Pre-Cooked Fiber-Enriched Wheat Flour Obtained by Extrusion: Rheological and Functional Properties

Abstract

Extrusion processing was utilized to pre-cook wheat flours substituted with 0, 10, 20, and 30% wheat bran in order to enhance their rheological properties and functionality with regards to production of cookies and tortillas. Two extrusion conditions, low-temperature-low-shear (LTLS), and high-temperature-high-shear (HTHS) were studied for pre-cooking the flours. Results showed that for all flours, as % bran increased, RVA peak viscosity (PV), mixograph peak time (PT_M), and peak height (PH) decreased. At all bran levels, PV, and PH were significantly lower for pre-cooked flours as compared to uncooked. As the percent bran and storage time (4 to 16 d) increased, the quality of cookies (weight and spread factor) and tortillas (specific volume, rollability, and extensibility) deteriorated for both uncooked and pre-cooked wheat flours. The quality of cookies and tortillas from pre-cooked flour were either similar or inferior to those from uncooked flour.

Keywords:

• Extrusion
• Pre-cooking
• Wheat bran
• Functional and Rheological Properties
• Cookies
• Tortillas

INTRODUCTION

Obesity is a global concern and the problem is reaching epidemic proportions.[1,2] It is a primary factor in a number of serious medical conditions, such as cardiovascular disease, cancer, and diabetes.[3] In the U.S. alone, over 127 million people are overweight, 60 million are obese, and 9 million are severely obese. This places an unnecessary burden on an already strained health system. Increased fiber consumption has been found to be important for lowering the risk of being overweight and associated with reduced body mass index, blood pressure, and fasting apo B and glucose concentrations.[4] Several studies have related consumption of dietary fiber and whole grains with a reduction in serum cholesterol, and a lower risk for coronary artery disease and certain forms of cancer.[5,6,7] Whole grains and other sources of dietary fiber—long an important part of the human diet—gained new stature in 1999, when the U.S. Food and Drug Administration authorized the following health claim: “Diets rich in whole grain foods and other plant foods and low in total fat, saturated fat, and cholesterol, may help reduce the risk of heart disease and certain cancers.”[6] A recent food industry report mentioned “high-fiber” as one of the top ten functional food trends in the U.S. market.[8] With increasing awareness of the benefits of dietary fiber, the demand for high-fiber functional foods is expected to continue increasing with projected growth of the fiber industry to $495 million by 2011, which would represent a more than two-fold increase over the preceding 7-year period.[8] It is, therefore, important to explore new ways of incorporating fiber into food products.

There are various definitions and classifications of fiber. According to one classification, total fiber comprises dietary fiber and functional fiber.[9] Dietary fiber consists of carbohydrates and lignins that are intrinsic and intact in plants and edible, yet resistant to digestion and absorption in the human small intestine with complete or partial fermentation in the large intestine. Functional fiber consists of isolated, non-digestible carbohydrates that have beneficial physiological effects in humans. Fiber can also be classified into soluble and insoluble fiber.[9] Soluble fibers, such as pectins and gums, are known to be effective in reducing total blood cholesterol and promoting satiety. Insoluble fibers, such as cellulose and lignin help in treating constipation and reduce the risk of colon cancer and diverticular disease. Wheat bran, which is the main fiber source for the baking industry, is mostly cellulosic. Other fiber sources such as soy hull,[9,10] oat bran,[9,11] citrus fruits,[9] and rice bran[12] are also used to boost the fiber content in baked products.

Despite its nutritional benefits, incorporation of fiber in foods, especially baked products such as bread, tortillas, and cookies, has been limited because of its poor functional properties and deleterious effects on the overall quality and consumer acceptance of the food product. Problems like poor extensibility, reduced loaf volume, and altered crumb structure are common in high-fiber baked products, because fiber disrupts the continuous visco-elastic dough matrix.[9,13,14,15] These deleterious effects depend on the bran type and its physical properties. Some studies have reported that functionality of bran from different sources can be improved by thermal processing (drum drying or jet cooking).[12,16] Physicochemical modification of fiber, using high temperature and shear conditions of extrusion processing, is another possible technique for enhancing its functional properties.[11,13,17,18,] This approach has been attempted to improve the functionality of corn fiber for use in cookies[13] and rice bran in snack products[18] with limited success.

A slight but significant variation to the above discussed studies would be the treatment of flours enriched with high levels of fiber for the improvement of functionality of the formulation as a whole. [19] The purpose of this study was to utilize extrusion processing for producing pre-cooked wheat flour substituted with high levels of wheat bran. The hypothesis was that by utilizing appropriate combinations of process parameters, the extrusion pre-cooking of fiber-enriched flour would lead to synergistic interactions between bran and other components and increase its functionality for use in baked products. Rheological properties of the pre-cooked flours, and their functionality with regards to making cookies and tortillas, were studied. Different extrusion processing conditions and post-extrusion drying methods (lyophilization and conventional oven drying) were also investigated.

MATERIALS AND METHODS

Materials and Formulation

Commercial hard red winter wheat flour, with moisture, protein, and ash contents of 14, 11.8, and 0.5%, respectively, was obtained from Horizon Milling, LLC (Wichita, KS). The flour was substituted with 0, 10, 20, or 30% hard red winter wheat bran obtained from the pilot mill in the Department of Grain Science and Industry, Kansas State University. All experiments were conducted using the bran substituted wheat flour as base ingredient. The other ingredients in the cookie formulation, including sugar (90%; United Sugar Corp., Clewiston, FL), all purpose shortening (60%; ACH Food Companies Inc., Cordova, TN), dry eggs (7%; Michael Foods Egg Products, Gaylord, MN), salt (1.8%; Morton Salt, Chicago, IL), sodium bicarbonate (1.8%; Sigma Aldrich Co., St.Louis, MO) and Butter Lemon Vanilla flavor (1.0%; Mothers Murphy's Labs, Greensboro, NC), were obtained from a local grocery store. This formulation was based on the method described by Payne.[20] The other ingredients in the tortilla formulation included all purpose shortening (11%), salt (1.5%), sodium bicarbonate (1.5%), potassium sorbate (0.5%), sodium propionate (0.5%), sodium stearoyl lactate (0.5%), fumaric acid (0.2%) and cysteine (0.003%). This formulation was based on the method described by Bello et al. [21] These ingredients, except for the first two, were obtained from Sigma Aldrich Co. (St.Louis, MO). The relative amounts of all materials for both cookie and tortilla formulation were expressed in baker percentages.

Flour Processing

The wheat flour substituted with 0–30% bran was processed in a laboratory scale twin-screw extruder (Micro 18; American Leistritz Extruder Corp., Somerville, NJ) under two different conditions: - 1) barrel temperatures of 30, 32, 34, 36, 38, and 40°C and screw speed of 200 rpm (low-temperature-low-shear or LTLS); and 2) barrel temperatures of 30, 40, 50, 60, 70, and 80°C and screw speed of 250 rpm (high-temperature-high-shear or HTHS). The extruder screw profile and barrel temperature zones are shown in Fig. 1. In-barrel moisture content was maintained at 30% (wet basis) for all treatments. The target moisture was achieved by mixing the flour with water in a bench-top mixer (KSM5; Kitchen Aid, St Joseph, MI), taking the initial moisture of the flour into account. The hydrated flour was stored overnight at 4°C for equilibration before extrusion. The ribbon-like extruded product was dried by two different methods—lyophilization and oven drying. The former was performed using a Labconco FTS System Inc. (Kansas City, MO) freeze drier, while a Thelco laboratory oven (160DM; Precision Scientific, Chicago, IL) set at 60°C was used for the latter. The moisture content of the dried flour ranged between 3.9–7.4 % (wet basis). The dried product was ground to pass through a 0.5 mm sieve using a Thomas Wiley laboratory mill (Model 4; Arthur H. Thomas Company, Philadelphia, PA). For control studies, uncooked flour was also substituted with 0–30% bran that was ground using the same mill to pass through a 0.5 mm sieve.

Figure 1 Schematic of the laboratory-scale extruder screw profile¹ and barrel temperatures.^{2 1}All screws are forward and intermeshing. ²LTLS process temperatures (shown above) were 30-32-34-36-38-40°C; and HTHS process temperatures were 30-40-50-60-70-80°C.

Cookie Preparation

The cookie dough preparation and subsequent baking was performed using a method outlined by Payne,[20] as summarized in Fig. 2. The ingredients were mixed using a 10-qt mixer (A-200; Hobart, Troy, OH). For both uncooked and pre-cooked flours, the water addition level was 20%. A moistened scoop (#20; Vollrath®, Sheboygan, WI) was used to drop dough onto a pan covered with a liner sheet. Cookies were baked in a reel oven (Reed Oven Co., Kansas city, MO) at 176.6°C (350°F) for 13–15 min.

Figure 2 Flow diagram of the cookie making method. ¹Mixer had speed settings from 1–3. ²BLV = Butter Lemon Vanilla.

Tortilla Preparation

The tortilla dough preparation, pressing, and subsequent baking were performed using the standard method described by Bello et al.,[21] as summarized in Fig. 3. For uncooked flour, water addition level was 61%. For the pre-cooked flours, a few modifications were made to the methodology. These included addition of 15 and 20% wheat gluten (MGP Ingredients, Atchinson, KS) to the pre-cooked LTLS and HTHS flours, respectively; and water addition levels increasing from 61 to 81% as bran levels increased from 0 to 30%. The dough balls were hot pressed for 15 s using a Dough Pro tortilla hot press (Process Corp., Paramount, CA). The top and bottom platen temperatures of the hot press were 80.6 and 78.3°C, respectively. The gap between the hot plates was set at “thin” (1.5–2.0 mm). The pressed dough was baked on griddle (Speedester, Walter & Carrell Mfg. Co., Denver, CO) for 40 sec on each side at a temperature of 173°C.

Figure 3 Flow diagram of the tortilla making method. ¹Mixer had settings from 1–3.

Rheological Properties of Flour

Rapid visco-analyzer. A Rapid Visco-Analyser (RVA-3D, Newport Scientific, NSW, Australia) was used to measure pasting properties of flour samples using the AACC Approved Method 22–08.[22] The flour sample (3.13–3.4 g) was added to 25.1–25.4 ml of water in an aluminum canister, and the RVA test was performed with a total run time of 13 min. The peak viscosity (PV) and peak time (PT_R) obtained from the pasting curve were used to infer the degree of cooking and degradation of flour during extrusion processing.

Mixograph. Dough rheological properties were measured using a Mixograph (National Mfg. Co., Lincoln, NE) as described by AACC Approved Method 54-40A.[22] The ratio of flour and water used in the mixograph studies was calculated based on the initial flour moisture and protein content. The flour and water were mixed in a 10-g mixograph bowl for 10 min according to the standard procedure. The peak height (PH; percentage of maximum) and peak time (PT_M) obtained from the mixogram curve were used to infer the dough strength.

Cookie and Tortilla Quality Parameters

Cookie properties. The weight (w_c), diameter or width (W) and thickness (T) of cookies were determined 24 h after baking as described by Payne.[20] The spread, defined as the W/T ratio, was also calculated.

Tortilla physical properties. The weight (w_T), diameter (D), and height (H) of freshly baked and cooled tortillas were measured and specific volume (V) calculated as described below.

(1)

Opacity was measured subjectively using a continuous scale where 100% was completely opaque (white) and 0% was completely translucent. Water activity (a_w) was determined using a water activity meter (CX2; Decagon Devices, Inc., Pullman, WA) after blending the tortilla in a coffee grinder.

Tortilla storage studies. Tortillas were packaged in sealed plastic bags and placed at room temperature (22°C) for storage. A previous study showed that this storage temperature corresponded with the fastest firming of flour tortillas.[23] Measurements for extensibility and rollability were conducted on tortillas stored for 4, 8, 12, and 16 days as described below.[24] Tortilla extensibility was measured using a texture analyzer (TA.XT2; Texture Technologies Corp., Scarsdale, NY). The test was conducted on a 60 × 35-mm tortilla strip, cut from the center of the tortilla, using the return-to-start option in tension mode, a trigger force of 0.05 N, and probe travel distance and speed of 10 mm and 1mm/s, respectively. The force and distance of rupture of the tortilla strip were recorded. Rollability of tortilla was evaluated using the subjective scoring method described by Friend et al.[25] A tortilla was wrapped around a wooden dowel of 1.0 cm diameter and its cracking and rollability were rated on a scale of 1–5 (1 = unrollable or worst to 5 = no cracking or best).

Experimental Design and Statistical Analysis

The effect of bran substitution and extrusion pre-cooking on the rheological and functional properties of wheat flour was investigated using a 4 × 3 complete factorial design, with four levels of bran substitution (0, 10, 20, and 30%) and three levels of processing (no pre-cooking or control, and LTLS and HTHS extrusion pre-cooking). Flour from each treatment was partitioned into three sub-samples for all subsequent tests. RVA and Mixograph tests were duplicated for each flour sub-sample. Physical parameters of cookies and tortillas were measured from batches based on each sub-sample. Rollability measurements were duplicated for tortillas from each sub-sample. Extensibility of tortillas was measured in three replicates from each treatment. The effect of drying methods (lyophilization and oven drying) was studied using only HTHS pre-cooked wheat flour without any bran substitution. Uncooked flour was used as the control. This comprised of an experiment design involving 3 treatments. The rheological properties of the flour, as well as the physical and textural properties of cookies and tortillas, were evaluated using two-way analysis of variance (ANOVA). Fisher's Least Square Difference (LSD) was used for multiple means comparisons (α = 0.05). SAS software (version 9.1; Cary, NC) was used to conduct all statistical analyses.

RESULTS AND DISCUSSION

Effect of Drying Methods

The rheological properties of uncooked flour and HTHS pre-cooked flour dried by lyophilization and oven drying are shown in Table 1. None of these flours were substituted with bran. RVA peak viscosity (PV) and peak time (PT_R) for the control (uncooked flour) were 2237 cP and 5.9 min, respectively. PV for the lyophilized and oven-dried pre-cooked flour was 34 and 48% lower, respectively, as compared to the control. The differences in PV for all three flours were significant (P < 0.05). PT_R for the pre-cooked flours was also lower than that for the control, although the differences were not marked. The RVA pasting parameters provide a relative measure of starch pasting, swelling and degradation. As starch is heated in the presence of water, granules swell leading to increase in viscosity of the starch-water suspension.[15,26] Uncooked starch would take a longer time (PT_R) to reach its PV. Also intact or relatively less degraded granules would swell more, leading to a higher PV. The pasting parameters indicated that HTHS extrusion processing followed by drying and grinding led to degradation of the flours, especially the starch fraction, irrespective of the drying method. It is well known that due to substantial mechanical energy input, extrusion processing not only leads to cooking or pasting of the starch but also ruptures the starch granules, often leading to some dextrinization.[27–31] The post-drying grinding process also probably contributed to flour degradation. It was clear from the RVA data that lyophilization led to significantly less degradation than oven drying. Very little physicochemical changes occur during the former process because of sub-zero temperatures. However, thermal processing can lead to deterioration in properties,[32,33] even though a relatively mild drying temperature of 60°C was employed in this study.

Table 1 Rheological properties ¹ of uncooked flour and HTHS ² pre-cooked flour ³ prepared by oven drying and lyophilization

Flour type	RVA peak visc (cP)	RVA peak time (min)	Mixograph peak height (%)	Mixograph peak time (min)
Uncooked	2237^a	5.92^a	45.4^a	3.5^a
Lyophilized HTHS	1480^b	5.5^b	22.9^b	0.5^b
Oven dried HTHS	1160^c	5.8^a	13.8^c	0.5^b
^1Mean, n = 3; means with same superscript in the same column are not significantly different (p < 0.05). ^2High-temperature-high-shear extrusion processing conditions. ^3All flours had 0% bran substitution.

Mixograph peak height (PH) and peak time (PT_M) for the control were 45.4% and 3.5 min, respectively. PH for lyophilized and oven-dried, pre-cooked flours were 50 and 70% lower, respectively, as compared to the control. The differences in PH for all three flours were significant (P < 0.05). PT_M for both pre-cooked flours was also substantially lower (by 85%) than the control, although there was almost no difference between the oven-dried and lyophilized flours. The mixograph parameters provide a measure of dough water absorption, viscoelastic strength, and stability or tolerance to over-mixing.[21,26] Usually, flours with good gas-holding properties and machinability for products like bread and tortillas have higher water absorptions, take longer times to mix, and have a better tolerance to overmixing than does poor quality flour.[26] Most of these attributes are a function of the flour protein content and quality, but also depend on the other ingredients in the dough, for example shortening.[21] PH is the height of the mixograph center curve at the highest point, and relates to water absorption and protein. PT_M is the time of mixing to reach PH or the dough development time. PH and PT_M of the mixograph curve approximately correspond to optimally developed dough. Mixograph results indicated that HTHS extrusion processing led to deterioration of protein quality, and thus, poor water absorption and viscoelastic strength as compared to the control. The high thermal and mechanical energy input to the flour during extrusion would cause denaturation of proteins, thus rendering them ineffective for dough development. It is also clear from PH data that lyophilized flour was significantly better in quality than oven-dried flour. The latter clearly led to further deterioration in protein quality, while lyophilization effectively prevents any additional physicochemical changes. Both RVA and mixograph data suggested that although HTHS pre-cooking led to poor flour quality and rheological properties, lyophilization was a better drying method than oven drying for the pre-cooked flour. Thus, only lyophilized pre-cooked flour was used for all subsequent experiments.

Rheological Properties of Pre-Cooked Flour with Different Bran Levels

Pasting properties of uncooked (control) and pre-cooked flours with 0–30% bran are shown in Table 2 a. PT_R did not show any significant trend with respect to bran level. For all three types of flours (control, LTLS, and HTHS), PV decreased as bran levels increased, with the exception of LTLS pre-cooked flour containing 20% bran. Although the general trend was not statistically significant, it was attributed to a decrease in the water-swelling starch fraction due to its replacement with bran. Another possible reason could be the interference of bran or fiber with the gelatinization or water absorption of starch granules. Similar results were observed in studies by Brennan and Samyue[15] on biscuit flour substituted by 0–10% dietary fiber (RS2 starch and inulin) and Arambula et al. [19] on tortilla flour substituted with 0–6% corn pericarp. Extrusion pre-cooking of flour led to decrease in PV at all bran levels, except in the case of LTLS pre-cooked flour with 20% bran. These differences were statistically significant in most cases, and were attributed to starch macromolecular degradation during the extrusion process as discussed earlier. Interestingly, at lower levels of bran substitution (0 and 10%), PV was higher for HTHS pre-cooked flour as compared to that of LTLS flour. This was contrary to expectation, as the more severe temperature and shear conditions during the HTHS process were expected to cause greater starch degradation and thus lesser swelling during pasting. This anomaly could be possibly due to some kind of synergistic effect of pre-cooked fiber and starch on water binding and swelling at lower bran levels.[19]

Table 2 Rheological data ¹ for uncooked and pre-cooked ² wheat flour with 0–30% bran substitution–using (a) rapid visco analyzer (RVA) and (b) mixograph

(a)
Flour type	Bran level (%)	Peak Visc (cP)	Peak Time (min)
Uncooked	0	1059^b	4.9^h
	10	998^bc	5.0^g
	20	935^bc	5.2^e
	30	930^bc	5.1^f
LTLS pre-cooked	0	786^d	5.3^d
	10	783^d	5.3^d
	20	1‘231^a	5.3^d
	30	709^cd	5.2^e
HTHS pre-cooked	0	982^bc	5.6^a
	10	894^cd	5.4^c
	20	601^e	5.5^b
	30	527^e	5.6^a
(b)
Flour type	Bran level (%)	Peak Time (min)	Peak Height (%)
Uncooked	0	4.1^bcd	44.6^b
	10	5.2^abcd	37.9^bc
	20	3.7^cde	33.2c^cd
	30	3.6^def	34.4^cd
LTLS precooked	0	0.8^h	26.8^d
	10	0.7^h	16.2^e
	20	6.0^a	31.3^cd
	30	1.0^gh	36.5^bc
HTHS Pre-cooked	0	5.8^ab	66.1^a
	10	4.9^abcd	66.6^a
	20	1.8^fgh	44.0^b
	30	2.2^efgh	61.6^a
^1Mean, n = 3; means with same superscript in the same column are not significantly different (P < 0.05).^2 HTHS = high-temperature-high-shear extrusion processing conditions; and LTLS = low-temperature-low shear extrusion processing conditions.

Mixograph parameters for control and pre-cooked flours with 0–30% bran are shown in Table 2 b. PH and PT_M for the control were lower by 15–26% and 10–13%, respectively, when substituted with 10–30% bran, with the exception of PT_M for flour with 10% bran. This was as expected because bran disrupts the continuous protein-starch matrix and negatively affected dough development, leads to poor visco-elastic strength. Similar results were obtained for HTHS pre-cooked flours with 10–30% bran as compared to no bran substitution, although the dough strength appeared to improve at 30% bran level. For LTLS pre-cooked flours, dough strength deteriorated with 10% bran but was substantially higher for 20–30% bran substitution, as compared to 0% bran.

Extrusion pre-cooking at HTHS and LTLS conditions decreased PT_M of flours with the same bran level, in most cases. LTLS pre-cooking of flours also decreased the PH, but on the other hand, HTHS pre-cooking led to an increase in PH. In general, mixograph data indicated that dough development and strength were negatively affected on substitution with bran. Extrusion pre-cooking led to protein denaturation, and thus negatively affected dough development. However it appeared that pre-cooking also led to synergistic interactions between bran and other components of the flour, resulting in improvements in the dough, especially using LTLS conditions for higher bran levels (20–30%) and HTHS conditions for lower bran levels (0–10%). Caprez et al. [34] reported an improvement in dough strength on boiling of wheat bran prior to substituting flours at 20% level. They attributed this synergistic effect to the partial gelatinization of starch present in the wheat bran. However, other thermal treatments (steam cooking, autoclaving, roasting, micronising and extrusion) of wheat bran led to deterioration in dough strength at the same substitution level. Bran levels other than 20% were not studied.

The pasting (RVA) and mixograph results revealed a complex relationship between degree of processing, bran substitution, and the resultant dough properties. Addition of bran led to a deterioration in dough strength, due to a reduction in the protein and starch fractions, and disruption of the continuous visco-elastic matrix. Extrusion pre-cooking led to degradation of starch and protein, but the data pointed toward some synergistic affects between bran and other flour components due to pre-cooking. This, however, needs to be investigated further.

Quality of Cookies from Pre-Cooked Flours with Different Bran Levels

The quality parameters for cookies from uncooked (control) and pre-cooked flours with 0–30% bran are given in Table 3. The average weight (w_c) of baked cookies varied within 9 and 7%, respectively, for the control and HTHS pre-cooked cookies. The weight variation was much higher (up to 22%) in the case of LTLS pre-cooked flours. The methodology for cookie making employed in the study was different than the standard methodology (AACC standard methods 10-52 and 10-53).[22] In the latter, the cookie dough is flattened and then cut with a cookie cutter, which results in more uniform cookies. In this study, however, a scoop was used to dispense the cookie dough onto the baking pan according to the method described by Payne.[20] Although care was taken to minimize variation in the volume of batter dispensed by the scoop, this departure from the standard “cookie-cutter” method could be one of the reasons for the variation in w_c. However, it is more likely that some other factors was responsible for the variation in w_c, as a definite trend was observed for w_c with respect to the bran level. For all three types of flour (control, LTLS and HTHS), w_c increased as bran levels increased from 0–30%. It is possible that cookies with higher bran level retained more water after baking because of binding of water by fiber. The effect of processing on w_c also exhibited an interesting trend with control > LTLS > HTHS. This was likely due to decreased water binding as the flour fraction was more degraded with higher degree of processing. Differences in cookie weight also corresponded with water retention and variation in final moisture in the study by Artz et al.[13]

Table 3 Quality parameters ¹ of cookies from uncooked and pre-cooked ² wheat flour with 0–30% bran substitution

	Bran level (%)	Weight (g)	Width (cm)	Thickness (mm)
Uncooked	0	49.8^e	11.0^a	0.7^f
	10	52.7^c	10.9^b	0.7^f
	20	53.9^b	10.4^c	0.7^f
	30	54.5^a	9.3^f	1.2^e
LTLS precooked	0	42.0^h	8.8^j	1.3^d
	10	42.4^g	9.0^i	1.2^e
	20	49.9^e	9.8^e	1.2^e
	30	51.1^d	8.0^l	1.7^b
HTHS precooked	0	41.0^i	8.7^k	1.8^a
	10	41.1^i	10.1^d	1.2^e
	20	43.8^f	9.2^g	1.5^c
	30	42.5^g	9.1^h	1.5^c
^1Mean, n = 3; means with same superscript in the same column are not significantly different (P < 0.05). ^2HTHS = high-temperature-high-shear extrusion processing conditions; and LTLS = low-temperature-low shear extrusion processing conditions.

For the control flour, cookie width (W) decreased and thickness (T) increased as the bran level increased from 0–30%. The W/T ratio or spread (Fig. 4) correspondingly decreased with an increase in bran. All differences in W and W/T were significant. Greater spread indicates better cookie quality. The spread mechanism in cookies is a function of the total availability of water.[35] Any change in composition of the flour which would make it more hydrophilic tends to decrease the spread, as less water is available for dissolving the sugar, which is the main spreading ingredient. In this study, higher bran levels in the control flour probably led to greater water binding as discussed earlier, resulting in reduced spread. Similar results have been obtained in previous studies on incorporation of resistant starch (RS), dietary fiber,[36] or rice bran (0–15%) [18] into cookies, and RS, inulin, or potato fiber (0–10%) into biscuits.[15] For both LTLS and HTHS pre-cooked flour, the trends for W, T and spread ratio with respect to bran levels were different than the control. In general, LTLS and HTHS flours with 10–30% bran had greater W, lower T and correspondingly higher spread ratio (Fig. 4), as compared to the pre-cooked flours without any bran substitution. Pre-cooking of flour under both LTLS and HTHS conditions led to a significant decrease in spread ratio. This was attributed to lesser water retention during baking.

Figure 4 Spread (W/T ratio) of cookies from uncooked and pre-cooked wheat flours with 0–30% bran substitution. ¹Y error bars denote the least significant difference, n = 3. ²HTHS = high-temperature-high-shear extrusion processing conditions and LTLS = low-temperature-low shear extrusion processing conditions.

Quality of Tortillas from Pre-Cooked Flours with Different Bran Levels

Common attributes that determine tortilla quality include opacity, height, diameter and shelf stability.[37] Consumers generally prefer opaque, fluffy tortillas that retain their freshness and flexibility for weeks. The quality parameters of tortillas from uncooked (control) and pre-cooked flours with 0–30% bran are shown in Table 4. As discussed earlier extrusion pre-cooking of flour led to poor-dough development and low visco-elastic strength. In order to make a machinable dough that was fit for baking tortillas, 15 and 20% gluten was added, respectively, to LTLS and HTHS flours. The level of gluten addition was determined based on ‘trial-and-error’ for optimum dough development and tortilla quality. The quality parameters, including weight (w_T), diameter (D), height (H), opacity, specific volume (V), and water activity (A_w) did not vary substantially. Weight of the tortillas (w_T) was not affected substantially by the bran level or degree of processing of the flour, as the same amount of dough was weighed while rounding the balls. The amount of water uptake increased on addition of bran. This led to evaporation of water during baking, and shrinkage of tortilla diameters, as was observed for both the control and pre-cooked flours. The height of tortillas from HTHS pre-cooked flour was significantly higher. In general, LTLS pre-cooked flour tortillas had greater opacity, but lower specific volume. The A_w of all the tortillas was very high (> 0.91). The optimum A_w for preventing mold growth is usually below 0.70. This underscores the need for preservatives to enhance the shelf-life.

Table 4 Quality parameters ¹ of tortillas from uncooked and pre-cooked ² wheat flour with 0–30% bran substitution

	Bran level (%)	Weight (g)	Diameter(cm)	Height (mm)	Opacity (%)	Sp.Vol (m^3/g)	Aw
Uncooked	0	37.7^c	18.1^a	1.5^e	76.3^b	1.01^b	0.91^f
	10	38.4^b	14.4^i	2.2^a	73.4^d	0.94^e	0.93^d
	20	38.7^a	15.3^d	1.9^c	76.2^b	0.95^d	0.94^c
	30	38.7^a	15.6^c	1.9^c	72.4^e	0.95^d	0.95^b
LTLS Precooked	0	37.6^cd	16.4^b	1.7^d	71.5^f	0.95^d	0.92^e
	10	36.6^f	14.3^j	2.0^b	77.4^a	0.88^f	0.94^c
	20	37.2^e	14.4^i	1.9^c	76.0^b	0.87^g	0.97^a
	30	36.3^g	15.3^d	1.7^d	77.8^a	0.87^g	0.95^b
HTHS Precooked	0	38.4^b	14.6^h	2.2^a	74.2^c	0.95^d	0.94^c
	10	37.6^cd	14.9^g	2.2^a	74.1^cd	1.02^a	0.95^b
	20	37.4^de	15.1^f	2.0^b	74.8^c	1.01^b	0.94^c
	30	37.4^de	15.2^e	2.0^b	72.2^ef	0.99^c	0.93^d
^1Mean, n = 3; means with same superscript in the same column are not significantly different (P ≤ 0.05). ^2HTHS = high-temperature-high-shear extrusion processing conditions; and LTLS = low-temperature-low shear extrusion processing condition.

Figure 5 Rollability of tortillas from (a) uncooked flour, (b) LTLS pre-cooked flour, and (c) HTHS pre-cooked flour, after storage for 4, 8, 12, and 16 d. ¹Y error bars denote the least significant difference, n = 3. ²HTHS = high-temperature-high-shear extrusion processing conditions and LTLS = low-temperature-low shear extrusion processing conditions.

Figures 5, 6, and 7 show the results from storage studies for the tortillas. In general, for all three flour types (control, LTLS and HTHS), rollability decreased with increasing bran levels and storage time (4 -16 d) (Fig. 5). Rollability of tortillas from control (Figure 5a) and LTLS flours (Fig. 5b) was similar, but rollability of tortillas from HTHS flours (Fig. 5c) was significantly lower. Tortilla's extensibility, as measured by force and distance for rupture, in shown in Fig. 6 and 7. In general, force for rupture was higher and rupture distance shorter for tortillas from flours substituted with 10–30% bran, indicating a decrease in extensibility. Extensibility also decreased with increases in storage time from 4 to 16 d. Pre-cooking of flours by extrusion lowered their extensibility. Good tortillas are characterized as soft, extensible, and flexible when fresh. The texture becomes firmer, less extensible, and less rollable, when the tortillas are stored at room temperature for a prolonged period of time.[23] The loss of freshness and increased firmness in tortillas with increasing storage time were due to retrogradation of starch,[23] and have been observed in previous studies.[10,37] Decreases in rollability and extensibility with the addition of bran were caused by poor dough development as discussed earlier, and also reported by previous studies.[10] Extrusion pre-cooking did not lead to any substantial improvement in tortilla quality or shelf stability.

Figure 6 Extensibility (force) data for tortillas from (a) uncooked flour, (b) LTLS pre-cooked flour, and (c) HTHS pre-cooked flour, after storage for 4, 8, 12, and 16 d. ¹Y error bars denote the least significant difference, n = 3. ²HTHS = high-temperature-high-shear extrusion processing conditions and LTLS = low-temperature-low shear extrusion processing conditions.

Figure 7 Extensibility (distance) data for tortillas from (a) uncooked flour, (b) LTLS pre-cooked flour, and (c) HTHS pre-cooked flour, after storage for 4, 8, 12, and 16 d. ¹Y error bars denote the least significant difference, n = 3. ²HTHS = high-temperature-high-shear extrusion processing conditions and LTLS = low-temperature-low shear extrusion processing conditions.

CONCLUSION

The rheological properties of wheat flour substituted with 0–30% wheat bran were investigated. Extrusion processing was utilized for pre-cooking the fiber-enriched flours. Results indicated deterioration in the rheological properties of pre-cooked flour. Also, the quality of cookies and tortillas from pre-cooked flour were either similar or inferior to those from uncooked flour. Complex inter-relationships were observed between bran level, processing conditions, and the rheological and functional properties of the flours. Future work should focus on extrusion pre-cooking of just the bran for improving its functionality.

ACKNOWLEDGMENTS

This study was supported by check-off funds from the Kansas Wheat Commission. The authors would like to thank Mr. Eric Maichel for his technical assistance, baking lab instructor, Mr. Dave Krishock, for providing lab facilities in the Department of Grain Science and Industry at Kansas State University, and Horizon Milling (Wichita, KS) for donating the flour ingredients. This is Contribution Number 07–237–J from the Kansas Agricultural Experiment Station, Manhattan, Kansas.

Notes

*Names are necessary to report factually on available data; however, the USDA neither guarantees nor warrants the standard of the product, and use of the name by the USDA implies no approval of the product to the exclusion of others that may also be suitable.