Development and characterization of couscous-like product using bulgur flour as by-product

Abstract

Couscous is produced traditionally by agglomeration of Triticum durum semolina with water. The aims of this study were: to produce couscous-like product by substitution of semolina with bulgur by-product (undersize bulgur); to find optimum quantity of bulgur flour and processing conditions. In order to determine the optimum processing parameters and recipes; 0, 25 and 50% of bulgur containing couscous-like samples were prepared. The color, yield, sensory properties, total phenol and flavonoid contents, bulk density, protein and ash content, texture properties were determined. Two different types of dryer e.g. packed bed and microwave were used. Optimum parameters were predicted as 50% of bulgur flour for packed bed (60 °C) and microwave (180 W) drying with 50% (w/w) of water according to yields, color (L*, a*, b*) values and sensory properties (color, odor, general appearance). For packed bed drying at 60 °C yields were 54.28 ± 3.78, 47.70 ± 1.73 and 52.57 ± 7.04% for 0, 25 and 50% bulgur flour containing samples, respectively. Lightness (L*) values of couscous-like samples were decreased with increasing the quantity of bulgur flour after both drying processes. Results of sensory analysis revealed that couscous-like bulgur were more preferable for consumers.

Introduction

Couscous, a world-wide known traditional cereal product, which is a staple food of North Africa (Aboubacar and Hamaker 2000; Rahmani and Muller 1996) and Middle East cuisines. It can be consumed as salad (tabulleh) and side dish with chicken and meat meals, as an alternative for pilaf. Depending on the formulation, processing technique and usage, there are three couscous such as Turkish, Arabic/African and short-cut pasta types. The basic industrial and traditional couscous processing steps are: (a) mixing and agglomeration of Triticum durum semolina with water, (b) steaming and (c) drying (Aboubacar et al. 2006; Debbouz and Donnelly 1996). Agglomeration is defined as the formation of aggregates by sticking together of smaller particles which is a method of size enlargement (Rhodes 2008). Wheat flour, semolina and flours of sorghum, millet, maize (Galiba et al. 1988) and barley (Kaup and Walker 1986) can be used in the couscous production.

Bulgur is a whole grain product, which is generally produced from Triticum durum wheat by cleaning, cooking in grain form, drying, tempering, debraning, milling and size classification (Bayram and Öner 2005, 2007; Hayta et al. 2003; Yıldırım et al. 2008a, b). Processing steps of wheat during bulgur production provide some functional characteristics as (1) resistance the mold contamination, (2) resist insect attacks, (3) inactivation of enzymes due to cooking process, (4) inactivation of microorganisms due to the cooking and drying (pasteurized), (5) numerous nutritional benefits, original wheat kernel nutrients are absorbed during cooking, (6) low fat, high protein, whole grain food, (7) appealing taste, (8) easy preparation and semi or ready to eat food, (9) long shelf life, because starch in wheat is gelatinized and the kernel is almost cooked, so it is more stable than wheat in hot and humid environments, (10) inexpensive and economical, (11) consumable as individually due to its nutritional properties and it is a good source for folic acid, (12) the best processing method to decrease the available phytic acid content in contrast to increasing bran content (Bayram et al. 2004). Therefore, it is nutritious, healthy and economical. Particle size of bulgur varies between 0.5 and 3.5 mm. After screening, bulgur is classified as undersize bulgur (< 0.5 mm), fine (1.0–0.5), medium (2.0–1.0 mm), pilaf (3.5–2.0 mm) and coarse (> 3.5 mm) (Yıldırım et al. 2008a, b). According to industrial survey (depend on yield, used process and technique), the amount of the by-product (undersize bulgur) is about 15% of total produced bulgur and its price around 100–200 USD/ton. This nutritionally valued product is only used as an animal feed, today (Bayram and Öner 2005). There is no published information in the literature about the use of this product. In this study, this by-product was used to produce couscous-like product. Nutritional and economical value of couscous might be increased due to the addition of undersize bulgur. Furthermore, particle size of bulgur flour may be increased by agglomeration process to avoid caking and lump formation.

In literature, traditional hand-made couscous production method is usually mentioned. As a raw material, sorghum is used in formulation (Aboubacar and Hamaker 2000; Aboubacar et al. 2006; Galiba et al. 1988). Debbouz and Donnelly (1996) compared home-made, commercial and extruded couscous samples according to their colors, water absorption indexes, degrees of starch gelatinization, cooking qualities and sensory attributes. They observed uniform size, intense yellow color and high degree of starch gelatinization for twin-screw-extruded couscous. In addition, cooking properties of chickpea flour used as a coating on couscous granules instead of wheat flour and then color and sensory evaluations were made (Demir et al. 2010). Moreover, gluten-free couscous is produced with rice, field bean, proteaginous pea (e.g. rich in protein, Pisum arvense) and chickpea flours (Benatallah et al. 2008).

The aims of this study are; to produce couscous-like product by substitution of semolina with undersize bulgur (bulgur flour, as by-product) using size enlargement technique; to find optimum quantity of bulgur flour and drying conditions. Moreover, huge amount of by-product such as bulgur flour can be converted to value-added product.

Materials and methods

Materials

Undersize bulgur (bulgur flour, < 0.8 mm) produced from Triticum durum wheat was obtained from a local factory in Gaziantep, Turkey. Semolina produced from Triticum durum wheat (the particle size of between 0.5 and 1.00 mm) was obtained from local market. Distilled water was used in all experiments and formulations. All chemicals and standards used in proximate and phytochemical analysis were obtained from Sigma-Aldrich Chemie GmbH (Germany).

Couscous-like product preparation

Experimental set-up is shown in Fig. 1. Couscous-like product was produced by manually by substituting Triticum durum semolina with undersize bulgur (bulgur flour, < 0.8 mm) at different quantities (25, 50, 75 and 100%). The quantities were given in total flour weight (w/w). The quantity of water to prepare dough was obtained by trial and error based on suitable dough stickiness and consistency. As a control (semolina-couscous), the couscous sample was produced using the quantity of 0% of bulgur flour. Formulations of couscous-like products: (1) 0% of bulgur flour, 100% of semolina and 30% of water; (2) 25% of bulgur flour, 75% of semolina and 46.75% of water; (3) 50% of bulgur flour, 50% of semolina and 50% of water.

Fig. 1.

Experimental setup

Drying

The samples were dried to 12% (w.b.) of moisture content using packed bed (MR II, Sherwood Scientific, Cambridge, England) and microwave (HMT84G421, Bosch, Germany) dryers. Drying was made at two different temperatures such as 60 and 80 °C (air velocity: 0.05 m/s, volumetric flow rate: 8.84 × 10⁻⁴ m³/s) for packed bed dryer and at two different power intensities such as 180 and 360 W for microwave dryer. Six hundred grams of sample was placed into the dryers. Weight of sample was recorded at the beginning of the drying operation. Weighing was made until the sample reaches to 12% of moisture content within 4 min intervals.

Determination of optimum processing parameters and formulations

The couscous samples were produced with 0, 25 and 50% as compositions by using packed bed (60 and 80 °C) and microwave (180 and 360 W) dryers. In order to determine the optimum processing parameters and formulations; yields, sensory attributes and color values (L*, a*, b*, YI and sensory color results) were used. The optimum recipes were used for further analysis (protein, ash, weight increase, bulk density, texture properties and sensory analysis).

Moisture content

Association of Official Analytical Chemists (AOAC) method was used for the determination of moisture content before drying using 3–4 g of sample in an oven (JS Research Inc., Korea) at 130 °C until constant weight reached up to 1.5–2 h (AOAC 1990).

Ash content

AOAC (1990) method was used to measure ash content of couscous (%, d.b.).

Protein content

AOAC (1990) Kjeldahl method was used to measure protein content of couscous (%, d.b.).

Color parameter

Color of the dry samples before the cooking operation was determined by measuring the L* (100 = white; 0 = black), a* (+, red; −, green) and b* (+, yellow; −, blue) and YI (Yellowness Index) values using QUEST II Minolta CR-400 (Minolta Camera, Co., Ltd, Osaka, Japan) with illuminate D65/10 as reference. The color results were the average of four measurements.

Screen analysis

Bulgur flour under the sieve with 0.80 mm of aperture opening was used. After size enlargement and each drying operations, the samples were sifted with 3.00/2.00/1.00 mm screens and each fraction was calculated by weighing the remaining couscous-like product on each sieve. Yields were determined for dried samples as weighing products between 2.00 and 1.00 mm screens, where weighing products between 3.00 and 1.00 mm screens was made for non-dried ones.

Analysis of % increase in weight (% WI)

Percent change in weight increase (% WI) was determined by the modified method of Demir et al. (2010). In this method, 6 ml of boiling distilled water was poured on 10 g of sample and waited for 5 min. WI was determined by differences between dry (uncooked, before cooking) and cooked weights (Eq. 1).

$$\text{WI}\left(\%\right)=100\times \frac{\left(\text{weight of cooked sample}-\text{weight of dry sample}\right)}{\text{weight of dry sample}}$$ 1

Bulk density

For bulk density measurement, 10 g of sample was poured into 50 mL graduated cylinder, the sample was packed well by hitting the cylinder on a cloth, such that the sample packs well, afterwards the volume of sample was measured. Bulk density was determined by the ratio of weight to volume of sample as kg/m³.

Sensory analysis

Color, odor and general appearance of dry samples were evaluated by 30 trained panelists. Scores were given between 1 (dislike mostly) and 10 (like mostly). Depending on color (L*, a*, b* and YI values) and sensory analysis, the optimization of formulas was made and this recipe was used for further analysis.

In sensory analysis, 120 ml of boiling distilled water was poured on 200 g of samples and kept for 10 min. The products were evaluated by 30 panelists, who are working/studying in Department of Food Engineering, Gaziantep University. Panelists evaluated each sample in duplicate using a hedonic scale. The panelists had access to water to help rinse their palates prior to proceeding to the next sample. Sensory evaluation was performed in a special sensory analysis room at the controlled temperature of 25 °C in open sitting. All samples were served at the same time on the same day. Panelists gave scores on a ten-point scale from 1 (dislike extremely) to 10 (like extremely) for chewiness, hardness, odor, taste, color and overall acceptability of the products.

Texture profile analysis (TPA)

TPA values of cooked samples were measured by using TA.XT2i Texture Analyzer (Stable Micro System Ltd., Surrey, UK) according to the method published by Altan and Maskan (2004). During test, compression rig was used. Compression was applied two times on the samples. Testing conditions were as follows: 2 mm/s of pre-test speed, 0.5 mm/s of test and post-test speed. Hardness, adhesiveness, springiness, cohesiveness, gumminess, chewiness and resilience were measured as the parameters of TPA.

Analysis of total phenolic content, total flavonoids content and DPPH (1,1diphenyl-2-picrylhydrazyl) scavenging activity (%)

According to the procedure of Caba et al. (2012), the phenolic compounds were extracted with 1 ml of 80% methanol containing 1% HCl using 100 mg sample. The extraction solvent and the ground sample were mixed in an orbital shaker at 200 rpm for 2 h at ambient temperature. The mixture was then centrifuged at 4000 rpm for 15 min. The aqueous phase was separated and collected in another tube. The extraction procedure was repeated on the precipitate. Combined extracts were used for further analysis.

Total amount of phenolic compounds in the products was determined by using the modification of the Folin–Ciocalteu method, using Gallic acid as a standard. Water (0.5 ml), 125 µl of sample extract and 125 µl of Folin–Ciocalteu reagent (diluted with distilled water, 1:10) were added in a test tube. After standing for 6 min, 1.25 ml of sodium carbonate solution (7.5%) and another 1 ml of water were added into the mixture solution. The solution was left in the dark place for 1.5 h. Absorbance was measured at 760 nm by using a spectrophotometer (Thermo Scientific Multiskan GO, USA) (Caba et al. 2012).

Total amount of flavonoids in the products was determined by using the modified method of Zhishen et al. (1999), using catechin as a standard. Water (1.25 ml distilled water) was added on 0.25 ml of sample extract. Then, 0.75 ml of NaNO₂ (5%) and 0.15 ml of AlCl₃·6H₂O (10%) were added to the mixture. After 6 min, 0.5 ml of 1 M of NaOH was added and homogenized by vortex for 10 s. Absorbance was measured at 510 nm by using a spectrophotometer (Thermo Scientific Multiskan GO, USA).

Total antioxidant capacity of the products was measured using DPPH radical scavenging activity percent method. According to the modified DPPH (1,1diphenyl-2-picrylhydrazyl) method, 1 ml of sample extract was added into a test tube containing 4 ml of methanol (80%) and 1 ml (containing 1 mM of DPPH) of freshly prepared DPPH solution. Then, the final concentration of DPPH solution was adjusted to 167 µmol. After that, tubes were left in the dark for 30 min and sample absorbance was measured by using spectrophotometer (Thermo Scientific Multiskan GO, USA) at 517 nm. Results were given for as ‘DPPH Scavenging Activity %’ (Caba et al. 2012).

Statistical analysis

Analysis of variance (ANOVA) was performed for all data to determine significant differences at the significance level of α = 0.05 by using SPSS software (version 22.0) (IBM Software, NY, USA). Duncan’s multiple range tests was carried out to determine the effect of parameters on all responses. All experiments were replicated and analyses were duplicated.

Results and discussion

To increase the value of bulgur flour which is a by-product, couscous-like product was produced by substitution of Triticum durum semolina with undersize bulgur (bulgur flour) at various quantities. The most important attributes (L*, a*, b*, YI, protein and ash content) of bulgur and semolina used as raw material in the experiments are given in Table 1. Due to the high bran content in bulgur flour, the values of lightness (L*) and yellowness (b*) were lower than that of semolina. However, the values of redness (a*) and cloudiness-opaqueness (YI, yellowness index) of bulgur flour were higher than semolina. Bulk densities of bulgur flour and semolina were 590 ± 0.00 and 790 ± 0.02 kg/m³, respectively. The difference between the bulk densities of bulgur flour and semolina was due to difference in their particle sizes. Theoretically, when the particle size decreases, surface area and void between particles increase, therefore bulk density decreases.

Table 1.

Color, physico-chemical properties and phytochemical studies of bulgur flour and semolina

	L*	a*	b*	YI	Protein content (d.b., %)	Ash content (d.b., %)	Bulk density (kg/m3)	Total phenol content (mg gallic acid/100 g sample)	Total flavonoid content (mg catechin/100 g sample)	% DPPH scavenging activity
Bulgur flour	64.86 ± 0.18	5.52 ± 0.04	25.01 ± 0.10	59.98 ± 0.13	12.32 ± 0.44	2.92 ± 0.07	590 ± 0.00	73.83 ± 1.62	105.88 ± 6.56	70.58 ± 0.60
Semolina	79.91 ± 0.37	3.33 ± 0.12	28.37 ± 1.27	54.84 ± 2.11	10.78 ± 0.37	0.76 ± 0.05	790 ± 0.02	83.53 ± 5.06	63.92 ± 5.92	68.67 ± 1.46

Color values are the averages of four replicates

Determination of optimum processing parameters and formulations

In order to determine the optimum processing parameters and recipes; 0, 25, 50, 75 and 100% of bulgur containing samples were prepared. However, yields of non-dried products with 75 and 100% of bulgur flour were lower than 50% which was our target yield to produce enough couscous for further processes. For that reason, samples with 25 and 50% of bulgur flour were only used. The prepared samples were dried using packed bed (60 and 80 °C) and microwave (180 and 360 W) dryers. After all products were obtained from the drying operations, screen analysis were made in order to determine the production yield, sensory analysis and color measurements (L*, a*, b*and YI) for determining the optimum processing parameters and formulas.

Optimization based on yield

After the preparation of the samples, screen analysis was made before the drying operations. Sieves with 3.00, 2.00 and 1.00 mm of apertures were used. The products, with granule sizes between 3.00 and 1.00 mm were accepted for the drying operations. Aboubacar et al. (2006) produced couscous from flours of different sorghum cultivars and they separated couscous into fine (< 1 mm), intermediate (1–2 mm) and coarse (> 2 mm) particles. They defined the couscous yield as the amount of 1–2 mm particles. Similar to their criteria, couscous-like products were analyzed. Yields, before drying operation, were 52.6 ± 2.28, 54.81 ± 2.44 and 50.5 ± 1.33% for the product consisting quantity of 0, 25 and 50% bulgur flour, respectively (Fig. 2). Quantity of bulgur flour was not significantly (p > 0.05) effective on the yield of the samples before the drying operations.

Fig. 2.

Yields of the products

For packed bed drying at 60 °C, the yields were 54.28 ± 3.78, 47.70 ± 1.73 and 52.57 ± 7.04% for the products consisting quantity of 0, 25 and 50% bulgur flour, respectively. Moreover, at 80 °C, the yields were 51.34 ± 0.77, 44.84 ± 1.18 and 46.13 ± 1.28% for the products consisting quantity of 0, 25 and 50% bulgur flour, respectively. Even though quantity of bulgur flour was not significantly (p > 0.05) effective on the yield when dried at 60 °C, it was significantly (p < 0.05) effective on the yield when dried at 80 °C. It can be said that increasing temperature caused decreasing yield of products for packed bed drying.

The amount of desired particle size product differed by the changes of the quantity of bulgur flour even though flour mixture was mixed well before the production. The reason of this variety can be resulted from the production technique. Moreover, the lower fraction of desired size product was due to the high amount of bran in bulgur flour competes with starch for water and results in less available water for starch gelatinization (Aboubacar et al. 2006).

For microwave drying at power intensities of 180 W, the yields were 48.26 ± 1.92, 52.56 ± 0.96 and 51.36 ± 3.03% for the product consisting quantity of 0, 25 and 50%, respectively. Additionally, for microwave drying at power intensity of 360 W, the yields were 48.18 ± 0.67, 56.60 ± 1.19 and 49.74 ± 2.06% for the product consisting quantity of 0, 25 and 50%, respectively. Even though quantity of bulgur flour was not significantly (p > 0.05) effective on the yield when dried at 180 W, it was also significantly (p < 0.05) effective on the yield for drying at 360 W. It can be said that increasing power intensity of microwave dryer resulted increase in the yield of desired particle size of the product.

The yields were slightly higher for microwave drying compared to packed bed drying. The reason of fewer yields can be the breakage of granule particles in the packed bed dryer. Even no fluidization of product occurred in the dryer; the velocity and the temperature of the drying air can cause breaking and separation of semolina/bulgur flour particles during drying. Due to the decreasing moisture content, the particles became more fragile.

Optimization based on color values

Color of couscous is an important quality parameter for consumers. Bright yellow color similar to couscous is usually preferred (Demir et al. 2010). The changes in L*, a*, b* and YI values based on the quantity of bulgur flour of samples are showed in Fig. 3. For all dried and non-dried product, the quantity of bulgur flour was significantly (p < 0.05) effective on lightness, yellowness and redness. The quantity of bulgur flour, when packed bed dried samples at 80 °C, has no significant (p > 0.05) effect on the cloudiness-opaqueness (Yellowness index, YI value). Lightness (L*) values of the samples were decreased with increasing the quantity of bulgur flour after both drying operations. Although lightness values of packed bed dried and non-dried products were similar, microwave dried ones had lower lightness values at different quantity of bulgur flour.

Fig. 3.

L*, a*, b* and YI values of uncooked products. Note For graphics SigmaPlot 12.0 was used

According to Duncan’s multiple range tests, L* values were significantly (p < 0.05) different for non-dried and microwave dried products at 180 and 360 W. L* values of packed bed dried samples at 60 and 80 °C were not significantly (p > 0.05) different for quantity of 25 and 50% bulgur flour. However, both were significantly (p < 0.05) different than the control sample (quantity of 0% bulgur flour). b* values before drying, packed bed dried at 60 and 80 °C and microwave dried at 360 W were not significantly different for quantity of 25 and 50% bulgur flour, however both were significantly different than control (quantity of 0% bulgur flour). Moreover, when microwave dried at 180 W, b* values were significantly (p < 0.05) different, a* values of non-dried, microwave dried (180 and 360 W) and packed bed dried (60 and 80 °C) products were not significantly (p > 0.05) different for quantity of 25 and 50% bulgur flour. However, both were significantly (p < 0.05) different than control (semolina-couscous). It is well known that yellow and brown color are correlated both to pigment content and enzymatic reactions, while the red index is strictly related to the development of Maillard reaction products (Oliver et al. 1993).

Yellowness (b*) values of microwave dried at 180 and 360 W were higher than non-dried product. However, packed bed dried products at 60 and 80 °C were lower than non-dried products.

Optimization based on sensory analysis

The sensory analysis after the drying operations were made and results are given in Table 2. It was found that the quantity of bulgur flour (i.e. 0, 25 and 50%) was not significantly (p > 0.05) effective on odor, in contrast to color and general appearance.

Table 2.

Results of sensory analysis of uncooked products

Quantity of bulgur flour (%)	Drying technique*	Drying parameters	Sensory attributes
			Color	Odor	General appearance
0	PBD	60 °C	6.33 ± 1.88abc	6.47 ± 1.25a	6.87 ± 1.73bc
		80 °C	6.20 ± 1.66abc	6.60 ± 1.30a	7.00 ± 1.46bc
	MW	180 W	6.87 ± 1.41bc	6.33 ± 1.54a	7.13 ± 1.36bc
		360 W	7.33 ± 1.54c	6.53 ± 1.51a	7.40 ± 1.30c
25	PBD	60 °C	5.47 ± 1.96ab	5.87 ± 1.19a	5.93 ± 1.71ab
		80 °C	5.33 ± 1.88a	6.07 ± 1.22a	5.73 ± 1.83ab
	MW	180 W	5.80 ± 1.74ab	6.60 ± 1.18a	6.07 ± 1.87abc
		360 W	5.93 ± 1.87abc	6.67 ± 1.35a	6.00 ± 1.56abc
50	PBD	60 °C	5.07 ± 1.94a	6.33 ± 1.45a	5.40 ± 2.13a
		80 °C	4.87 ± 2.07a	6.33 ± 1.11a	5.40 ± 2.06a
	MW	180 W	4.93 ± 1.87a	6.40 ± 1.24a	5.20 ± 2.08a
		360 W	4.93 ± 1.67a	6.60 ± 1.40a	5.33 ± 1.88a

* PBD packed bed dryer, MW microwave. Each column followed by different superscripts is significantly different (p < 0.05)

Optimum results

According to color (L*, a*, b* and YI) and sensory values after the drying operations, the optimum processing parameters and formulations were determined such as (1) 50% (w/w) of water for dough preparation, 50% of bulgur flour, 60 °C drying temperature for packed bed dryer; (2) 50% (w/w) of water for dough preparation, 50% of bulgur flour, 180 W drying power for microwave dryer. As a note, the quantity of water was calculated according to total flour weight to obtain suitable dough stickiness and consistency.

According to these optimum conditions, the other analyses were made for the optimum products.

Analysis of the optimum products

For the optimum products, chemical and physical analyses were made to determine some functional and engineering properties of the products. In order to compare the results, the control sample results were also used.

Bulk density

Bulk density is very important technological attribute of solid bulk products to evaluate product size, particle shape, smoothness, particle surface area and void fraction. Additionally, it can be used to design the processing equipment of this new granule product.

It was found that bulk densities of control samples were a little higher than bulgur compositions, which were produced with packed bed (at 60 °C) and microwave (180 W) dryers. However, the quantity of bulgur flour was not significantly (p > 0.05) effective on bulk density. Additionally, the control samples were not significantly (p > 0.05) different from bulgur formulations. Bulk density of granular products increases with the increase in the size of granules due to low surface area and void fraction between particles. Due to porous form and small particle size in couscous-like product, their bulk densities were lower, for example, than the hammer (663.9 kg/m³), stone (741.6 kg/m3) and disc milled (735.0 kg/m³) coarse bulgur (Bayram and Öner 2005).

Weight increase

Weight increase is used to analyze the water absorption ability of the products. It was found that there was no significant (p > 0.05) difference for the packed bed and microwave dried semolina and bulgur samples (Table 3). However, it seems that there was a small difference between both in contrast to results of the study made by Demir et al. (2010). They observed that increasing the quantity of chickpea flour resulted to decreasing weight increase value of couscous samples which were between 124 and 161%.

Table 3.

Functional properties and phytochemical studies in optimized products

Quantity of bulgur flour (%)	Drying technique*	Drying parameters	Bulk density (kg/m3)	Weight increase (%)	Protein content (d.b. %)	Ash content (d.b. %)	Total phenol content (mg gallic acid/100 g sample)	Total flavonoid content (mg catechin/100 g sample)	% DPPH scavenging activity
Quality properties							Functional properties
0	PBD	60 °C	610.32 ± 49.94a	70.06 ± 0.11a	10.95 ± 0.07a	0.93 ± 0.01a	60.43 ± 2.43b	35.27 ± 4.99b	73.55 ± 0.61a
50			519.95 ± 11.24a	70.76 ± 2.46a	11.92 ± 0.1c	2.01 ± 0.07b	50.99 ± 6.79ab	54.26 ± 5.07c	71.48 ± 1.80a
0	MW	180 W	628.28 ± 67.49a	70.48 ± 0.98a	11.05 ± 0.16ab	0.88 ± 0.03a	32.98 ± 11.07a	22.05 ± 2.77a	71.57 ± 1.44a
50			593.08 ± 38.91a	70.63 ± 1.50a	11.58 ± 0.32bc	1.90 ± 0.04b	52.93 ± 6.25ab	39.55 ± 0.84b	73.00 ± 2.76a

Quantity of bulgur flour (%)	Drying technique*	Drying parameters	Hardness (N)	Adhesiveness (N.s)	Springiness	Cohesiveness	Gumminess	Chewiness	Resilience
Texture properties
0	PBD	60 °C	122.54 ± 36.14a	− 172.16 ± 218.20a	0.52 ± 0.21a	0.36 ± 0.07a	45.48 ± 21.95a	26.08 ± 20.86a	0.09 ± 0.01a
50			216.92 ± 61.33ab	− 59.20 ± 32.23a	0.28 ± 0.03a	0.34 ± 0.00a	73.74 ± 20.90ab	20.58 ± 3.91a	0.12 ± 0.00a
0	MW	180 W	276.35 ± 81.47ab	− 20.24 ± 0.25a	0.53 ± 0.24a	0.30 ± 0.03a	84.38 ± 32.67ab	48.99 ± 37.59a	0.11 ± 0.02a
50			433.75 ± 125.83b	− 14.10 ± 0.26a	0.50 ± 0.05a	0.41 ± 0.04a	181.56 ± 67.43b	91.73 ± 42.13a	0.16 ± 0.01b

Quantity of bulgur flour (%)	Drying technique*	Drying parameters	Chewiness	Hardness	Odor	Taste	Color	General acceptability
Sensory properties
0	PBD	60 °C	4.57 ± 2.40a	4.60 ± 2.33a	5.52 ± 2.45a	5.57 ± 2.52a	5.93 ± 2.33a	5.12 ± 2.59a
50			6.80 ± 1.98b	6.20 ± 2.10b	7.10 ± 0.97b	7.42 ± 1.24b	7.67 ± 1.02b	7.28 ± 1.12b
0	MW	180 W	7.03 ± 1.56b	6.70 ± 1.44b	6.63 ± 2.29b	7.18 ± 1.98b	7.20 ± 2.10b	7.25 ± 1.95b
50			8.38 ± 1.15c	8.50 ± 0.98c	8.48 ± 1.57c	8.82 ± 1.05c	8.72 ± 1.24c	8.97 ± 0.94c

* PBD packed bed dryer, MW microwave. Each column followed by different superscripts is significantly different (p < 0.05)

Protein and ash content

Protein and ash contents of bulgur containing products were significantly (p < 0.05) higher than semolina containing products due to the raw materials properties. Bulgur has higher protein and ash contents than semolina; therefore it caused a good gain to finished product. Ash comes from bran layer of wheat and bran is not removed completely during bulgur production. In addition, phytic acid is very low in bulgur bran due to its processing properties (Bayram et al. 2004). Therefore, bulgur is a good dietary fiber source (Bayram 2005) for couscous by increasing bran content without phytic acid. Protein and ash contents of bulgur flour were similar to durum wheat (the level of 8% debranned), where Singh and Singh (2010) observed as between 12.9 and 13.2% (d.b.) and between 1.40 and 1.56% (d.b.), respectively. Moreover, durum wheat has the highest ash content, i.e. 1.95% against a lowest value of 1.70% for bread wheat according to the study made by Singh et al. (1998). Rahmani and Muller (1996) measured ash and protein contents of five traditional and four commercial couscous samples and found that protein and ash contents were between 12.2–14.3 g/100 g sample (d.b.) and 1.03–1.49 g/100 g sample (d.b.), respectively.

Total phenolic and total flavonoid contents and % DPPH scavenging activity

Phenolics provide essential functions in the reproduction and growth of the plants and acting as defense mechanisms. In addition, phenolic substances in our diet may provide health benefits associated with reduced risk of chronic diseases. Phenolic compounds in whole grains contribute to antioxidant activity (Liu 2007). Total phenolic and flavonoid contents of bulgur flour and semolina were determined as 73.83 and 83.53 mg gallic acid/100 g sample and 105.88 and 63.92 mg catechin/100 g sample, respectively. Their % DPPH scavenging activities were 70.58% for bulgur flour and 68.67% for semolina (Table 1). According to the statistical analysis, total phenolic content and % DPPH scavenging activity were not significantly (p > 0.05) affected by the quantity of bulgur flour in contrast to the total flavonoid content. Moreover, antioxidant activities of semolina and bulgur containing products were not significantly (p > 0.05) different from each other (Table 3).

The results of the present study are similar with the results of other studies made on cereal commodities. Yu et al. (2002) examined and compared three hard winter wheat varieties for their free radical scavenging properties and total phenolic contents. They found that the total phenolic contents were in range of 48.8–92.8 mg gallic acid/100 g of grain. Caba et al. (2012) determined total phenolics and DPPH scavenging activity  % of five different bulgur brands and they found that the average total phenolic content and  % DPPH values are 59.5 ± 5.2 mg gallic acid/100 g dry sample and 22.2 ± 2.4%, respectively. Higher percentage of DPPH activity resulted due to the high quantities of bran in bulgur flour. Similarly, due to the higher pigment content of semolina, antioxidant activities of control sample were high. Where radical scavenging activities of higher pigment content containing red sorghum and black rice were 92 and 87%, respectively; non-pigmented cereals have DPPH activity of 7–67% (Choi et al. 2007).

Texture profile analysis (TPA)

Textural parameters of cooked couscous-like samples are given in Table 3. Hardness, adhesiveness, springiness, cohesiveness, gumminess, chewiness and resilience were measured using texture analyzer. It was found that the resilience values of bulgur and semolina containing samples were significantly (p < 0.05) changed. In contrast to the resilience, the hardness, adhesiveness, cohesiveness, springiness, gumminess and chewiness of the product were not changed significantly (p > 0.05). The hardness of bulgur containing samples was higher than semolina containing samples for each drying technique.

Adhesiveness is defined as negative force to pull the compression rig from the sample, where springiness refers the height of sample recovers between the end of the first bite and the start of the second where chewiness is the energy needed to masticate couscous samples (Altan and Maskan 2004). Adhesiveness of packed bed dried couscous samples was higher (not preferred) than microwave dried ones. Moreover, the samples containing 0 and 50% of bulgur flour dried by using packed bed and microwave dryers were not significantly (p > 0.05) different for adhesiveness, springiness, cohesiveness, chewiness and resilience. Microwave dried bulgur containing samples had the highest values for TPA parameters of hardness, cohesiveness, gumminess, chewiness and resilience. According to the study made by Singh and Singh (2010), the increase in the debraning level of wheat caused decrease in textural properties. They observed that cohesiveness and chewiness values of wheat varieties ranging from 0.27 to 0.37 and 4.16 to 9.62 N, respectively, where similar values of cohesiveness of couscous-like products were observed ranging from 0.30 to 0.41. However, chewiness values of couscous-like products were higher than wheat varieties, i.e. between 20.58 and 91.73.

Sensory analysis

Sensory analysis showed that the quantity of bulgur flour was significantly (p < 0.05) effective on all sensory attributes (Table 3). It was obtained that bulgur containing samples had higher sensory scores than semolina containing products generally. Microwave dried bulgur containing samples had the highest scores for all sensory attributes. However, packed bed dried semolina containing products had lowest scores for all sensory attributes. The samples with bulgur flour dried with packed bed and microwave dryers were more preferable for consumers.

Conclusion

It was found that lightness (L*) values of samples were decreased with increasing the quantity of bulgur flour after both drying operations. Yellowness (b*) of microwave dried at 180 and 360 W were higher than non-dried. The quantity of bulgur flour (i.e. 0, 25 and 50%) did not show significant effect on odor, in contrast to color and general appearance. As optimum recipes and drying parameters; (1) 50% (w/w) of water for dough preparation, 50% of bulgur flour, 60 °C drying temperature for packed bed dryer; (2) 50% (w/w) of water for dough preparation, 50% of bulgur flour, 180 W drying power for microwave dryer were selected.

Results indicated that protein and ash content of bulgur containing samples were significantly (p < 0.05) higher than semolina containing samples due to the raw materials properties. The quantity of bulgur flour was significantly effective (p < 0.05) on sensory attributes (chewiness, hardness, odor, taste, color and general acceptability), protein and ash contents, total flavonoid content and resilience.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Human and animal rights

This article does not contain any studies with human or animal subjects.