRS/St) ] ∂2/[∂(% PVP/St)2] ∂2/[∂(H2O/St)2]

dpn0.5 (µm)

latent heat of fusion (J/g)

+3.24 +0.46 +0.17

-121.07 -0.07 -4.11

As aforementioned, the PRS paraffin wax/styrene mass ratio (PRS/St), percentage of polyvinylpyrrolidone/styrene mass ratio (% PVP/St), and water/styrene mass ratio (H2O/St) were chosen as factors due to their remarkable importance in the microcapsules synthesis process. Average particle size in number (dpn0.5), which represents 50% of microcapsule particles whose average numeric diameter is lower than this value and latent heat of fusion of the microcapsules were chosen as response variables. In this study, according to Carro et al.,26 a sequential exploration of the response variables were chosen and it was carried out in two stages. In the first stage of the process, the relative influence of PRS/St mass ratio ranging from 0.36 to 1.00, % PVP/St from 5.03 to 10.37, and H2O/St from 5.03 to 10.37 and their interactions in an experimental design were established. This involves eight basic experiments (three factors in two levels) undertaken in random order plus three central points. In the second stage, six star points to the above 23 factorial design were added. The star points represent extreme values (low and high) for each factor in this design. The values corresponding to the low (-), high (+), central (0), and (R points for each factor, and the final values for the design factors are shown in Table 1. The experiments were performed in random order to avoid systematic error. 3. Results and Discussions Thermoregulatory properties of the microcapsules have been studied to improve thermal comfort of end-use products. The thermal capacity (latent heat) of the PCM microcapsules depends on the specific quantity used and the particle size of the PCM microcapsules determines its application. For that reason, it is interesting to know how the experimental conditions affect the amount of PCM encapsulated and also the microcapsules size. The initial screening design allowed to detect those synthesis variables having significant influence on the microcapsules properties. An analysis of statistical significant variables showed that PRS/St mass ratio was the main influential variable for the latent heat and both the % PVP/St and H2O/St were the most important on the particle size in number of microcapsules. This behavior was observed in previous works.10,18 Moreover response hypersurfaces in the multidimensional factorial space exhibits curvature in the domain of the experimental design. Thus, a central composite design was required. Table 1 shows the factorial design of experiments with three factors in two levels, 17 experiments including three central repeats, changing PRS/St mass ratio from 0.25 to 1.11 and % PVP/St and H2O/St mass ratios from 4.09 to 11.31. The average particle size in number and latent heat of microcapsules were the response variables. The experimental results of the central composite design were fitted using a second-order polynomial equation. Results shown in Figure 2 illustrate a good fitting with a medium error of 7% for the latent heat and around 8% for

the dpn0.5. The equation and the parameters for the both fittings are shown in Table 2. Table 3 shows the first and second derivative equations obtained of second-order polynomial fitting with respect to PRS/ St, % PVP/St, and H2O/St mass ratios. The mathematical treatment of the model proposed for latent heat allows to obtain an optimum for PRS/St, % PVP/St, and H2O/St mass ratios of 1.02, 9.43, and 8.23, respectively. According to the second derivative, this point corresponds to a maximum response of 102.42 J/g indicating that about a 50% w/w of paraffin wax has been successfully encapsulated. At this optimal condition, an average particle diameter size in number of 4.80 µm was obtained. On the other hand, the mathematical treatment for the dpn0.5 yielded a minimum of 4.03 µm corresponding to the synthesis variables values of PRS/St 0.76, % PVP/St 10.65, and H2O/St 9.14 mass ratios. 3.1. Influence of Synthesis Variables on the Heat Capacity of Microcapsules. The 3D response surfaces of the latent heat model are shown in Figure 3. They were drawn holding constant each fixed variable at its optimal condition. Figure 3a shows the influence of the PRS/St and % PVP/St mass ratios on the latent heat of microcapsules. As can be seen the PRS/St has higher influence than % PVP/St and mainly at high PRS/St mass ratio. The highest storage thermal capacity of microcapsules is obtained for the highest PRS/St mass ratio and the lowest % PVP/St mass ratio values studied. In the same way, the maximum storage thermal capacity is obtained at the highest PRS/St mass ratio (Figure 3b). Moreover it can be seen that the water content on the synthesis has a great influence on the latent heat of the microcapsules a maximum latent heat in the range of 6.0-9.0 for all the range of the PRS/St mass ratio studied. Results indicated that in the studied range the thermal capacity of the microcapsules increases as higher the PRS/St is used. On the other hand, Figure 3c confirms that the variable H2O/St induces the curvature showing maximum values ranging from 6.0 to 9.0 and that % PVP/St mass ratio has a negligible influence. According to this figure, the absolute maximum is close to the optimum got above by the mathematical treatment. These results are in agreement with the previous work18 for PRS/St mass ratios lower than 1.0 where thermal capacity of the microcapsules always increased as higher the PRS/St was used. Furthermore, the same trend was observed by other authors.27,28 Ma et al.28 reported that as the amount of hexadecane increased, hexadecane was more easily encapsulated by the polymer. The above results suggested that the phase separation was promoted by increasing PCM amount. As it has previously been shown, the thermal capacity depends on the total amount of paraffin encapsulated. In this study, about 50% w/w of paraffin was encapsulated with a latent heat of 102.42 J/g. In this sense, other authors in the literature have

Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008 9787

Figure 3. (a) Effect of PRS/St and % PVP/St mass ratios at 8.23 of H2O/St mass ratio, (b) effect of the H2O/St and PRS/St mass ratios at 9.43 of % PVP/St, and (c) effect of the H2O/St and % PVP/St mass ratios at 1.02 of PRS/St on the latent heat of microcapsules.

reported 20 wt % of paraffin encapsulated by using complex coacervation or spray-drying methods.8 Xing et al.27 found an

Figure 4. (a) Effect of PRS/St and % PVP/St mass ratios at 8.23 of H2O/St mass ratio, (b) effect of the H2O/St and PRS/St mass ratios at 9.43 of % PVP/St, and (c) effect of the H2O/St and % PVP/St mass ratios at 1.02 of PRS/St on the average particle size in number of microcapsules.

optimum of 56 wt % of paraffin by in situ polymerization method. According to the storage capacity of the microcapsules obtained, this method seems to be appropriate to produce microcapsules to be incorporated into textiles.

9788 Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008

Figure 5. SEM micrographs of the optimal microcapsules obtained. (a) General view of batch of microcapsules. (b) Cross-section of a single microcapsule.

3.2. Influence of Synthesis Variables on the Particle Size of Microcapsules. The particle size and size uniformity are important factors for the incorporation of microcapsules to textile materials.6 According to different authors, it is possible to use particle sizes of microcapsules with a wide range from 0.5 to 100 µm for textile application.3-5 Sarier and Onder29 recommended wax-based particles smaller than 10 µm for

incorporating in textile by coating. It seems obvious that as lower the particle size of microcapsules containing PCM better properties, such as softness, a homogeneous heat absorbing or heat releasing, and good washing durability, of the smart temperature adaptable textiles independent of the incorporating method used. As it was shown above, the mathematical treatment for the dpn0.5 yielded a minimum of 4.03 µm, which is similar to the particle size in number obtained by using the synthesis variables values for the maximum of latent heat (4.80 µm). Therefore with the aim to produce microcapsules containing PCM with the highest thermal storage capacity, the synthesis variables values corresponding to the maximum latent heat PRS/St 1.02, % PVP/St 9.43, and H2O/St 8.23 mass ratios were selected. 3D simulations from the second-order polynomial model for the dpn0.5 are shown in Figure 4 holding constant the corresponding fixed variable on the aforementioned selected values. Figure 4a shows that the % PVP/St mass ratio has a larger influence on the average particle size in number of microcapsules as compared with the PRS/St mass ratio. Mainly at values lower than 8.0 a decrease of dpn0.5 with increasing the amount of suspension agent (% PVP/St) is observed. This could be attributed to the interfacial tension between polymer and aqueous phase that decreased due to an increase of the content of polyvinylpyrrolidone in the polymerization medium.21 Moreover even at synthesis conditions PVP/St higher than 6.0 and the highest PRS amount, it is possible to develop microcapsules with a particle size in the required range. Figure 4b shows the 3D response surface at a 9.43 value of the % PVP/St mass ratio. It can be observed that the average diameter in number of the microcapsules is strongly depending on the PRS amount at the lowest H2O/St value. On the other hand, the H2O/St mass ratio seems to present a high influence on the diameter particle size in number at lower the PRS/St ratio is used in the microcapsules synthesis. Changes of PRS/St and H2O/St mass ratios produce variations of viscosity, interfacial tension, and density of each different phase, parameters that have a strong influence on the particle diameter size.19,21 Figure 4c was drawn at a fixed PRS/St mass ratio of 1.02. According to this figure, the particle size of the microcapsules increases as higher H2O/St and lower % PVP/St mass ratios are used. Besides, the particle size can be outside the desire range for textile applications depending on the both mass ratios. That is, a % PVP/St higher than 7.5 should be used in the synthesis for the whole H2O/St mass ratio to obtain particles size with a diameter lower than 10 µm.

Figure 6. Particle size distribution of microcapsules synthesized using the selected variables’ values.

Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008 9789

Furthermore, the influence of H2O/St mass ratio can be negligible when the amount of paraffin wax and the amount of polyvinylpyrrolidone are high (Figure 4b and c). Nevertheless, its effect is important when paraffin wax and polyvinylpyrrolidone are present in small proportions (Figure 4a and c). 3.3. Synthesis of Microcapsules Using the Selected Variables’ Values. In order to verify the effectiveness of the model prediction, some experiments were carried out using the optimal formulation PRS/St 1.02, % PVP/St 9.43, and H2O/St 8.23 mass ratios. The average particle size in number of microcapsules obtained was 4.53 µm (5.6% error related to the dpn0.5 predicted), as can be observed in Figure 6. On the other hand, DSC measurements showed an average latent heat of 104.7 J/g, resulting in a 2.2% error related to latent heat. From SEM micrographs (Figure 5a and b), it is evident that the microcapsules obtained were spherical, with smooth surfaces. Figure 5b shows the internal structure of a big microcapsule microtomized by a cross-sectioning. A great paraffin wax inside the microcapsule has been observed; this fact confirms the high quantity of paraffin wax incorporated. The particle core constitutes around 42% of the volume particle, while the outer shell is approximately 5 µm thick. Microcapsules obtained in this study have similar latent heat, smaller particle sizes, and narrower particle size distribution in number than those reported in our previous work,18 without using a SPG membrane. On the other hand, Ma et al.12,28 prepared microcapsules by the Shirasu porous glass emulsification technique, and the encapsulation efficiency reported is slightly lower than that obtained in this work. These facts make the PCM microcapsules reported in this work potentially useful for textile applications. 4. Summary The application of a central composite design resulted in a useful tool for the characterization and optimization of PCM microcapsules prepared by the Shirasu porous glass (SPG) emulsification technique and a subsequent suspensionlike polymerization process. The second-order polynomial model provides a good and simple fitting of the average particle size in number and the latent heat of microcapsules. The empirical equations that describe the behavior of the system allowed the prediction of the amount of paraffin wax encapsulated and mean particle size in number as a function of synthesis variables with an error of less than 8%. It was observed that the principal influence corresponds to changes in the mass ratio of PRS/St for the latent heat of microcapsules. Moreover, the H2O/St mass ratio is the variable that induces the curvature, and the % PVP/St mass ratio has a negligible influence. On the other hand, the average particle size is strongly influenced by % PVP/St, but changes of the PRS/St and H2O/St mass ratios produce variations of the viscosity, interfacial tension, and density of each different phase, which have a strong influence on the particle size. The mathematical treatment for the latent heat showed a maximum response of 102.42 J/g, whereas the dpn0.5 yielded a minimum of 4.03 µm. The synthesis variables’ values corresponding to the maximum latent heat PRS/St 1.02, % PVP/St 9.43, and H2O/St 8.23 allowed the particle size in number of 4.80 µm to be obtained. Then, this formulation was selected as the optimum in order to produce microcapsules to be applied in textiles. These conditions were used in a repeatability study to assess the effectiveness of the model prediction. Only a 5.6% error related to the main diameter in number predicted and a 2.2%

error related to the latent heat predicted were obtained. Furthermore, DSC thermograms showed high encapsulation efficiency (50 wt % of paraffin encapsulated), and SEM micrographs of experiments carried out showed that microcapsules had spherical shapes and low particle size. Acknowledgment Financial support from ASINTEC S.A. and the fellowship and grant of Ref. PBC08-0243-1458 from Consejerı´a de Ciencia y Tecnologı´a (JCCM) are gratefully acknowledged. Literature Cited (1) Nelson, G. Application of micro-encapsulation in textiles. Int. J. Pharm. 2002, 242, 55. (2) Nelson, G. Microencapsulates in textile coloration and finishing. ReV. Prog. Color. 1991, 21, 72. (3) Ghosh, S. K. Functional Coatings by Polymer Microencapsulation; WILEY-VCH: Weinheim, 2006. (4) Colvin, D. P.; Bryant, Y. G. Protective clothing containing encapsulated phase change materials. Advances in heat and mass transfer in biotechnology (HTD). ASME IMECE Proc. 1998, 362, 123. (5) Pause, B. Nonwoven protective garments with thermo-regulating properties. J. Ind. Textil. 2003, 33, 93. (6) Shin, Y.; Yoo, D.; Son, K. Development of thermoregulating textile materials with microencapsulated phase change materials (PCM). IV. Performance properties and hand of fabrics treated with PCM microcapsules. J. Appl. Polym. Sci. 2005, 97, 910. (7) Cox, R. Dec. Sypnopsis of the new thermal regulating fiber outlast. Chem. Eng. Res. Des. 1998, 48, 475. (8) Hawlader, M. N. A.; Uddin, M. S.; Khin, M. Microencapsulated PCM thermal-energy storage system. Appl. Energy 2003, 74, 195. (9) Farid, M. M.; Khudhair, A. M.; Razack, S. A. K.; Al-Hallaj, S. A review on phase change energy storage: materials and applications. Energy ConVer. Manage. 2004, 45, 1597. (10) Sa´nchez, L.; Sa´nchez, P.; de Lucas, A.; Carmona, M.; Rodrı´guez, J. F. Microencapsulation of PCMs with a polystyrene shell. Colloid Polym. Sci. 2007, 285, 1377. (11) Omi, S.; Katami, K.; Yamamoto, A.; Iso, M. Synthesis of polymeric microspheres employing SPG emulsification technique. J. Appl. Polym. Sci. 1994, 51, 1. (12) Ma, G. H.; Nagai, M.; Omi, S. Study on preparation of monodispersed poly(styrene-co-N-dimethylaminoethyl methacrylate) composite microspheres by SPG (Shirasu Porous Glass) emulsification technique. J. Appl. Polym. Sci. 2001, 79, 2408. (13) Omi, S.; Matsuda, A.; Imamura, K.; Nagai, M.; Ma, G. H. Synthesis of monodisperse polymeric microspheres including polyimide prepolymer by using SPG emulsification technique. Colloids Surf., A. 1999, 153, 373. (14) Nuisin, R.; Ma, G. H.; Omi, S.; Kiatkamjornnwong, S. Dependence of Morphological Changes of Polymer Particles on Hydeophobic/Hydrophilic Additives. J. Appl. Polym. Sci. 2000, 77, 1013. (15) Yuyama, H.; Yamamoto, K.; Shirafuji, K.; Ma, G. H.; Nagai, M.; Omi, S. Preparation and analysis of uniform emulsion droplets using SPG membrane emulsification technique. J. Appl. Polym. Sci. 2000, 77, 2237. (16) Palasota, J. A.; Deming, S. N. Central composite experimental designs: Applied to chemical systems. J. Chem. Educ. 1992, 69, 560. (17) Barnabas, I. J.; Dean, J. R.; Tomlinson, W. R.; Owen, S. P. Experimental Design Approach for the Extraction of Polycyclic Aromatic Hydrocarbons form Soil Using Supercritical Carbon Dioxide. Anal. Chem. 1995, 67, 2064. (18) Sa´nchez, L.; Sa´nchez, P.; Carmona, M.; de Lucas, A.; Rodrı´guez, J. F. Influence of operation conditions on the microencapsulation of PCMs by means of suspension-like polymerization. Colloid Polym. Sci. 2008, 286, 1019. (19) Ullmann′s Encyclopedia of Industrial Chemistry; WILEY-VCH: Weinheim, 2003. (20) Vivaldo-Lima, E.; Wood, P. E.; Hamielec, A. E. An Updated Review on Suspension Polymerization. Ind. Eng. Res. 1997, 36, 939. (21) Yuan, H. G.; Kalfas, G.; Ray, W. H. Suspension Polymerization. ReV. Macromol. Chem. Phys. 1991, 283, 215. (22) Mlynek, Y.; Resnick, W. Drop Sizes in an Agitated Liquid-Liquid System. AIChE J. 1972, 18, 122. (23) Nakashima, T.; Shimizu, M.; Kukizaki, M. Membrane emulsification operation manual; Industrial Research Institute of Miyazaki Prefecture: Japan, 1991.

9790 Ind. Eng. Chem. Res., Vol. 47, No. 23, 2008 (24) Christov, N. C.; Ganchev, D. N.; Vassileva, N. D.; Denkov, N. D.; Danov, K. D.; Kralchevsky, P. A. Capillary mechanisms in membrane emulsification: Oil-in-water emulsions stabilized by Tween 20 and milk proteins. Colloids Surf., A 2002, 209, 83. (25) Lo´pez, O. M. Me´todos iterativos de resolucio´n de ecuaciones. Teorema del punto fijo y aplicaciones. Me´todo de Newton; Alhambra: Madrid, 1986. (26) Carro, M. A.; Cobas, J. C.; Rodrı´guez, J. B.; Lorenzo, R. A.; Cela, R. Application of chemometric techniques to the optimization of the solidphase extraction of 27 pesticides before GC-MIP-AES analysis. J. Anal. Atom. Spectrom. 1999, 14, 1867. (27) Xing, J.; Hongyun, L.; Shujun, W.; Lu, Z.; Hua, C. Preparation and thermal properties of form stable paraffin phase change material encapsulation. Energy ConVers. Manage. 2006, 47, 2515.

(28) Ma, G. H.; Su, Z. G.; Omi, S.; Sundberg, D.; Stubb, J. Microencapsulation of oil with poly(styrene-N,N-dimethylaminoethyl methacrylate) by SPG emulsification technique: Effects of conversion and compositon of oil phase. J. Colloid Interface Sci. 2003, 266, 282. (29) Sarier, N.; Onder, E. Thermal characteristics of polyurethane foams incorporated with phase change materials. Thermochim. Acta 2007, 452, 149.

ReceiVed for reView July 18, 2008 ReVised manuscript receiVed September 23, 2008 Accepted October 6, 2008 IE801107E