Revista Mexicana de Ingeniería Química

CONTENIDO

Volumen 8, número 3, 2009 / Volume 8, number 3, 2009

213 Derivation and application of the Stefan-Maxwell equations

(Desarrollo y aplicación de las ecuaciones de Stefan-Maxwell)

Stephen Whitaker

Biotecnología / Biotechnology

245 Modelado de la biodegradación en biorreactores de lodos de hidrocarburos totales del petróleo

intemperizados en suelos y sedimentos

(Biodegradation modeling of sludge bioreactors of total petroleum hydrocarbons weathering in soil

and sediments)

S.A. Medina-Moreno, S. Huerta-Ochoa, C.A. Lucho-Constantino, L. Aguilera-Vázquez, A. Jiménez-

González y M. Gutiérrez-Rojas

259 Crecimiento, sobrevivencia y adaptación de Bifidobacterium infantis a condiciones ácidas

(Growth, survival and adaptation of Bifidobacterium infantis to acidic conditions)

L. Mayorga-Reyes, P. Bustamante-Camilo, A. Gutiérrez-Nava, E. Barranco-Florido y A. Azaola-

Espinosa

265 Statistical approach to optimization of ethanol fermentation by Saccharomyces cerevisiae in the

presence of Valfor® zeolite NaA

(Optimización estadística de la fermentación etanólica de Saccharomyces cerevisiae en presencia de

zeolita Valfor® zeolite NaA)

G. Inei-Shizukawa, H. A. Velasco-Bedrán, G. F. Gutiérrez-López and H. Hernández-Sánchez

Ingeniería de procesos / Process engineering

271 Localización de una planta industrial: Revisión crítica y adecuación de los criterios empleados en

esta decisión

(Plant site selection: Critical review and adequation criteria used in this decision)

J.R. Medina, R.L. Romero y G.A. Pérez

Revista Mexicanade Ingenierıa Quımica

1

Academia Mexicana de Investigacion y Docencia en Ingenierıa Quımica, A.C.

Volumen 10, Numero 2, Agosto 2011

ISSN 1665-2738

1Vol. 10, No. 2 (2011) 301-312

MORPHOMETRIC CHARACTERIZATION OF SPRAY-DRIED MICROCAPSULESBEFORE AND AFTER α-TOCOPHEROL EXTRACTION

CARACTERIZACION MORFOMETRICA DE MICROCAPSULAS SECADAS PORASPERSION ANTES Y DESPUES DE LA EXTRACCION DE α-TOCOFEROL

M.X. Quintanilla-Carvajal1, L.S. Meraz-Torres1, L. Alamilla-Beltran1, J.J. Chanona-Perez1, E. Terres-Rojas2,H. Hernandez-Sanchez1, A.R. Jimenez-Aparicio3 and G.F. Gutierrez-Lopez1∗

1Escuela Nacional de Ciencias Biologicas del Instituto Politecnico Nacional. Carpio y Plan de Ayala s/n. Col.Santo Tomas. Mexico, D. F. CP. 11340.

2Instituto Mexicano del Petroleo, Laboratorio de Microscopia Electronica de Barrido Ambiental.3Centro de Desarrollo de Productos Bioticos del Instituto Politecnico Nacional. Km 35, Yautepec-Morelos, Mexico.

Received 5 of June 2011; Accepted 14 of June 2011

AbstractMicrocapsules were obtained by spray-drying of α-tocopherol (AT) - in- water emulsions subjected to single anddouble atomization yielding single atomized microcapsules (SAM) and double atomized microcapsules (DAM)which had different core to wall material ratios. Micrographs of whole and AT extracted of SAM and DAM obtainedfrom Environmental Scanning Electron Microscopy (ESEM) were image analyzed and the Feret diameter, projectedarea, perimeter, maximum perimeter, shape factor and Fractal Dimension of contour (FDc) and texture (FDt) weredetermined. Whole DAM presented higher FDc and FDt than SAM, while SAM presented higher FDc values for allthe microcapsules and higher DFt values for the AT to wall materials ratios 1:4 after the AT extraction. Most of themicrocapsules displayed significantly different perimeter but non- significant projected areas, whereas the extractedmicrocapsules tended not to show significant differences between these two parameters.

Keywords: morphometric parameters, spray drying, image analysis, alpha tocopherol.

ResumenSe obtuvieron microcapsulas mediante secado por aspersion de emulsiones de α-tocoferol (AT) - en-agua sometidasa atomizacion simple o doble de las que se obtuvieron microcapsulas de atomizacion simple (MAS) y microcapsulasde atomizacion doble (MAD) con diferentes relaciones entre los materiales de pared y el AT a encapsular. Seobtuvieron micrografıas de MAS y MAD antes y despues de la extraccion del AT por Microscopia Electronica deBarrido Ambiental (MEBA) a las que se les determino el diametro de Feret, el area proyectada, el perımetro, elperımetro maximo, el factor de forma, la Dimension Fractal de contorno (DFc) y la Dimension Fractal de textura(FDt). Las MAD presentaron mayores valores de DFc y DFt que las MAS, mientras que MAS presentaron mayoresvalores de DFc para las relaciones de AT a materiales de pared de 1:4 despues de la extraccion del AT. La mayorıade las microcapsulas presentaron valores de perımetros significativamente diferentes, pero no de areas proyectadas,mientras que las microcapsulas a las que se les extrajo el AT no presentaron diferencias significativas entre estos dosparametros.

Palabras clave: parametros morfometricos, secado por aspersion, analisis de imagenes, alfa-tocoferol.

∗Corresponding author. E-mail: gusfgl@gmail.com

Publicado por la Academia Mexicana de Investigacion y Docencia en Ingenierıa Quımica A.C. 301

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

1 Introduction

Microencapsulation techniques offer the possibility ofprotection and controlled release of active compounds.In this process, small particles are surroundedby a coating or embedded in a homogeneous orheterogeneous matrix to produce small capsules whichmay have a number of useful characteristics. Infood and pharmaceutical applications, spray dryingis often the final step of the microencapsulationprocess inducing the liquid droplets, solid particlesor aroma compounds to be entrapped into thin filmsof food or chemical grade microencapsulating agents(Gharsallaoui et al., 2007; Champagne y Fustier 2007;Perez-Alonso et al., 2008).

Extensive research aiming to understandingthe fundamentals and applications of spray dryingin encapsulation processes has been done andthe progress in development of modern foodmaterials has been enhanced by the advances inpowder technologies and improvement in powdercharacterization methods (Woo et al., 2010).Nowadays, it is expected to understand powderbehavior through physical characterization such assize, shape and surface area and to correlate theseproperties with encapsulation efficiency (Perea-Floreset al., 2010).

In the case of spray-dried emulsions, the amountof non-encapsulated core material is a key parameterin determining product quality. It has been shownthat in fat-containing dairy powders, the surfaceof individual powder particles is almost completelycovered with a thin layer of fat (Kim et al., 2009)which determines the flowability and wettability ofthese powders (Vega and Roos, 2006). There aredifferent methods for the determination of free orsurface lipid materials in spray-dried emulsions whichare based on the extraction of different fractions ofthe non-encapsulated oil (Drusch and Berg, 2008;Lekago and Dunford, 2010). Information aboutmorphology, structure and microstructure of thesecapsules can be obtained from image analysis throughdifferent computer vision systems. It is possiblethen, to identify the structure-function relationshipsto understand these complex systems (Aguilera, 2007;Chanona et al., 2008).

Irregular surfaces and textures of different foodparticles such as coffee, milk, or maltodextrin powdershave been characterized successfully through fractalanalysis. The key to quantify the irregularity ofthe contours and surfaces of food materials is theevaluation of their fractal dimension (D) by extracting

structural and microstructural features from differentimages (Perea-Flores et al., 2010; Tapia-Ochoateguiet al., 2011). However, the relationships betweenthe nature of the wall materials used in encapsulationprocesses and the fractal dimension of the resultingcapsules have not been reported yet. Gum arabicand maltodextrin are commonly used in encapsulationprocesses since they are important carriers and coatingagents used for the spray drying of flavors andcolorants that need to be protected from differentoxidative reactions (Pitalua et al., 2010). Also, themicrostructure of different kind of emulsions hasbeen described (Murillo-Martinez et al., 2011) butno quantitative descriptors of the properties of theemulsion have been proposed. The objective ofthis work was to characterize by image analysis, themorphology of capsules of alpha tocopherol usingdifferent ratios of wall materials (gum arabic andmaltodextrin).

2 Materials and methods

2.1 Materials

(±)α-tocopherol (AT; HPLC grade, Sigma-Aldrich,Toluca, State of Mexico, Mexico) was used as corematerial. Maltodextrin DE 20 (MD; ComplementosAlimenticios, Naucalpan, State of Mexico, Mexico)and gum arabic from Acacia Senegal (GA; Alfred L.Wolf, S.A. de C.V., Mexico City, Mexico) were usedas biopolymers.

Chloroform (C), hexane (H), isopropanol (IP) andmethanol (M; HPLC grade) were all purchased fromSigma-Aldrich (Toluca, State of Mexico, Mexico).Grade I water was used in all assays.

2.2 Emulsion preparation

MD-GA in different proportions (3:1, 2:1, 1.5:1,1:1) were dissolved in water and left to stand for24 h to allow complete hydration. Afterwards,AT was added dropwise, in darkness conditions,to the aqueous biopolymer solutions with help ofa Rival homogenizer (model IB901-MIX, Florida,USA) operated at 3000 rpm during 6 min. Theemulsions were then added water to bring solidscontents to 20% (w/w).

2.3 Spray-drying

A laboratory scale spray-dryer equipped with a 2-fluid atomizing nozzle was employed for drying the

302 www.rmiq.org

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

emulsions. Inlet temperature of air was 190 ± 3 ◦C,outlet temperature of air was 100 ± 3 ◦C, the flow ratewas 1.14L/h, and atomizer air pressure 0.00012 Kg/m2

(Perez-Alonso et al., 2009; Pulido and Beristain,2010).

Emulsions were processed in two manners:

a) Spray-dried immediately after manufacture and

b) First cold-sprayed, and then spray-dried,using in all cases the experimental conditionsmentioned above.

The single atomized microcapsules were coded asSAM3:2:1, SAM3:2:2, SAM3:1:1, SAM2:2:1, SAM2:1:1,while the double atomized microcapsules werecoded as DAM3:2:1, DAM3:2:2, DAM3:1:1, DAM2:2:1,DAM2:1:1, respectively. All microcapsules were putinto ziplock bags and covered with Reynold aluminumpaper and put into a dessicator until required analysis.

2.4 AT extraction

The surface and inner free AT was extracted fromSAM and DAM by adding 10 mL of chloroform to100 mg of each powder, and stirred with vortex atroom temperature for 2 min. The resulting solutionwas filtered through cotton to retain wall material andwashed out with 2 mL of chloroform. The organicphase containing the surface and inner free AT wasevaporated to dryness in a Buchi Rotavapor (modelR205, Flawil, Switzerland) operated under reducedpressure and 50 ◦C.

Encapsulated AT was determined by adding 10mL of water to the cotton containing the filtrate(microcapsules matrices) followed by the addition of12 mL of chloroform for disrupting the microcapsulesmatrices. The filtrate was collected in a glass beaker,and added 10 mL of chloroform. This mixturewas transferred to a burette and the lower organicphase was separated into a beaker containing 1 g ofanhydrous Na2SO4. Finally the organic phase wasevaporated to dryness as mentioned before, remainingthe encapsulated AT.

2.5 Morphometric analysis

2.5.1. Morphometric parameters

Image processing was carried out according toPedreschi et al, (2004). Images of the powders of2048 × 1536 pixels were captured using a digitalcamera (Nikon Digital Sight DS-2Mv, TV Lens0.55X, DS, Japan) fitted to a optical microscope

(Nikon 50I, USA) and by means of the softwareNIS-Elements F V2.30. The images of the powderswere converted to grey scale (8 bits) maps andthen to binary images (black and white) by usingthe software ImageJ 1.34 (National Institutes ofHealth, Bethesda, MD, USA). Thresholding wasapplied to each image by using default settings (95-111). Fill holes tool was then used and the imagesinverted (Mery and Pedreschi, 2005). The followingmorphological parameters were determined: Feretdiameter, perimeter, maximum perimeter and shapefactor. With these results, the particle size distributioncould be evaluated.

The area-perimeter relationship was alsoevaluated: Area-perimeter methods are generally usedto estimate the fractal dimension (D) of objects (inthis case agglomerates) to evaluate their complexity.This method measures the extent that patch perimeters“fill” the two-dimensional plane. The perimeter-arearelationship for a given set of agglomerates is givenby:

P = kAD/2 (1)

Where the area A is the number of pixels making up agiven object, the perimeter P is a count of the numberof pixel edges, and k is a scaling constant. The slopeof the log-log area-perimeter plot for a set of objectsgives an “average” fractal dimension (Burrough,1986). Agglomerates with perfectly square objects(low perimeter: area ratio) have a fractal dimensionD = 1, while those containing highly complexconvoluted objects (high perimeter:area ratios) havefractal dimensions approaching 2. The method can beused to determine the relative “edginess” of an image.For single agglomerates, the perimeter dimensionreduces to:

DFc =2 log(P)log(A)

(2)

Where DFC is the Contour Fractal Dimension and wasobtained adding 1 to the slope of the equation (Barleitaand Barbosa-Canovas, 1993; Bellouti et al., Jimenez etal., 2005).

2.5.2 Microstructural analysis of microcapsules

For the structure analysis of the tablets, sampleswere mounted on cylindrical stubs which were fittedwith double coated conductive carbon tape. Anenvironmental scanning electron microscope (FEIQuanta 600, FEI Co., Hillsboro, Oregon., U.S.A.)was used to analyze the surface morphology of thecapsules before and after the partial extraction ofthe extractable tocopherol. Digital images where

www.rmiq.org 303

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

obtained at 2 magnifications, 500× and 2000×.Images were stored as bit-maps in a gray scalewith brightness values between 0 and 255 for eachpixel constituting the image. A generalization ofthe Box Counting method was used to evaluate thefractal dimension of the images. In this work,the shifting differential box-counting method (SDBC)(Chen et al., 2003) was used to evaluate the fractaldimension of texture of ESEM images using theImageJ 1.34 software. Image segmentation andextraction were also performed using the Image J 1.34software to evaluate macroscopic structural changes.Image segmentation included cropping, conversionof color image to grey-scale values, and backgroundsubtraction to obtain the binary image from theoriginal color image.

2.6 Statistical analysis

All analytical tests and microencapsulationexperiments were carried out in duplicate and inrandomized order. Results were reported as themean ± standard deviation. Significant differencesamong results were determined by analysis of variance(ANOVA) at α = 0.05. All the analyses wereperformed using MS Excel 2007.

3 Results and discussion

3.1 Morphometric parameters

As stated in Section 2, all the images of DAM andSAM were processed and analyzed to obtain thedifferent morphometric parameters. An example of

the binarization process is presented in Fig. 1. Eachoriginal image represents the DAM3:2:1 and SAM3:2:1.In this figure, the gray-scale images represent theoriginal emulsion, while the black and white ones arethe processed ones according to the conversion carriedout from grey scale (8 bits) maps to binary images(black and white).

Fig. 2 shows the particle size distribution (in termsof the Feret diameter) of DAM and SAM and theirrespective distribution of frequencies. Apparentlythe graphs showed no evidence of a significantdifference in particle size distribution between thetwo processes. To verify this assumption, a one-way ANOVA (α = 0.05) was applied to the variousmorphometric parameters.

Tables 1 and 2, present the average valuesof the morphometric parameters obtained fromimage analysis of DAM and SAM respectively.Results suggested that homogenization influencessignificantly (p < 0.05) the morphometriccharacteristics of the powders without showing anyspecific trend with wall material to tocopherol ratios.Evaluated Feret diameters varied between 2 and 10µm.

DAM

SAM Figure 1. Images of DAM3:2:1 and SAM3:2:1 (40X)

 Fig. 1. Images of DAM3:2:1 and SAM3:2:1 (40X)

Table 1. SAM and DAM morphometric parameters

Microcapsules Area (µm2) Maximum perimeter (µm) Perimeter (µm) Shape factor

3:2:1 DAM 1.35E-01 ± 2.30E-05a 50.8± 3.70E-03a 42.4 ± 3.14E-03a 0.7442 ± 1.37E-02a

SAM 9.99E-02 ± 1.57E-05a 37.4 ± 2.66E-03b 31.0 ± 2.23E-03b 0.7509 ± 1.10E-02a

3:2:2 DAM 5.67E-02 ± 2.10E-05a 22.5 ± 1.94E-03c 18.6 ± 1.68E-03c 0.8472 ± 1.15E-02a

SAM 8.10E-02 ± 2.99E-05a 29.9 ± 3.51E-03d 24.7 ± 2.93E-03c 0.8130 ± 1.43E-02a

3:1:1 DAM 1.10E-02 ± 7.14E-06a 48.7 ± 2.09E-03e 40.4 ± 1.75E-03d 0.7686 ± 1.18E-02b

SAM 1.24E-02 ± 2.99E-05a 31.4 ± 3.34E-03 f 26.2 ± 2.82E-03e 0.8448 ± 1.37E-02c

2:2:1 DAM 1.25E-02 ± 1.43E-05a 46.3 ± 3.02E-03g 38.6 ± 2.54E-03c 0.7475 ± 1.10E-02d

SAM 1.37E-02 ± 8.18E-06a 45.1 ± 1.71E-03g 37.6 ± 1.43E-03c 0.7730 ± 7.23E-03e

2:1:1 DAM 6.07E-02 ± 6.59E-06b 33.4 ± 1.81E-03g 27.7 ± 1.52E-03c 0.7607 ± 9.20E-03 f

SAM 1.04E-02 ± 2.35E-05c 32.2 ± 3.14E-03g 26.7 ± 2.65E-03c 0.8131 ± 1.67E-02g

Superscripts with the same letter within the same column presented significant differences (p < 0.05)

304 www.rmiq.org

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

Figure 2. Particle size distribution of a) DAM3:2:1 and SAM3:2:1, b) DAM3:2:2 and SAM3:2:2,

c) DAM3:1:1 and SAM3:1:1, d) DAM2:2:1 and SAM2:2:1 and e) DAM2:1:1, and SAM2:1:1 

0.00%

20.00%

40.00%

60.00%

-4.16E-1 0.02 0.04 0.06 0.08 0.1 0.12

Fre

cuen

cy (%

)

Feret Diameter (mm)

DAM

SAM

0.00%

20.00%

40.00%

60.00%

80.00%

0 0.02 0.04 0.06 0.08 0.1 0.12

Fre

cuen

cy (%

)

Feret Diameter (mm)

DAM

SAM

0.00%

20.00%

40.00%

60.00%

80.00%

-4.16E-1 0.02 0.04 0.06 0.08 0.1 0.12

Fre

cuen

cy (%

)

Feret Diameter (mm)

DAM

SAM

0.00%

20.00%

40.00%

60.00%

-4.16E-1 0.02 0.04 0.06 0.08 0.1 0.12

Fre

cuen

cy (%

)

Feret diameter (mm)

DAM

SAM

0.00%

20.00%

40.00%

60.00%

-4.16E-1 0.02 0.04 0.06 0.08 0.1 0.12

Fre

cuen

cy (%

)

Feret Diameter (mm)

DAM

SAM

b)a)

d)c)

e)

Fig. 2. Particle size distribution of a) DAM3:2:1 and SAM3:2:1, b) DAM3:2:2 and SAM3:2:2, c) DAM3:1:1 and SAM3:1:1,d) DAM2:2:1 and SAM2:2:1 and e) DAM2:1:1, and SAM2:1:1.

Table 2. SAM and DAM Feret diameters

Microcapsules DAM (µm) SAM (µm)

3:2:1 10.8aA ± 0.623 7.87aB ± 0.498

3:2:2 4.82bC ± 0.343 6.47aD ± 0.6193:1:1 10

aE ± 0.31 6.46aE ± 0.568

2:2:1 3.8bF ± 0.497 9.61bG ± 0.3342:1:1 6.84c ± 0.270 6.51a ± 0.556

Lowercase superscripts indicate significant difference inthe same column (p < 0.05)Uppercase superscripts indicate significant difference inthe same row (p < 0.05)

www.rmiq.org 305

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

In Table 3, it is possible to observe that forpowders having the higher amounts of maltodextrinin the matrix also had higher sizes (evaluated asFeret diameters). Similar findings were reported byVignolles et al, (2009). On the other hand, Yeet al, (2007) reported that the powder particle sizeis independent of the size of the original emulsionglobule, which coincide with our results. Thisshrinkage or swelling could be due to a combinationof different ratios of wall material and tocopherolas Klaypradit and Huang suggested (2008). Theyreported a decrease in the size of the capsule whenmaltodextrin was present in high proportions, resultsin agreement with those obtained in this work.

Feret diameter, perimeter and maximum perimeterof those powders obtained from emulsions preparedwith MD:GA:AT of 3:2:1, 3:2:2 and 3:1:1, presentedsignificant differences when comparing DAM

and SAM. However, the projected area for bothmicrocapsules did not show significant differences.These results indicated that microcapsules havingdifferent shapes, presented fairly similar areas(different contour) and it is possible to point out thatthe evaluation of the size of the particles must includemorphological indicators of the shape and dimension.The shape factor which describes the irregularity ofthe objects, in most of the cases, was higher for SAM.The powders that passed twice through the nozzle,presented more irregularity. Values of morphologicalparameters were significantly different for DAM(p ≤ 0.05) at least in one morphometric parameter.

Surface and inner free AT extraction of DAMand SAM induced morphometric differences betweenthese two materials as compared with the originalmicrocapsules.

 

0%

5%

10%

15%

20%

25%

30%

0 0.005 0.01 0.015 0.02

Fre

cuen

cy (%

)

Feret diamter (mm)

DAM WITH EXTRACTION

SAM WITH EXTRACTION

0%5%

10%15%20%25%30%35%40%45%

0 0.005 0.01 0.015 0.02

Fre

cuen

cy (%

)

Feret diameter (mm)

DAM WITH EXTRACTION

SAM WITH EXTRACTION

0%

5%

10%

15%

20%

25%

30%

35%

40%

0 0.005 0.01 0.015 0.02

Fre

cuen

cy (%

)

Feret diameter (mm)

SAM WITH EXTRACTION

DAM WITH EXTRACTION

0%5%

10%15%20%25%30%35%40%45%

0 0.005 0.01 0.015 0.02

Fre

cuen

cy (%

)

Feret diameter(mm)

SAM WITH EXTRACTION

DAM WITH EXTRACTION

0%5%

10%15%20%25%30%35%40%45%

0 0.005 0.01 0.015 0.02

Fre

cuen

cy (%

)

Feret diameter (mm)

SAM WITH EXTRACTION

DAM WITH EXTRACTION

a) b)

c) d)

e)

Fig. 3. Particle size distribution of a) DAM3:2:1 and SAM3:2:1, b) DAM3:2:2 and SAM3:2:2, c) DAM3:1:1 and SAM3:1:1,d) DAM2:2:1 and SAM2:2:1 and e) DAM2:1:1, and SAM2:1:1 after AT extraction.

306 www.rmiq.org

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

The particle size distribution changed frommonomodal to multimodal (Fig. 3) and in mostpowders, 3 peaks without any specific trend wereevident. To verify this assumption, the variousmorphometric parameters were statistically analyzed(ANOVA) with α = 0.05.

Table 3 shows morphometric parameters for theSAM and DAM materials after AT extraction. In Table4 the Feret diameter for the same materials after theextraction are presented. These results suggest thatthe stage of atomization modified the morphometriccharacteristics of the microcapsules during surface andinner free AT extraction. Although, the particle sizesof the microcapsules varied significantly, DAM, for allcases, showed larger particle sizes. For SAM, particlessizes were significantly lower than for DAM with sizesfrom 2 to 8 microns (20% decrement).

All morphometric parameters showed significantdifferences (p < 0.05) for most SAM and DAM.SAM were more irregular than DAM based onobtained values for the shape factor and in general,all microcapsules showed significant differences inat least one morphometric parameter which is a

consequence of the process and composition of thecapsule.

FDc was significantly (p ≤ 0.05) higher for DAM,as shown in Fig. 4. The FDC value in all casesconfirmed the fractal nature of the agglomerates. Notethat in this sense, the FDc is a relation betweengeometric properties (area and perimeter) and providesevidence of how smooth, rough, convoluted, tortuousor sinuous the perimeter of an object is in relationto its area covered. The results demonstrated thatFDc is apparently very sensitive to changes in oneor other of the above descriptors, which is why eachpowder had a specific value of FDc. On the otherhand, the FDc values for powders without extractabletocopherol presented a significant decrease in thisparameter for all cases. This can be explained by thereduction of the area and perimeter of each powdersince the extraction of the AT helped to deagglomeratethem (Fig. 5). It has been observed that in the caseof powders containing lipid phases, their surface isalmost completely covered by a thin layer of oil (Kimet al., 2003).

Table 3. SAM and DAM morphometric parameters after AT extraction

Microcapsules Area (µm2) Maximum perimeter (µm) Perimeter (µm) Shape factor

3:2:1 DAM 5.87E-02 ± 3.47E-03a 35.8 ± 0.0014a 29.7 ± 1.15a 0.786 ± 0.140a

SAM 1.13E-02 ±2.37E-03a 12.8 ± 1.17b 10.5 ± 0.97b 0.804 ± 0.213b

3:2:2 DAM 7.14E-02 ± 2.29E-03b 8.2 ± 1.0c 6.6 ± 0.872c 0.913 ± 0.233c

SAM 2.31E-03 ± 4.79E-03c 17.7 ± 1.71d 14.5 ± 1.43d 0.865 ± 0.195d

3:1:1 DAM 3.03E-02 ±2.33E-03d 23.8 ± 1.1e 19.7 ±0.98e 0.829 ± 0.209e

SAM 9.01E-03 ± 2.26E-03e 10.8 ± 1.07 f 8.80 ± 0.89 f 0.857 ± 0.224e

2:2:1 DAM 1.40E-02 ± 2.17E-03 f 15.3 ± 1.1g 12.7 ± 0.9g 0.835 ± 0.166e

SAM 8.10E-03 ± 2.21E-03g 9.64 ± 1.03h 7.90 ± 0.86h 0.855 ± 0.223e

2:1:1 DAM 1.03E-02 ± 2.14E-03h 12.3 ± 1.1i 10.1 ± 0.91i 0.857 ±0.209e

SAM 9.72E-03 ± 2.31E-03i 10.8 ± 1.15 j 8.82 ± 0.96 j 0.815 ± 0.254e

Superscripts with the same letter within the same column presented significant differences (p < 0.05)

Table 4. SAM and DAM Feret diameters after AT extraction

Microcapsules DAM (µm) SAM (µm)

3:2:1 8.2aA ± 2.77E-04 2.9aB ± 2.49E-043:2:2 1.9bC ± 2.35E-04 4.0bD ± 3.63E-043:1:1 5.6cE ± 2.78E-04 2.4aF ± 2.35E-042:2:1 3.5d ± 2.33E-04 2.2a ± 2.34E-042:1:1 2.8d ± 2.29E-04 2.4a ± 2.60E-04

Lowercase superscripts indicate significant difference in thesame column (p < 0.05)Uppercase superscripts indicate significant difference in thesame row (p < 0.05)

www.rmiq.org 307

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

 

 

 Figure 4. Ln (Perimeter) vs Ln (Area) for determining FDc Determination for a) DAM3:2:1 and SAM3:2:1, b) DAM3:2:2 and SAM3:2:2, c) DAM3:1:1 and SAM3:1:1, d) DAM2:2:1 and SAM2:2:1 and

e) DAM2:1:1, and SAM2:1:1

 

FDc of DAM= 1.546y = 0.546x + 2.065

R² = 0.980

FDc of SAM= 1.517y = 0.517x + 1.817

R² = 0.991

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area) DAM SAM

FDc of DAM= 1.527y = 0.527x + 1.888

R² = 0.995

FDc of SAM= 1.521y = 0.521x + 1.847

R² = 0.991

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)DAM SAM

FDc of DAM= 1.528y = 0.528x + 1.91

R² = 0.988

FDc of SAM= 1.522y = 0.522x + 1.860

R² = 0.994

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area) DAM SAM

FDc of DAM= 1.538y = 0.538x + 2.019

R² = 0.987

FDc of SAM= 1.519y = 0.519x + 1.814

R² = 0.995

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)DAM SAM

FDc of DAM= 1.519y = 0.519x + 1.860

R² = 0.991

FDc of SAM= 1.538y = 0.538x + 2.029

R² = 0.983

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area) DAM SAM

a) b)

c) d)

e)

Fig. 4. Ln (Perimeter) vs Ln (Area) for determining FDc Determination for a) DAM3:2:1 and SAM3:2:1, b) DAM3:2:2and SAM3:2:2, c) DAM3:1:1 and SAM3:1:1, d) DAM2:2:1 and SAM2:2:1 and e) DAM2:1:1, and SAM2:1:1.

It has been established that characteristics such aswettability, fluidity and agglomeration are highlyaffected by the amount of material that is encapsulatedon the surface of the individual particles (Vegaand Roos, 2006). Martins and Kieckbusch (2008)highlighted the influence of lipid phases on theagglomeration mechanism of carbohydrate basedmaterials.

3.2 Microstructural analysis of themicrocapsules

Significant differences were found in the texturefractal dimension for SAM and DAM, at the twomagnifications (500X and 2000X), after and before

the removal of the extractable AT and between allthe ratios of wall materials. Figure 6 shows allthe microcapsules images at each condition with therespective FDt and plot profile. It is importantto note that all FDt texture were higher at the2000X magnification for the powders before theextraction as can be observed in all the images.Smaller values were measured at 500X, where theagglomerations are more evident. Although a specifictendency was not found, in most of the samples thetexture fractal dimension was higher for DAM; thesedifferences might be due to size and morphologyof agglomerates and the matrix accommodation.Image analysis results provided more informationabout the relationship between wall materials andmicrostructure (Perea-Flores et al., 2010).

308 www.rmiq.org

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

 

     

 Figure 5. Ln (Perimeter) vs Ln (Area) for determining FDc for a) DAM3:2:1 and SAM3:2:1, b)

DAM3:2:2 and SAM3:2:2, c) DAM3:1:1 and SAM3:1:1, d) DAM2:2:1 and SAM2:2:1 and e) DAM2:1:1, and SAM2:1:1 after the AT extraction 

 

 

 

 

 

 

 

FDc of DAM= 1.511y = 0.511x + 1.743

R² = 0.986

FDc of SAM= 1.516y = 0.516 + 1.745

R² = 0.976

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)

DAM WITH EXTRACTION

SAM WITH EXTRACTION

FDc of DAM= 1.510y = 0.510x + 1.708

R² = 0.984

FDc of SAM= 1.523y = 0.523 + 1.843

R² = 0.995

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)

DAM WITH EXTRACTION

SAM WITH EXTRACTION

FDc of DAM= 1.505y = 0.505x + 1.630

R² = 0.989

FDc of SAM= 1.522y = 0.522 + 1.869

R² = 0.986

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)

DAM WITH EXTRACTION

SAM WITH EXTRACTION

FDc of DAM= 1.516y = 0.516x + 1.767

R² = 0.985

FDc of SAM= 1.520y = 0.520 + 1.847

R² = 0.987

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)

DAM WITH EXTRACTION

SAM WITH EXTRACTION

FDc of SAM= 1.511y = 0.511x + 1.712

R² = 0.981

FDc of DAM= 1.508y = 0.508x + 1.726

R² = 0.981

-8

-6

-4

-2

0

2

4

-18 -16 -14 -12 -10 -8 -6 -4 -2 0

Ln

(Per

imet

er)

Ln (Area)

DAM WITH EXTRACTION

SAM WITH EXTRACTIONe)

d)c)

a) b)

Fig. 5. Ln (Perimeter) vs Ln (Area) for determining FDc for a) DAM3:2:1 and SAM3:2:1, b) DAM3:2:2 and SAM3:2:2,c) DAM3:1:1 and SAM3:1:1, d) DAM2:2:1 and SAM2:2:1 and e) DAM2:1:1, and SAM2:1:1 after the AT extraction.

ConclusionsIt was possible to differentiate by using morphometricparameters, capsules obtained by doble andsingle atomized microcapsules. Double atomizedmicrocapsules had rougher surface texture andmore irregular contour than those produced byapplying only single atomization process. Moreover,extraction of surface tocopherol decreased, in allcases, the roughness of the particles by possiblyexposing cavities previously filled out with tocopherol.Homogenization processes affected quantitative and

qualitatively the shape of capsules and the form inwhich the core material is trapped.

Acknowledgements

Authors wish to thank financial support fromCONACYT and SIP-IPN-Mexico. Authors X.Quintanilla and S. Meraz acknowledge study grantsfrom CONACYT-Mexico. Also authors wish to thankDr. Jaime Vernon for useful contributions towards thiswork.

www.rmiq.org 309

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

Fractal Dimension of texture (FDt)

SAMPLE BEFORE EXTRACTION AFTER EXTRACTION

500X 2000X 500X 2000X

DAM

3:2:1

2.0473 2.2091 2.0598 2.0499

3:2:2

2.0773 2.2169 2.1183 2.0322

3:1:1

2.1664 2.1122 2.1744 2.1024

2:2:1

2.1070 2.2548 2.2060 2.1143

2:1:1

2.0699 2.1435 2.0860 2.0570

SAM

3:2:1

2.0842 2.2642 2.1260 2.0944

3:2:2

2.1050 2.2883 2.1391 2.0794

3:1:1

2.1097 2.1865 2.1615 2.0566

2:2:1

2.0820 2.2212 2.1369 2.0895

2:1:1

2.0950 2.1972 2.1295 2.1337

 

Fig. 6. ESEM images at 500X and 2000X of whole and AT extracted DAM and SAM

310 www.rmiq.org

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

ReferencesAguilera, J.M. (2007). Microstructure and food

product engineering. Food Technology 54(11),56-65.

Barleita, B.J. and Barbosa-Canovas, G. (1993).Fractal analysis to characterize ruggednesschanges in tapped agglomerated food powders.Journal of Food Science 58(5), 1030-1046.

Bellouti, M., Alves, M.M., Novais, J.M. and Mota,M. (1997). Flocs vs. granules: differentiation byfractal dimension. Water Research 31(5), 1227-1231.

Burrough, P.A. (1986). Principles of geographicalsystems for land resources assessment. In:Principles of Geographical Information Systemsfor land resources assessment. Pp. 193.Clarendon Press, Oxford.

Champagne, P. and Fustier, P. (2007).Microencapsulation for the improved deliveryof bioactive compounds into foods. CurrentOpinion in Biotechnology 18,184-190.

Chanona-Perez, J., Quevedo, R., Jimenez-Aparicio,A.R., Gumeta-Chavez, C., Mendoza-Perez,J.A., Calderon-Domınguez, G., Alamilla-Beltran, L. and Gutierrez-Lopez, G.F. (2008).Image processing methods and fractal analysisfor quantitative evaluation of size, shape,structure and microstructure in food materials.In: Food Engineering Integrated Approaches,(G. F. Gutierrez-Lopez, G. V. Barbosa-Canovas,J. Welti-Chanes, E. Parada-Arias, eds.), Pp.277-285. Springer.

Chen, W., Yuan, S., Hsiao, H. and Hsieh, C. (2003).Algorithms to estimating fractal dimension oftexture images. IEEE International Conferenceon Acoustics, Speech and Signal Processing3,1541-1544.

Drusch, S. and Berg, S. (2008). Extractable oilin microcapsules prepared by spray-drying:Localization, determination and impact onoxidative stability. Food Chemistry 109(1),17-24.

Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley,A., and Saurel, R. (2007). Applications ofspray-drying in microencapsulation of foodingredients: An overview. Food ResearchInternational 40, 1107-1121.

Jimenez, J.C., Salcedo, M.G., Martınez, B.B.,Chanona, P.J., Alamilla, B.L., Arenas, O.M.,Gutierrez, L.G. y Jimenez, A.R. 2005.Naturaleza fractal de celulas en suspension deBeta vulgaris L., sometidas a un proceso depermeabilizacion. Biotica 2, 39-53.

Kim, E., Dong, D. and Pearce, D. (2003). Onthe mechanisms of surface formation andthe surface compositions of industrial milkpowders. Drying Technology 21, 265-278.

Kim, E.H.-J., Chen, X.D. and Pearce, D. (2009).Surface composition of industrial spray-driedmilk powders. 3. Changes in the surfacecomposition during long-term storage. Journalof Food Engineering 94, 182-191

Klaypradit, W. and Huang, Y. (2008). Fish oilencapsulation with chitosan using ultrasonicatomizer. LWT -Food Science and Technology41, 1113- 1139.

Lekago, J. and Dunford, N. L. (2010). Effectof Spray Nozzle Design on Fish Oil-WheyProtein Microcapsule Properties. Journal ofFood Science 75(6), E94-E97.

Martins, P. and Kieckbusch, T. (2008). Influenceof lipid phase on steam jet agglomeration ofmaltodextrin powders. Powder Technology 185,258-266.

Mery, D. and Pedreschi, F. (2005) Segmentation ofcolour food images using a robust algorithm.Journal of Food Engineering 66, 353-360.

Murillo-Martınez, M.M., Pedroza-Islas, R., Lobato-Calleros, C., Martınez-Ferez, A. and Vernon-Carter, E.J. (2011). Designing W1/O/W2double emulsions stabilized by proteinpolysaccharide complexes for producing ediblefilms: Rheological, mechanical and watervapour properties. Food Hydrocolloids 25, 577-585.

Pedreschi, F., Mery, D., Mendoza, F. and Aguilera,J.M. (2004). Classification of potato chips usingpattern recognition. Journal of Food Science 69,E1-E7.

Perea-Flores, M.J., Chanona-Perez, J.J., Terres-Rojas, E., Calderon-Dominguez, G., Garibay-Febles, V., Alamilla-Beltran, L. and Gutierrez-Lopez, G.F. (2010) Microstructure structurecharacterization of milk powders and their

www.rmiq.org 311

M.X. Quintanilla-Carvajal et al./ Revista Mexicana de Ingenierıa Quımica Vol. 10, No. 2 (2011) 301-312

relationship with rehydration properties. In:Spray Drying Technology, (M.W. Woo, A.S.Mujumdar y W.R.W Dau, eds.), Pp 197-219.Editorial Transport Processes Research (TPR),Singapore.

Perez-Alonso, C., Cruz-Olivares, J., Barrera-Pichardo, J., Rodrıguez-Huezo, M., Baez-Gonzalez, J. and Vernon-Carter, E. (2008). DSCthermo-oxidative stability of red chili oleoresinmicroencapsulated in blended biopolymersmatrices. Journal of Food Engineering 85, 613-624.

Perez-Alonso, C., Fabela-Moron, M.F., Guadarrama-Lezama, A.Y., Barrera-Pichardo, J. F., Alamilla-Beltran, L. and Rodrıguez-Huezo, M.E. 2009.Interrelationship between the structural featuresand rehydration properties of spray driedmanzano chilli sauce microcapsules. RevistaMexicana de Ingenierıa Quımica 8(2),187-196.

Pitalua, E., Jimenez, M., Vernon-Carter, E.J. andBeristain, C.I. (2010). Antioxidative activityof microcapsules with beetroot juice using gumarabic as wall material. Food and BioproductsProcessing 88, 253-258.

Pulido, A. and Beristain, C.I. 2010. Spray driedencapsulation of ascorbic acid using chitosan aswall material. Revista Mexicana de IngenierıaQuımica 9(2), 189-195.

Tapia-Ochoategui, A.P., Camacho-Dıaz, B.H., Perea-Flores, M.J., Ordonez-Ruiz, I.M., Gutierrez-Lopez, G.F. and Davila-Ortiz, G. 2011.Morfometric changes during the traditionalcuring process of vanilla pods (Vanillaplanifolia; Orchidaceae) in Mexico. RevistaMexicana de Ingenierıa Quımica 10(1), 105-115.

Vega, C. and Roos, Y.H. (2006). Invited review:spray-dried dairy and dairy-like emulsionscompositional. Journal of Dairy Science 89(2),383-401.

Vignolles, M. L., Lopez, C., Madec, M. N., Ehrhardt,J. J. Mejean, S., Schuck, P. and Jeantet, R.(2009). Fat properties during homogenization,spray-drying, and storage affect the physicalproperties of dairy powders. Journal of DairyScience 92, 58-70.

Woo, M.W., Mujumdar, A.S. and Dau, W.R.W.(2010). Spray Drying Technology. EditorialTransport Processes Research (TPR),Singapore.

Ye, X., Fernando, S., Wilson, W. and Singh,A. (2007). Application of amphiphiliccatalysts, ultrasonication, and nanoemulsionsfor biodiesel production process. ChemicalEngineering and Technology 30(11), 1481 -1487.

312 www.rmiq.org