Effect of extrusion conditions on physicochemical characteristics and anthocyanin content of blue corn third-generation snacks

Abstract

The aim of this study was to evaluate the effect of barrel temperature (BT, 98.8–141.2°C) and feed moisture (FM, 19.93–34.07%) as independent factors on physicochemical characteristics of microwave-expanded extruded third-generation (3G) snacks obtained from blue corn and corn starch. Single-screw laboratory extruder and a central, composite, rotatable experimental design were used. Both independent factors showed significance (p ≤ 0.01) on most of the analyzed responses. The mathematical models showed values of R²_Adj ≥ 0.76 and p of F_(model) ≤ 0.001. The optimum area of the extrusion process ranged from 120°C to 126°C for BT and from 23.8% to 25.2% for FM. In optimal conditions, the product showed an expansion index of 4.8, a penetration force of 12.42 N, a specific mechanical energy of 169.08 kJ/kg, and 71.09 mg of total anthocyanin content/kg. The developed 3G snack presented high-quality physicochemical characteristics, with the potential health benefits derived from nutraceutical characteristics (dietary fiber and anthocyanins) of the whole blue corn used.

El objetivo de este estudio fue evaluar el efecto de la temperatura de barril BT (98,8–141,2 °C) y la humedad de alimentación FM (19,93–34,07%) como factores independientes, sobre características fisicoquímicas de botanas extrudidas, de tercera generación (3G), expandidas por microondas, obtenidas a partir de maíz azul y almidón de maíz. Se utilizó un extrusor de laboratorio de tornillo simple y un diseño experimental central compuesto, rotable. Ambos factores independientes mostraron significancia (p ≤ 0,01) en la mayoría de las respuestas analizadas. Los modelos matemáticos mostraron valores de R²_aj ≥ 0,76 y p de F_(modelo) ≤ 0,001. La zona óptima para el proceso de extrusión varió de 120–126 °C de BT y de 23,8–25,2% de FM. En condiciones óptimas el producto mostró un índice de expansión de 4,8, una fuerza de penetración de 12,42 N, una energía mecánica específica de 169,08 kJ/kg y 71,09 mg de contenido total de antocianina/kg. La botana 3G desarrollada presentó características físico-químicas de alta calidad, con los beneficios potenciales para la salud, derivados de las características nutracéuticas (fibra dietaria y antocianinas) del maíz azul integral utilizado.

Keywords:

• extrusion
• blue corn
• third-generation snack
• anthocyanin

Palabras clave:

• extrusión
• maíz azul
• botana de tercera generación
• antocianina

Introduction

A snack is defined as a small, lightweight food that is easy to manipulate, ready to eat, accessible, and, most importantly, able to satisfy the appetite sensation for a moment (Hurtado, Escobar, & Estévez, 2001). Snack foods are widely consumed, regardless of social status, age, or gender. The industrial sector of snacks in Mexico is booming with an annual market value of 3419 million dollars, offering various kinds of snacks, mainly, the potato and corn (dough and tortilla) derivatives (http://inegi.gob.mx). Among the main types of snacks are the third-generation (3G) snacks, also known as intermediate snacks or pellets, which are cheap and easy to prepare at home (Hollingsworth, 2001). In the processing of 3G snacks, the dry ingredients are mixed with water (22–35%) to form a dough. The 3G snacks are prepared by extrusion, formed at low pressure to avoid expansion, and dried to a final moisture content of 10–14% to form a glassy pellet. The extrusion process consists of a 3-step temperature profile, starting with a low-temperature step at the feed zone (70–80°C), continuing with a high-temperature step at the mixing and cooking zone (90–145°C), and ending with a low-temperature step at the output die (75–95°C) (Bastos-Cardoso, Zazueta-Morales, Martínez-Bustos, & Yoon, 2007; Delgado-Nieblas et al., 2012). The 3G snacks have a long shelf life, being capable of retaining a good quality for at least one year, provided that a proper storage is given. As pellets, they require less storage space due to their less volume in relation to their size after expanding when compared with directly expanded snacks (Arias-García et al., 2007). However, they require a further expansion process, which may be done by hot oil, hot air, or microwave exposure. During this latter intensive heating step, the moisture in the pellet will start to boil and vapor bubbles are formed, which will expand the pellet. The expansion gives the snack a porous structure (Boischot, Moraru, & Kokini, 2003). Taking advantage of the greater consumer acceptance for 3G snacks, they can be used as nutrient carriers in order to offer an added value product with high nutritional/nutraceutical properties. Balasubramanian, Borah, Singh, and Patil (2012), Delgado-Nieblas et al. (2012), Limón-Valenzuela, Martínez-Bustos, Aguilar-Palazuelos, Caro-Corrales, and Zazueta-Morales (2010) have reported studies aiming this purpose, where milk proteins, legume seeds, and vegetable flour have been used.

Blue corn composition is similar to white corn, with the advantage of containing anthocyanin and phenolic compounds (Pedreschi & Cisneros-Zevallos, 2007; Yan & Zhai, 2010). These are phytochemicals that are synthesized in plants by secondary metabolism, and there is great interest in them because of their antioxidant and bioactive properties. Their consumption has been correlated with health benefits, chronic, and degenerative illness prevention, such as cancer, cardiovascular diseases, and cataracts (He & Giusti, 2010).

The extrusion process has become very important in food processing because of its extensive technical advantages of cost, in addition to being a high-temperature short-time process, which allows for less destruction of heat-sensitive components (White, Howard, & Prior, 2010). Furthermore, extrusion technology has been successfully used in the production of both directly (second generation) or indirectly expanded snack (3G). Studies have been conducted about the addition of anthocyanin to directly expanded extruded foods. Khanal, Brownmiller, Howard, and Prior (2009) studied an extrusion process (temperature 160–180°C and screw speed 150–200 rpm) using mixtures of grape seed, cranberries, and white sorghum flour. In this report, the extrusion process reduced the anthocyanin content up to 42% at the high-temperature range. Camire, Chaovanalikit, and Dougherty (2002) reported a total anthocyanin loss, in the range of 64–90% caused by the extrusion process, when concentrated cranberry was incorporated into extruded breakfast cereals. Zazueta, Martínez, Jacobo, Ordorica, and Paredes (2001) studied the effect of the addition of calcium hydroxide on some characteristics of extruded directly expanded blue corn. They found that it is possible to obtain an extruded directly expanded blue corn product, fortified with calcium. However, the scientific literature on the development of blue corn-based 3G snacks, and the effect of extrusion processing on their physicochemical, structural, and nutritional properties, is scarce. The aim of this study was to evaluate the effect of extrusion variables on physicochemical characteristics and anthocyanin content of 3G snacks elaborated from blue corn flour and corn starch.

Materials and methods

Materials

Blue corn (Zea mays L.), “Chalqueño” race, from the local market of Pachuca, Hidalgo, and corn starch produced by IMSA (Industrializadora de maíz S.A de C.V. Puebla, México) were used.

Preparation of samples

Integral flour (≤250 μm) from blue corn was obtained using a hammer mill (Mini 100 Pulvex S.A de C.V. México) and was mixed with corn starch in a 65:35 proportion. The criteria used for the selection of the mixture was to obtain a product with high expansion, maximum blue corn content, and minimum addition of corn starch, which was established in preliminary studies. The mixture was added to 0.1% of monoglycerides (Bioproceso Company, Culiacán, México). The addition of emulsifiers, such as saturated monoglycerides, contributes to the lubricant effect, which decreases the mechanical degradation produced by the extruder and improves the texture of the cooked product (Bastos-Cardoso et al., 2007). This monoglyceride concentration was chosen because, in a preliminary study, it showed desirable effect for 3G snacks. The prepared mixtures were homogenized at medium speed (~8 min) in a laboratory mixer (Kitchen Aid, model K5SS, Michigan, USA); they were stored in sealed polyethylene bags and kept under refrigeration (8–10°C) for 12 h before processing.

Extrusion process

The extrusion process was performed using a single-screw laboratory extruder (Brabender 20DN, model 8-235-00, Duisburg, Germany). A rectangular aperture output die with internal measures of 20 mm wide × 1.0 mm high × 100 mm long, a screw with 2:1 compression ratio, a screw speed of 80 rpm, and feed rate of 2.8–3.0 kg/h, were used. Temperatures in both the feed zone and in the output die were 75°C, while temperatures of the intermediate zone (mixing/cooking zone) and feed moistures varied according to the experimental design (Table 1). The extruded materials (pellets) were manually cut into approximately 2.5 cm long strips, dried at room temperature (24–26°C, 50–70% relative humidity) for 48 h, up to a moisture content of about 9–13%, stored in sealed plastic bags, and kept in darkness under refrigeration (6–8°C) until analysis.

Table 1. Experimental design used for blue corn extrusion.

Tabla 1. Diseño experimental utilizado para la extrusión de maíz azul.

Treatment	Independent variables
	Coded		Actual
	X1	X2	BT (°C)	FM (%)
1	−1	−1	105	22
2	1	−1	135	22
3	−1	1	105	32
4	1	1	135	32
5	−1.414	0	98.79	27
6	1.414	0	141.21	27
7	0	−1.414	120	19.93
8	0	1.414	120	34.07
9	0	0	120	27
10	0	0	120	27
11	0	0	120	27
12	0	0	120	27
13	0	0	120	27

Note: BT = Barrel temperature (°C); FM = Feed moisture (%).

Nota: BT = Temperatura de barril (°C); FM = Humedad de alimentación (%).

Expansion pellets

The cut and extruded pellets were expanded using a commercial microwave oven (LG^®, R-501CW, Monterrey México, 900 W and 2450 Hz) for 26 s, according to preliminary tests.

Proximate analysis

Official methods of AOAC (1999) were used to analyze moisture (925.09), protein (979.09), ash (923.03), lipids (923.05), and fiber (962.09). The carbohydrate content was calculated by difference.

Expansion index (EI) and bulk density (BD)

Expansion index (EI) and bulk density (BD) were determined using the specific volume of non-expanded (Vnep) and expanded (Vep) pellets as EI = (Vep–Vnep)/Vnep. The specific volume was determined using the seed displacement test, according to Penfield and Campbell (1990) and Boischot, Moraru and Kokini (2003). Results were the mean of 30 determinations by treatment.

Penetration force (PF)

Penetration force (PF) was measured in microwave-expanded products with a penetrometer (Chatillon, model TCD 200, Surrey, UK). A 2 mm diameter flat-tip probe with a penetration speed of 0.8 mm/s was used. The required force (N) to penetrate a depth of 3 mm was registered, with 30 replicates for treatment.

Specific mechanical energy (SME)

The energy required for extruder screw rotation (kJ/kg) was calculated from values of torque (t, N·m), screw extruder speed (ss, rpm/min), and feed flow (F, kg/h), according to Batterman-Azcona, Lawton, and Hamaker (1999).

Water absorption index (WAI) and water solubility index (WSI)

They were performed in the microwave-expanded products using 2.5 g of sample according to Anderson, Conway, Pfeifer, and Griffin (1969), where the quantity of dissolved material and the proportion of absorbed water are gravimetrically calculated after stirring a suspension at room temperature.

Total anthocyanin content (TAC)

Total anthocyanin content (TAC) was measured in raw materials and microwave-expanded snacks, using the method described by Abdel-Aal and Hucl (1999). TAC was expressed in mg of cyanidin-3-glucoside/kg (db).

Experimental design

A central, composite, rotatable experimental design for response surface methodology, with 13 treatments and value α = 1.414 (Table 1), was used.

The response surface superposition methodology was utilized to find the optimal processing conditions, in order to obtain a high-quality and high expanded product, using the Design-Expert software (7.0). Pearson correlations were performed using Statistica 7.0 software (Stat-Ease, Inc., Minneapolis, MN).

Results

Proximate composition

Table 2 shows the proximate composition of blue corn, corn starch, and the mixture of blue corn flour and corn starch. The proximate composition of blue corn is consistent with values reported by Zazueta-Morales et al. (2001) and Escalante-Aburto et al. (2013), who used blue corn for developing a directly expanded second-generation snack. The results are also consistent with those reported by Nava, Jimenez, and Hernández (2008) and Hoover and Manuel (1996) for blue corn and corn starch.

Table 2. Proximate analysis of raw materials.

Tabla 2. Análisis proximal de materias primas.

Component (g/100g dry base)	BCF	CS	Mixture^1 (BCF + CS)
Protein	8.91 ± 0.27	0.09 ± 0.06	5.8 ± 0.15
Fat	5.79 ± 0.03	0.18 ± 0.01	3.82 ± 0.03
Ash	1.65 ± 0.05	0.02 ± 0.01	1.08 ± 0.03
Fiber	2.32 ± 0.04	−	1.51 ± 0.02
Carbohydrates*	81.30 ± 0.31	99.70 ± 0.01	87.79 ± 1.16

Note: ¹65% BCF + 35% SC (db); BCF = Blue corn flour; CS = Corn starch.*Calculated by difference.

Nota: ¹65% BCF+ 35%SC(bs); BCF = Harina de maíz azul; CS = Almidón de maíz.*Calculado por diferencia.

Regression coefficients and ANOVA

The Regression coefficients for the responses analyzed are shown in Table 3. Both factors, barrel temperature (BT) and feed moisture (FM), showed a highly significant effect (p ≤ 0.01) in their linear (b₁) and quadratic (b₁₁) terms on the majority of the responses studied, except b₂ and b₂₂ to WSI and b₂₂ for BD. Furthermore, for TAC in the microwave-expanded extruded snacks, only the terms b₂₂ and b₂₂b₁ of the mathematical model had a significant effect. The interactions terms of the models, in general, were significant on various responses. Additionally, Table 4 shows the analysis of variance (ANOVA) for the analyzed response variables. The models were accurate enough for all responses, with values of R²_Adj > 0.76, p of F_(model) < 0.001, and variability coefficient (VC) < 14.78 (except for PF, 20.41). However, it can be seen that BD, SME, and WAI showed lack of fit (p ≥ 0.082).

Table 3. Regression coefficients of the models and significance levels for analyzed responses.

Tabla 3. Coeficientes de regresión de los modelos y niveles de significancia para las respuestas analizadas.

Coefficients
Response	Intercept	Linear		Quadratic		Interaction
	b0	b1	b2	b11	b22	b12	b11b2	b22b1
EI	−314.39	+5.18 (0.01)*	+9.54 (<0.01)	−0.02 (<0.01)	−2.73E-3 (0.03)	−0.15 (<0.01)	6.17E-4 (<0.01)
BD	+237,07.33	−355.82 (<0.01)	−774.03 (<0.01)	+10.64 (<0.01)	+1.33 (0.074)	+10.64 (<0.01)	−0.035 (<0.01)	−0.033 (<0.01)
PF	+3551.98	−56.99 (<0.01)	−117.75 (<0.01)	+0.23 (<0.01)	+0.06 (<0.01)	+1.87 (<0.01)	−7.62E-3 (<0.01)
SME	+3524.49	+17.44 (<0.01)	−269.41 (<0.01)	−0.39 (<0.01)	+8.22 (<0.01)	+0.13 (<0.01)	0.02 (<0.01)	−0.07 (<0.01)
WAI	−166.87	+2.33 (<0.01)	+6.97 (0.145)	7.29E-3 (<0.01)	−0.08 (0.19)	−0.08 (0.07)	1.70E-4 (0.08)	6.91E-4 (0.03)
WSI	−1015.84	+18.16 (<0.01)	+30.69 (<0.01)	−0.07 (<0.01)	+0.09 (<0.01)	−0.59 (<0.05)	0.09 (<0.01)	2.42E-3 (<0.01)
TAC	+4449.51	−44.17 (0.87)	−310.23 (0.57)	+0.039 (0.11)	+5.29 (<0.01)	+3.028 (0.47)	−1.82E-3 (0.41)	−0.048 (<0.01)

Note: IE= Expansion index; BD = Bulk density; PF = Penetration force; SME = Specific mechanical energy; WAI = Water absorption index; WSI = Water solubility index; TAC = Total anthocyanin content; b₀, b₁, … b_n = Regression coefficients.*Number in parentheses are p values

Nota: IE = Índice de expansión; BD = Densidad aparente; PF = Fuerza de penetración; SME = Energía mecánica específica; WAI = Índice de absorción de agua; WSI = Índice de solubilidad en agua; TAC = Contenido de antocianinas totales; b₀, b₁, … b_n = Coeficientes de regresión.*Número entre paréntesis corresponde a los valores p.

Table 4. Variance analysis for analyzed responses.

Tabla 4. Análisis de varianza para las respuestas analizadas.

Response	R^2 adjusted	CV (%)	F value	p of F(model)	Lack of fit
EI	0.84	11.49	201.98	<0.001	<0.001
BD	0.90	8.24	269.35	<0.001	0.082
PF	0.76	20.41	207.11	<0.001	<0.001
SME	0.99	2.24	387.68	<0.001	0.089
WAI	0.96	3.01	41.48	<0.001	0.990
WSI	0.87	14.78	43.77	<0.001	<0.001
TAC	0.86	7.80	27.84	<0.001	0.029

Note: CV = Coefficient of variation; IE = Expansion index; BD = Bulk density; PF = Penetration force; SME = Specific mechanical energy; WAI = Water absorption index; WSI = Water solubility index; TAC = Total anthocyanin content.

Nota: CV = Coficiente de variabilidad; IE = Índice de expansión; BD = Densidad aparente; PF = Fuerza de penetración; SME = Energía mecánica específica; WAI = Índice de absorción de agua; WSI = Índice de solubilidad en agua; TAC = Contenido de antocianinas totales.

Expansion index (EI)

The effect of BT and FM on expansion index (EI) is shown in Figure 1. It can be seen that in almost the entire experimental interval of FM, EI increased with increasing BT from 98°C to 125°C. BT higher than 125°C favored a decrease in values of EI. Furthermore, at low BT (<115°C), EI values increased as FM increased, and the highest EI values were obtained at low FM (~20%) and intermediate BT (120–125°C). These results are in agreement with Moraru and Kokini (2003), who mention that the expansion usually occurs at high BT and low FM as a result of several events such as structural transformations of biopolymers, phase transitions and nucleation, swelling, growth, and collapse of air bubbles, all of them contributing to the expansion phenomenon. Furthermore, Lee, Lim, Lim, and Lim (2000) observed that starch gelatinization degree and the moisture content of the pellets are two important factors in determining the shape, bulk density, and expansion of microwave-expanded products. These authors found higher expansion values when BT ranged from 90°C to 110°C, in comparison with expansion values obtained at low BT (70°C), in microwave-expanded pellets made from corn starch. They attributed the high EI to the starch gelatinization degree of the high BT obtained samples, and they found the highest expansion at a gelatinization percentage of ~50%. These conditions allowed for the easier formation of air cells and later expansion by increasing the steam pressure without rupture of its cellular structure. Aguilar, Zazueta, and Martínez (2006) found a similar pattern to that reported in this study for microwave-expanded pellets, elaborated from potato starch, quality protein maize and soybean flour mixtures. Our results also were in agreement with Delgado-Nieblas et al. (2012), who reported that EI increased concomitantly with BT increment. From our results also, it may be theorized that an increase in BT above 120°C might induce a higher level of starch granules breakdown, leading to a reduction of viscosity and of extruder residence time. It is also possible to consider that high BT would lead to a decreased viscosity due to the heating effect and thus the severity of the process could be reduced, as suggested by Chang, Martínez-Bustos, Park, and Kokini (1999).

Figure 1. Effect of barrel temperature (°C) and feed moisture (%) on expansion index (EI) of blue corn microwave-expanded snack.

Figura 1. Efecto de la temperatura de barril (°C) y la humedad de alimentación (%), sobre el índice de expansión (EI) de botanas de maíz azul expandidas por microondas.

The expansion of the extruded products is enhanced by an increase in temperature up to a peak and decreases thereafter. This is due to the physicochemical changes in starch-protein systems induced by temperature increment (Amaya-Llano, Morales-Hernández, Castaño-Tostado, & Martínez-Bustos, 2007; Moraru & Kokini, 2003). The temperature of maximal expansion is dependent on the ingredients being used. The decrease in EI may also be related to dietary fiber content of the mixture.

In our study, EI ranged from 1.91 to 4.80 (a commercial sample showed an EI of 4.75). These values are close to those reported by several authors for 3G microwave-expanded snacks (Bastos-Cardoso et al., 2007; Delgado-Nieblas et al., 2012), and the differences that were found in BT for maximum expansion of the pellets might be attributed to raw material formulations as well as to different types of extruders used.

Bulk density (BD)

Increasing BT from 98°C to ~125°C resulted in a decreased BD throughout the whole FM experimental interval. Above 125°C, the BD of the expanded products tends to increase. Furthermore, it can be seen that at lower BT (<110°C), BD decreased when FM increased. The values of BD in this study ranged from 178.2 to 428.6 kg/m³ (a commercial sample showed a BD of 130 kg/m³). The lowest values of BD of expanded pellets were shown at BT ~125°C, in about the whole FM range studied (Figure 2). In this work, an inverse behavior can be observed between BD and EI. The BD showed a high negative Pearson-correlation with EI (r = −0.95, p < 0.05). According to Ramírez-Ascheri, Ciacco, Ríaz, and Lusas (1995), BD of 3G extruded products is inversely related to expansion degree and starch gelatinization degree, thereby lower BD values corresponded with higher EI values. These authors correlated extrusion variables with starch gelatinization degree, finding that the best product, with the lowest density, showed approximately 50% gelatinization. Aguilar-Palazuelos et al. (2006) elaborated a 3G snack, and they found that BD decreased by an interaction effect between BT and FM. Several authors have reported that when the extrusion temperature is increased, BD values tend to decrease, which is attributed to starch degradation as an effect of thermal process (Altan, McCarthy, & Maskan, 2008).

Figure 2. Effect of barrel temperature (°C) and feed moisture (%) on bulk density (BD) of blue corn microwave-expanded snack.

Figura 2. Efecto de la temperatura de barril (°C) y la humedad de alimentación (%), sobre la densidad aparente (BD) de botanas de maíz azul expandidas por microondas.

Özer, Ibanoglu, Ainsworth, and Yagmur (2004) reported that by decreasing the moisture content the BD decreased, while in the present study the effect of FM was significant only at BT lower than 110°C. Ding, Ainsworth, Plunkett, Tucker, and Marson (2006) found that as FM in the samples increased, BD values of the expanded snack products elaborated from wheat also increased. These authors attributed this behavior to a smaller starch gelatinization, leading to a lower EI and a high BD. In addition, Stojceska, Ainsworth, Plunkett, and İbanoğlu (2009) reported a similar behavior in extruded snacks made from wheat flour and corn starch added with red cabbage and by-products from brewing beer process.

Penetration force (PF)

Figure 3 shows the effect of BT and FM on PF. It can be seen that in the range from 18–26% of FM, PF decreased by increasing BT from 98°C to ~120°C. Above that temperature, PF tended to increase. On the other hand, at low BT (<110°C) and high BT (>130°C), as FM increased, PF decreased. The lowest experimental value of PF showed in this study was 8 N; this value was higher than the PF of a commercial sample (3.52 N). It has been documented that the incorporation of dietary fiber in the extruded products significantly reduces expansion volumes and increases density of extruded products, leading to harder textures (Robin, Schuchmann, & Palzer, 2012). The high PF values reported in the present study, could be due to a relatively high dietary fiber contained in whole blue-corn flour used. In this work, PF showed moderate negative Pearson correlation with EI (r = −0.63, p < 0.05) and an important negative correlation with BD (r = 0.71, p < 0.05), thereby the low PF values corresponded with higher EI values and lower BD values. It is also observed that PF is a dependent variable of BT and FM, as reported by Martinez-Bustos et al. (1998). These behaviors can be related to starch gelatinization degree and the dietary fiber content present in the corn pericarp and interactions between lipids and proteins that make up the corn grain. These findings are consistent with those reported by Arias-Garcia et al. (2007), who reported on products made from mixtures of wheat flour and corn starch. Those with higher EI and lower BD were the softer products. PF is a strength that reflects the resistance of bubble walls to be broken, which depends on the number of bubbles formed per volume unit and on the resistance of the formed cell-wall-type structures which are thinner as EI is increased (Pérez-Navarrete, Cruz-Estrada, Chel-Guerrero, & Betancur-Ancona, 2006). Several authors have reported a relationship between PF and EI of extruded products, since softer products have a higher EI (Delgado-Nieblas et al., 2012; Hsieh, Mulvaney, Huff, Lue, & Brent, 1989).

Figure 3. Effect of barrel temperature (°C) and feed moisture (%) on penetration force (PF) of blue corn microwave-expanded snack.

Figura 3. Efecto de la temperatura de barril (°C) y la humedad de alimentación (%), sobre la fuerza de penetración (PF) de botanas de maíz azul expandidas por microondas.

The extruded pellets, after expansion, can acquire a volume 2–9 times the original size (Mercier & Feillet, 1975). This expansion is related to the fragility of the piece being chewed. When the elaboration of the pellets is not within technological patterns, its expansion is very low and the product is very hard (Ramírez-Ascheri & Carvalho-Wanderley, 1997).

Specific mechanical energy (SME)

Figure 4 shows the effect of BT and FM on the SME. It is observed that with increasing FM at low temperatures (~<110°C), SME decreased. This is due to a reduced shear force and mechanical energy input. On the other hand, at low FM (<25%), BT increased and SME decreased. These results agree with those reported by Chang et al. (1999), who reported that an increase in BT causes a decrease in the SME. This was attributed to the increased temperature in the extrudate by increasing BT, which resulted in a decrease of the material viscosity and the required equipment energy. In addition, various authors have reported that an increase in FM during extrusion resulted in a decrease in material viscosity and in the required SME for the process (Rosentrater, Muthukumarappan, & Kannadhason, 2009; Singh, Smith, & Frame, 1998). In this study, SME showed a moderate positive correlation with PF (r = 0.58, p < 0.05). SME value is indicative of the extrusion process severity, and it has been reported that this parameter is correlated with properties of extruded products such as EI, BD, and PF (Altan et al., 2008; Onwulata, Konstance, Smith, & Holsinger, 2001). Furthermore, it has been reported that SME is dependent on the process parameters such as FM, BT, screw speed, and feed rate, FM being the most significant factor (Ding, Ainsworth, Tucker, & Marson, 2005).

Figure 4. Effect of barrel temperature (°C) and feed moisture (%) on specific mechanical energy (SME) for the production of blue corn third-generation snack.

Figura 4. Efecto de la temperatura de barril (°C) y humedad de alimentación (%), sobre la energía mecánica específica (SME) para la producción de botanas de tercera generación, elaboradas de maíz azul.

The starch gelatinization degree varies with the SME applied by the extruder. High SME facilitates intermolecular rupture of hydrogen bonds, and the hydrophilic groups of starch are exposed to water, thereby the gelatinization is favored (Gropper, Moraru, & Kokini, 2002). The SME decrement by moisture effect is mainly due to the lubricating effect of water. FM plays an important role in controlling the extrusion process because of its impact on mixing, on viscosity, and on the retention time of the dough in the extruder barrel. Additionally, it has been reported that a reduction in the FM at low BT increases the SME required for the extrusion process, which is explained by the high viscosity of the dough processed under these conditions (Ryu, 2001). This behavior is consistent with the data obtained in this study. The experimental values of SME for the different treatments of the present study were within the range of 80.4–316.7 kJ/kg. SME values for different materials processed by extrusion ranged from 160 to 2108 kJ/kg (Bastos-Cardoso et al., 2007; Delgado-Nieblas et al., 2012; Gropper et al., 2002). It has been documented that, in efficiency terms, the SME values for extrusion process must be lower than 1000 kJ/kg.

Water absorption index (WAI)

The effect of BT and FM on WAI is shown in Figure 5. It can be seen that throughout the studied FM range WAI increased, reaching its highest level at approximately 130°C and decreasing thereafter. The highest WAI increment rate was shown between the range of 20–27% of FM. This effect could be explained by the rearrangement in the starch structure, which facilitates water absorption. Some researchers have suggested that when the temperature is increased in the presence of moisture, the chains of amylose and amylopectin are separated and form an extended matrix, which results in a higher water absorption capacity (Balandrán-Quintana, Barbosa-Cánovas, Zazueta-Morales, Anzaldúa-Morales, & Quintero-Ramos, 1998; Colonna, Tayeb, & Mercier, 1989; Kokini, Lai, & Chedid, 1992).

Figure 5. Effect of barrel temperature (°C) and feed moisture (%) on water absorption index (WAI) of blue corn microwave-expanded extruded products.

Figura 5. Efecto de la temperatura de barril (°C) y humedad de alimentación (%), en el índice de absorción de agua (WAI) de productos extrudidos de maíz azul, expandidos por microondas.

Water absorption index determines the quantity of water (in grams) that is bound to one gram of dry sample and indicates the integrity of the starch in an aqueous dispersion (Anderson et al., 1969; Mason & Hoseney, 1986). Water absorption depends on the availability of hydrophilic groups, which bind water molecules, and on the gel-forming capacity of macromolecules (Gomez & Aguilera, 1983).

When damaged, starch granules are capable of absorbing a great deal of water at room temperature and swell resulting in increased viscosity (Colonna et al., 1989). One of the most important phenomena on the extruded food starch is gelatinization, the conversion of raw starch into a cooked and digestible material by the application of heat and water. The water is absorbed and bound to the starch molecule, causing a change in the starch granule structure. The temperature and moisture are the factors that exert an important effect on gelatinization. According to Lawton (1972), maximum degree of gelatinization can occur in both conditions, high moisture and low temperature or low moisture and high temperature.

Furthermore, the soluble starch increased with increasing extrusion temperature and decreasing FM (Mercier & Feillet, 1975). Agustiniano-Osornio et al. (2005) indicated that the extrusion process produces a complete starch gelatinization at low FM, when BT is between 110°C and 135°C. In the present study, at BT higher than 130°C, WAI decreases. This effect may be attributed to the fact that an increase in BT favors the material fluidity, decreasing the residence time within the extruder, and enables the retention of the starch granular structure, as suggested by values of SME found in this study. Also, in this study, WAI presented a moderate positive correlation with BT (r = 0.58, p < 0.05). An increase in BT produces an increase in viscosity of the dough that is fed into the extruder and the time required to pass through the extruder barrel is increased, causing a greater mechanical damage in the starch granule. In the same way, this response (WAI) showed an important correlation with EI (r = 0.71, p < 0.05), with BD (r = −0.82, p < 0.05), with PF (r = −0.76, p < 0.05), and with SME (r = −0.69, p < 0.05). Lee, Ryu, and Lim (1999) reported that WAI was affected by BT, FM, and screw speed (in rpm). These authors found that WAI rapidly increased with increasing BT up to 90°C, decreasing thereafter. Similar results have been reported by different authors (Ding et al., 2005; Lee et al., 1999). The increase in WAI as BT increases may be due to high temperatures coupled with shear force generated by the extruder screw during the process, leading to starch degradation, which produces fragmented granules that absorb water at room temperature.

The WAI reaches a maximum value and then decreases as a result of the high starch dextrinization degree, due to the thermal–mechanical damage suffered during the extrusion process (Linko, Vuorien, & Linnko, 1980). For high-moisture samples combined with high processing temperatures, the level of starch degradation is low, consequently, WAI decreases (Carvalho, Ramírez-Ascheri, & Cal-Vidal, 2002). In this study, the experimental values obtained for WAI showed a maximum of 5.18 g aw/g ds at BT = 135°C and FM = 32%, while the minimum value (2.85 g aw/g ds) was obtained at BT = 98.79°C and FM = 27%.

Water solubility index (WSI)

The effect of extrusion parameters on WSI is shown in Figure 6. At FM lower than 30%, WSI increased as BT increased, peaking approximately at 120°C. This may be due to the starch degradation, which causes a rise in WSI as a result of a reduction in the molecular size of starch fragments. WSI showed a moderate negative correlation with FM (r = −0.62, p < 0.05). This correlation may be due to the fact that an increase in FM reduces friction of the dough in the extruder, so the material fragmentation is limited. Furthermore, the lubricating effect supplied by the water causes the sample to pass faster through the extruder, and the shearing effect of the BT and the extruder screw is not high enough to degrade the starch in high levels, obtaining a lower WSI as a result. WSI also showed a moderate positive correlation with EI (r = 0.71, p < 0.05), a moderate negative correlation with BD (r = −0.60, p < 0.05), and a moderate negative correlation with WAI (r = 0.57, p < 0.05). At low moisture content, SME increased. WSI is related to the amount of soluble solids in a dry sample, allowing for the verification of the severity of the extrusion process, which depends on the degradation, gelatinization, and dextrinization of starch (Carvalho et al., 2002; Yang, Peng, Lui, & Lin, 2008). In this study, the highest WSI value (20.47%) was shown at BT = 120°C and FM = 20%, while the lowest WSI value (5.46%) was obtained at BT = 99°C and FM = 27%. These results agree with those reported by several authors (Colonna & Mercier, 1983; Colonna, Doublier, Melcion, De Monredon, & Mercier, 1984; Gomez & Aguilera, 1983), who found that at low FM and high BT the water solubility of the materials is increased, and the viscosity decreases with respect to raw materials or extruded materials at high FM and low BT.

Figure 6. Effect of barrel temperature (°C) and feed moisture (%) on water solubility index (WSI) of blue corn microwave-expanded extruded products.

Figura 6. Efecto de la temperatura de barril (°C) y humedad de alimentación (%), el índice de solubilidad en agua (WSI) de productos extrudidos de maíz azul, expandidos por microondas.

Balandrán, Barbosa, Zazueta, Anzaldúa, and Quintero (1998) reported that a reduction in FM, in conjunction with high BT, causes severe thermal–mechanical damage, causing the breakage of amylose and amylopectin chains into smaller molecules. When the moisture content is limited, viscosity is increased and, consequently, the shearing force is increased, resulting in a decrease in molecular weight of the components (Guha, Zakiuddin, & Bhattacharya, 1997; Thymi, Krokida, Pappa, & Marinos-Kouris, 2008). WSI increases with increasing temperature, regardless of the concentration of starch present. This increase in soluble solids content suggests a disintegration of the starch granules (Palav & Seetharaman, 2006). Singh-Gujral, Singh, and Singh (2001) elaborated sweet corn grits by extrusion and found that WSI increased with increasing BT and that WSI decreased with increasing FM, showing a greater FM effect. This behavior was attributed to the fact that increasing FM decreases SME, and this leads to low starch solubility. Agustiniano-Osornio et al. (2005) and Yağcı & Göğüş (2008) reported that the extrusion process produces a complete starch gelatinization at low FM, when BT is in the range of 110–135°C. According to these authors, a combination of heat treatment and mechanical shear could explain the disappearance of granular structure and crystallinity of starch in extruded materials. WAI and WSI are important parameters to define the possible application of extrudates. A high WSI is related to thickener characteristics of extruded products (Hashimoto & Grossmann, 2003).

Total anthocyanin content (TAC)

Blue corn flour showed a TAC of 374 ± 9.60 mg/kg db, whereas the mixture (BCF + CS) showed a TAC of 248.67 ± 4.33. TAC of blue corn flour is consistent with that reported by Del Pozo-Insfran, Brenes, Serna, and Talcott (2007), but higher than TAC values reported by Aguayo-Rojas et al. (2012) and Mora-Rochin et al. (2010) for Mexican blue corn. The effect of BT and FM on TAC of microwave-expanded pellets is shown in Figure 7.

Figure 7. Effect of barrel temperature (°C) and feed moisture (%) on total anthocyanin content (TAC) of blue corn microwave-expanded snack.

Figura 7. Efecto de la temperatura de barril (°C) y humedad de alimentación (%), en el contenido total de antocianinas (TAC) de botanas de maíz azul expandidas.

It can be seen that throughout the FM experimental interval, TAC decreases with increasing BT. This decrease may be due to the poor stability of anthocyanin to heat. On the other hand, at low BT (<110°C), in general, FM had no effect on TAC levels. However, above this temperature, at intermediate conditions of FM (~27%), highest TAC values were observed. It is probable that during the increase of FM from 18% to 27%, the severity of the extrusion process was reduced, due to the lubricant effect of water. However, an increase of FM higher than 27% provided moisture sufficient for starch gelatinization that causes paste formation, provoking the slowing of the material flow, and a longer exposure to the action of high temperatures and mechanical shear, conducive to TAC degradation. The minimum (41.16 mg/kg db) and maximum (82.3 mg/kg db) TAC values corresponded to 105°C of BT/22% of FM and 135°C of BT/32% of FM, respectively. These data indicate a TAC decrease in 70–85% in relation to the raw material (248.67 ± 4.33 mg/kg). Losses of 64–90% for TAC caused by extrusion have been reported where blueberry and cranberry concentrates were incorporated to extruded corn and extruded breakfast cereals (Camire et al., 2002; Chaovanalikit, Dougherty, Camire, & Briggs, 2003). Aguayo-Rojas et al. (2012) found a TAC loss of 53.5% in tortillas elaborated from lime-cooking extruded blue corn. Due to heat treatments, anthocyanin may suffer important structural changes such as conversion to colorless chalcones (Wrolstad, Durst, Giusti, & Rodriguez-Saona, 2002), and due to their thermolability, chalcones may be instantly degraded into phenolic acids (Sadilova, Carle, & Stintzing, 2007). On the other hand, polymerization and browning also lead to a decrease in TAC (Singh, Gamlath, & Wakeling, 2007).

Optimization

The optimization of the extrusion process was carried out by the response surface superposition methodology. The selected responses for this procedure were EI, PF, SME, and TAC. The main criteria for determining the optimal area of surface superposition was the finding of processing conditions corresponding to the highest values of EI and TAC and the lowest values of SME and PF.

The area corresponding to the optimal conditions for obtaining expanded snacks, elaborated from blue corn flour and corn starch, ranged from 120°C to 126°C of BT and from 23.80% to 25.20% of FM, selecting as the central point the following conditions: 122.3°C of BT and 25.20% of FM (Figure 8). To validate the models, one experimental assay was carried out with the central point conditions. The predicted values by the mathematical models for each response were EI = 4.10 ± 0.04, PF = 12.42 ± 0.31 N, SME = 169.08 ± 1.85 kJ/kg, and TAC = 71.09 ± 1.10 mg/kg. The experimental values of the obtained products (pellets) were EI = 4.47 ± 0.07, PF = 11.45 ± 0.49 N, SME = 185 ± 5.5 kJ/kg, and TAC = 61 ± 1.74 mg/kg. There is no significant difference (p = 0.05) between the predicted and the experimental values, except for TAC. Therefore, the tested model showed a good fit in finding the best conditions of BT and FM for the elaboration of BCF + CS-expanded snacks by the extrusion process.

Figure 8. Area of superposition of responses (EI, PF, SME, and TAC) for BT and FM in the optimization process of blue corn snack.

Figura 8. Área de la superposición de las respuestas (EI, PF, SME y TAC) de BT y FM, en el proceso de optimización de botanas de maíz azul.

Conclusions

The mathematical models used in the analysis of responses showed suitable values (R² ≥ 0.76), although some responses showed lack of fit. BT and FM had a significant effect on all the studied responses, except for WAI and TAC. The EI values showed by expanded blue corn products were similar to those exhibited by a commercial product, while the BD and PF were higher. The dietary fiber and anthocyanin derived from the addition of whole blue corn flour confer eventually nutraceutical characteristics to the expanded snacks. However, studies are needed to evaluate their nutraceutical potential. To our knowledge, this is the first report about the utilization of blue corn for the elaboration of 3G snacks.

Acknowledgments

This research was financed by Programa de Fomento y Apoyo a Proyectos de Investigación (PROFAPI-2009/029) from Universidad Autónoma de Sinaloa. Camacho-Hernández thanks PROMEP due to the scholarship support for PhD studies. Authors thank Dr Armando Carrillo-López for writing assistance.