- Email: [email protected]
Rheological properties of wheat flour dough and French bread enriched with wheat bran
Journal of Cereal Science 65 (2015) 167e174
Contents lists available at ScienceDirect
Journal of Cereal Science journal homepage: www.elsevier.com/locate/jcs
Rheological properties of wheat flour dough and French bread enriched with wheat bran F. Le Bleis a, b, L. Chaunier a, H. Chiron a, G. Della Valle a, *, L. Saulnier a a b
INRA, UR 1268 Biopolym eres Interactions Assemblages (BIA), 44 316 Nantes, France Food Development, Conseils en Innovation Alimentaire, 44 316 Nantes, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 November 2014 Received in revised form 3 June 2015 Accepted 27 June 2015 Available online 16 July 2015
Wheat flour doughs were elaborated with wheat-bran in various contents, up to 20%, and particle sizes of fractions, in order to study the specific role of rheological properties in processing high fibre breads. The addition of wheat bran, especially more than 10%, decreased the specific mechanical energy developed by the mixer, which was attributed to a deficient formation of the gluten network. It increased the elongational viscosity of the dough, measured by biaxial extension tests, likely through a solid particles effect. These changes explained the lower increase of porosity during proofing, assessed by digital camera and 2D image analysis. The loss of dough stability was rather attributed to the destabilizing effect of bran particles on the films separating gas bubbles. The resulting changes of bread texture, determined by image analysis and mechanical testing of breads, including crust and crumb, were governed by bread density, which was established at the end of proofing. These results help to understand the impact of wheat bran on dough rheological properties in order to design French breads with increased fibre content. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Density Extensional viscosity Gluten Particle size Texture
1. Introduction High dietary fibre diets and whole grain consumption have been associated with a lower risk of various diet related diseases such as cardiovascular disease, diabetes, hypertension, obesity, and gastrointestinal disorders (Anderson et al., 2009). For a healthy diet the recommended daily dose of dietary fibre (DF) is at least 25 g/ day that is far higher than the consumption observed in most western countries. Cereals products and especially bread are staple food that provides opportunities to deliver health benefits to large populations and therefore bread is a priority target for enrichment in dietary fibre. Among the numerous sources of DF, the most prevalent insoluble DF source is the wheat bran of the cereal kernel (Galliard and Gallagher, 1988). However, the addition of wheat bran causes severe problems on dough rheology, texture and sensory quality of bread (Pomeranz et al., 1977). Addition of bran in bread production has been widely studied, but its effect on dough rheological behaviour, proofing and baking has not been fully explored. The breadmaking process is based on three main operations: (1) mixing, in which ingredients are transformed into a macroscopic
* Corresponding author. E-mail address: [email protected] (G. Della Valle). http://dx.doi.org/10.1016/j.jcs.2015.06.014 0733-5210/© 2015 Elsevier Ltd. All rights reserved.
homogeneous medium with viscoelastic properties, mainly due to gluten network formation; (2) proofing step over which dough expands due to gas production by yeast activity, until porosity reaches about 70%; gas retention capacity of the dough depends on the rheological properties of the gluten matrix built during mixing (van Vliet, 2008); (3) baking does not increase much the porosity but sets the cellular structure by dough/crumb transition and crust formation (Bloksma, 1990). Density, or inversely, specific volume, is a common technological target. Numerous negative effects of bran addition on dough processing and breads properties have been reported such as: dough stickiness and loaf weight increase, mixing and fermentation tolerances decrease, specific volume reduction, coarser crumb texture, darker crumb colour and reduced crumb softness (Poutanen et al., 2014). Loss of crust crispiness is also inferred in the case of French bread (Chaunier et al., 2014). But the main problems are the major increase of density and crumb elasticity (Pomeranz et al., 1977), which seem to explain the lower consumer's acceptability of dietary fibre-enriched breads (Zhang and Moore, 1999). Different mechanisms have been inferred to explain the effects of bran addition: bran particles would weaken gluten network (Pomeranz et al., 1977) rather than dilute it (Galliard and Gallagher, 1988), by preventing protein aggregation during mixing (Noort et al., 2010). This could possibly be due to ferulic acid release, size depletion
168
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
structure and the mechanical properties of the crumb. Nomenclature 2. Materials and methods a, b, c, d a0 , b0 , c0 Ebread Es K M n P(t) Pm(t) S(t) Tf, Tt
DT εb ε_ b
hE rs r* sm, s2/3,
g_
Parameters of the Gompertz model for porosity Parameters of the exponential decay for stability Apparent modulus of bread (Pa) Specific mechanical energy (kJ/kg) Consistency index (Pa.sn) Total dough mass during mixing (kg) Flow index () Dough porosity () Value of mechanical power during mixing (kW) Dough shape ratio or stability () Dough temperature after pre-mixing and texturing steps, respectively ( C) Temperature rise during mixing ¼ (TtTf) ( C) Elongational strain (biaxial) () Elongational strain rate (biaxial) (s1) Elongational viscosity (Pa s) Density of gas-free dough Density (g cm3) sr Apparent stress at the end of relaxation (Pa) Shear rate in the mixer (s1)
effects, and competition with gluten for water due to their high water binding capacities (Wang et al., 2002, 2003); in addition, during proofing, bran particles could also restrict gas cells expansion (Gan et al., 1995), destabilize the interface between gas bubbles in the fermented dough, thus limiting dough expansion (Cavella et al., 2008), while larger particles may also pierce gas cells (Courtin and Delcour, 2002). As solid particles act like a charge in a suspension, bran addition also increases the extensional and shear viscosities (Cavella et al., 2008; Bonnand-Ducasse et al., 2010), which is unfavourable to the growth of the gas cells. Campbell et al. (2008) also suggested that bran acted during baking rather than during proofing by releasing extra water available for starch gelatinisation, and thereby lowering the final bread volume. Various means can help to reduce these drawbacks. Reducing particle size has been claimed to increase bread volume (Lai et al., 1989), but opposite effects have been observed by Noort et al. (2010) and Zhang and Moore (1999). Noort et al. (2010) explained this negative impact by an increase of surface interaction and water absorption rate, and also liberation of reactive compounds diminishing aggregation of gluten proteins. Pre-treatments of bran, such as pre-fermentation or heat treatments (Zhang and Moore, 1999; Salmenkallio-Marttila et al., 2001), and addition of vital gluten, or the use of surfactants and enzymes (Shogren et al., 1980), can also minimize the negative effect of bran addition on dough rheology and bread volume. Finally, literature survey reveals that the effects of the incorporation of dietary fibres on dough and bread properties depend on the type of bran, the addition level and the breadmaking process. Meanwhile, very few studies addressed the effect of bran addition on the dough properties relevant at different steps of the breadmaking process. In this context, the aim of the study was to determine the influence of wheat bran addition on wheat flour dough rheological properties, in order to ascertain its consequence on bread texture. In this purpose, wheat bran fractions having different particle size were added at various levels to wheat flour. Gluten addition was also tested, as a possible mean to balance the effect of bran addition. The effects on the rheological properties of the dough were, investigated at the mixing and proofing steps. Subsequently, the effects on bread texture were assessed by studying the cellular
2.1. Raw materials Commercial wheat bran, a coarse fraction (CB) and a fine fraction (FB), was supplied by Moulins Soufflet (F91-Corbeil-Essonnes). FB fraction was obtained by grinding CB. Median particle size was 1.8 mm (CB) and 18 mm (FB). DF content, according to AOAC 985-2 method, was 47% (CB) and 40% (FB); protein content was 18% for both fractions. Water-binding capacity was 6.1 g/g for CB, 4.0 g/g for FB and both fractions were mainly constituted of insoluble fibres (>90%). Standard commercial bread flour was supplied by Moulins Soufflet (F44-Pornic). Its protein content was 10.5% and initial moisture content 15% (total wet basis). Dry vital gluten (Moulins Soufflet, F91-Corbeil-Essonnes), fresh yeast (Lesaffre, F94-Maisons Alfort), salt (K þ S, Hanovre) and tap water were used in the breadmaking experiments. 2.2. Breadmaking procedure Control dough was obtained by mixing 2000 g wheat flour, 1240 g tap water, 28 g fresh yeast, 28 g salt and 0.04 g acid ascorbic. Bran samples were substituted for flour at levels of 5%, 10%, 15% or 20% on flour weight basis (Table 1). In case of recipe containing dry vital gluten, 4% of the flour was replaced. To take into account water absorption increase due to bran addition, our expert baker adjusted the amount of water added, at the end of pre-mixing stage in view of dough behaviour and mechanical power measurements, in agreement with the French breadmaking procedure (AFNOR standard V03-716). This led to add an amount of water of about half the level of bran to the control water amount (Campbell et al., 2008). The water temperature was adjusted to give a final dough temperature of 25 ± 2 C. Mixing involved two stages: pre-mixing at 80 rpm for 4 min, and texturing at 160 rpm during 7 min, in a spiral mixer (Diosna SP12, Osnabrück, Germany) and the mechanical power supplied to the dough and of the temperature of the dough were recorded. As detailed by Shehzad et al. (2012), mixing conditions were assessed by computing the final temperature increase from the initial value during texturing (DT, C), and the specific mechanical energy (Es, kJ/kg) from the mechanical power Pm:
Es ¼
X
. Pm ðtÞ:Dt M
(1)
where Pm(t) is the average value of the mechanical power at time t, during interval Dt, with SDt ¼ t, total texturing time, and M the total dough mass. Three replicates were carried out for all 18 formulas (Table 1). After a first bulk fermentation of 20 min at 27 C and the relative humidity (rh) of 75% in proofing cabinet (Panem, che), the dough was divided into 350 g pieces. These F79-La Cre pieces were hand-rounded and mechanically moulded (Tregor, F35-Noyal sur Vilaine) and finally proofed for 100 min at 27 C and 75% (rh). After scarification, the dough pieces were baked in a traditional Bongard deck oven for 25 min at 245 C (hearth and vault temperatures), with 300 mL of steam just before and after loading. Breads were cooled at room temperature for 2 h before characterization. 2.3. Dough rheological properties Rheological measurements were performed at large biextensional deformations by Lubricated Squeezing Flow test. At
31.6 19,6 46.4 27.9 53.7 32.9 25.6 20.2 43.4 28.4 78.4 41.2 34.7 23.3 63.7 59.8 155.3 83.9 ± 1.4 ± 2.1 ± 0.3 ± 2.4 ± 0.1 ± 0.8 ± 3.4 ± 0.0 ± 2.6 ± 1.8 ± 7.1 ± 2.2 ± 3.5 ± 3.9 ± 0.7 ± 0.4 ± 1.3 ± 4.1 45.5 50.4 63.7 53.6 47.6 51.8 64.0 65.3 45.2 48.9 48.4 47.7 47.8 47.5 44.9 47.9 51.6 48.4 0.65 0.75 0.65 0.68 0.61 0.72 0.67 0.64 0.67 0.64 0.53 0.59 0.60 0.61 0.58 0.57 0.54 0.54 ± 0.3 ± 0.8 ± 1.9 ± 0.5 ± 0.9 ± 1.8 ± 1.4 ± 1.1 ± 1.2 ± 1.5 ± 1.2 ± 0.2 ± 0.9 ± 0.9 ± 0.2 ± 2.8 ± 0.2 ± 0.1 16.6 13.0 18.0 14.8 16.7 15.3 14.1 14.8 14.9 17.3 21.0 18.7 15.6 15.3 17.2 19.0 20.3 18.3 ± 1.1 ± 0.6 ± 0.1 ± 0.9 ± 0.1 ± 0.5 ± 0.6 ± 1.4 ± 0.2 ± 0.2 ± 0.3 ± 0.3 ± 0.9 ± 0.8 ± 0.2 ± 0.1 ± 0.1 ± 0.2 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 100 96 95 91 90 86 95 91 90 86 85 81 90 86 85 81 80 76 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
62 65 67 70 70 72 68 71 71 73 73 73 71 73 73 74 75 76
0 0 0 0 0 0 5 5 5 5 5 5 10 10 10 10 10 10
0 0 5 5 10 10 0 0 5 5 10 10 0 0 5 5 10 10
kPa.s kJ/kg
23.6 29.2 20.1 24.6 17.1 20.6 21.0 25.6 17.0 22.0 15.7 18.1 18.4 22.7 16.0 19.0 14.9 16.4
()
± 0.01 ± 0.02 ± 0.08 ± 0.00 ± 0.02 ± 0.05 ± 0.02 ± 0.00 ± 0.05 ± 0.02 ± 0.02 ± 0.03 ± 0.06 ± 0.04 ± 0.01 ± 0.01 ± 0.01 ± 0.00
min
0.0067 0.0068 0.0055 0.0061 0.0071 0.007 0.0059 0.0053 0.0074 0.0064 0.0055 0.0057 0.0068 0.0064 0.0063 0.0055 0.0057 0.0057
min
c
1
b a
n
K Es Vital gluten Coarse bran Fine bran Water Flour Formula
Table 1 Dough composition and results of dough and bread properties.
Ingredients are expressed as mass on flour/bran/gluten basis. The increase of bran contents is represented in grayscale with two symbols (- with or C without vital gluten supplementation). Es: specific mechanical energy; K: consistency index; a: porosity increase; b: maximum volume expansion growth rate (SD < 6 104); c: time for inflection point; b0 : starting time of the stationary phase; c0 : the stability at t/þ ∞; Ebread: bread apparent modulus; sr: residual stress; r*: bread density (SD < 0.01). The highest and lowest values of each variable are represented in bold. All measurements were duplicated, except bread density which was measured five times. Minima and maxima values of the variables are indicated in bold italics.
0.18 0.16 0.21 0.18 0.21 0.20 0.21 0.18 0.24 0.19 0.29 0.23 0.26 0.21 0.32 0.28 0.36 0.33 ± 0.4 ± 0.0 ± 0.1 ± 0.0 ± 0.1 ± 0.2 ± 0.0 ± 0.2 ± 0.1 ± 0.2 ± 0.1 ± 0.1 ± 0.1 ± 0.0 ± 0.1 ± 0.1 ± 0.1 ± 0.1 1.76 1.60 2.15 1.50 1.56 1.72 2.05 1.92 2.03 1.63 1.93 1.91 1.81 1.85 2.08 1.93 2.23 1.86 ± 1.0 ± 0.4 ± 2.6 ± 0.2 ± 0.4 ± 0.6 ± 1.7 ± 0.6 ± 0.9 ± 0.2 ± 2.4 ± 1.2 ± 1.1 ± 0.6 ± 3.6 ± 1.9 ± 2.2 ± 1.2 0.13 0.16 0.19 0.19 0.18 0.15 0.12 0.09 0.17 0.13 0.27 0.24 0.13 0.10 0.28 0.23 0.47 0.30
() min
± 1.0 ± 1,1 ± 2.3 ± 1.3 ± 0.4 ± 4.9 ± 3.4 ± 0.0 ± 5.6 ± 3.6 ± 6.0 ± 2.5 ± 1.7 ± 1.4 ± 10 ± 4.9 ±1 ± 8.9
a0 c0 b0
± 0.01 ± 0.01 ± 0.01 ± 0.00 ± 0.02 ± 0.00 ± 0.01 ± 0.00 ± 0.01 ± 0.01 ± 0.04 ± 0.00 ± 0.02 ± 0.01 ± 0.02 ± 0.01 ± 0.02 ± 0.02
6.9 8.0 11.8 6.9 7.8 9.2 15.4 11.5 16.7 11.2 29.3 13.6 10.9 10.6 32.6 18.8 38.2 28.7
g/cm3 10 Pa 10 Pa
5
Ebread
sr
5
r*
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
169
the end of mixing, unyeasted dough samples (z5 g) were loaded in Teflon cylinders (h ¼ 14 mm, Ø ¼ 20 mm), lubricated with paraffin oil (110e230 mPa s at 20 C) and kept at room temperature for 30 min to reach temperature equilibrium (Launay and Michon, 2008). The homogeneous samples were then removed from the cylinders and placed between two parallel plates of Teflon (Ø ¼ 20 mm) lubricated with paraffin oil. The upper plate was attached with movable crosshead of a traction/compression machine (Instron type #1122) equipped with force sensor in the range [0; 100 N] (Instron Corporation, Canton, MA, USA). The cylindrical samples were compressed until a final height of 1 mm, at a constant speed (5, 10 and 100 mm/min). Measurements were repeated 3 times at each speed and 2 replicates were performed for the 18 formulas. The force applied to samples was recorded as a function of displacement. Data were processed according procedure described by van Vliet (2008), in order to represent stressestrain curves for bi-extensional flow and derive the variations of biextensional viscosity, hE, with strain rate, ε_ b , which were fitted by the power law equation (Eq. (2)):
hЕ ¼ K: ð_εb Þn1
for constant strain εb
(2)
where K (Pa.sn) is the consistency index and n the flow behaviour index, n < 1 reflecting a strain-thinning behaviour of the dough.
2.4. Proofing follow-up Images of a rounded dough piece (m ¼ 25 g) during proofing in a controlled ambience (T ¼ 27 C, rh ¼ 75%) were acquired every 5 min through digital camera for 240 min, according to the procedure described in detail by Shehzad et al. (2010). The resulting volume V was converted into porosity, P(t), according to Eq. (3), rs being the density of gas-free dough (rs z 1.23 g/cm3 according to Shehzad et al., 2010). The dough shape ratio was defined by S(t) ¼ H/Lmax, H and Lmax being, respectively, the height and the maximum width of the dough sample at a given time. The porosity curves can be modelled using a Gompertz function, as proposed by Romano et al. (2007) (Eq. (3)):
b e ðt cÞ þ d PðtÞ ¼ 1 m=ðrs :VÞ ¼ a:expð exp a (3) where a is an approximation of the final porosity increase from the initial value, b is the maximum volume expansion growth rate, i.e., the slope at inflection point, c is the time for inflection point, and d is such as (a þ d) ¼ P(t ¼ þ∞) with d<
0 0 0 0 SðtÞ ¼ a c :expt=b þ c
(4)
where a0 and c0 being the stability at t ¼ 0 et t/þ ∞, respectively, (a0 c0 ) is the overall loss of dough stability, and b0 is the starting time of the stationary phase. To avoid any bias due to hand rounding of the dough prior to measurement, all curves were homothetically shifted to the same value of a0 , here 0.6. The exponential model was applied on the first 90 min of the proofing, in agreement with proofing time during breadmaking. For the 18 formulas, two replicates were carried out.
170
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
2.5. Bread characterisation
2.6. Experimental design
2.5.1. Density Bread volume based on seed displacement in a volumeter (Chopin, F92-Villeneuve-la-Garenne) was determined after 2 h cooling. Bread density, r* (g/cm3), was calculated from the ratio mass/volume. This measure was performed five times for each bread recipe.
Three factors were tested: wheat bran content (0%, 5% and 10%) and particle size (1.8 mm and 18 mm), and the level of dry vital gluten addition (0% and 4%). The experimental design resulted in 18 different combinations, allowing the addition of both fine and coarse brans, up to an overall level of 20% (Table 1). Statistical analysis was performed using XLStat software (v2.04, Addinsoft, NY, USA) on several variables of doughs (Es, K, a and a0 c0 ) and breads (r* and Ebread). These variables are selected because they are representative of each operation or property of dough and bread processing: mixing, rheology, porosity and stability kinetics and texture, respectively. It was also checked, thanks to the Spearman correlation matrix (see sup. Material), that they were correlated to other variables, not further taken into account in the statistical analysis. The relative standard deviation for each variable was computed from the replicates of experiments. In order to check whether particle size and content of brans and gluten addition can have an impact on dough properties (energical and rheological) during the breadmaking process (mixing, proofing and baking), and on bread properties, a three-way analysis of variance (ANOVA) was achieved at the 5% significance level. When a significant effect was revealed, formulations were compared using the StudentNewman-Keuls test (SNK) at a significance level of 5%, 1% and 0.1%.
2.5.2. Image analysis The objective was to define crumb grain from visual texture of bread. Images of crumb were acquired in 2D at a macroscopic scale by using a flatbed scanner (Epson Perfection, V100) and covered by a black box. The resolution was 400 dpi. Then, grey level granulometric method from mathematical morphology was applied. As described in detail by Lassoued et al. (2007), mathematical morphology is a set of transformations based on probing images with masks of given size and shape, called “structuring elements”. The basic transformations were erosion and dilatation with squared structuring elements. After n erosion or dilatation, the size of the square is 2nþ1. By applying erosions of increasing size to the original image, bright objects progressively disappeared and a curve of the sum of grey level, V(i) according to the erosion step, i, can be build. Conversely, the value V increased when dilatations were applied and the dark objects progressively disappeared. A granulometric curve was obtained by normalizing the sum of grey level, separately for dilatations and erosions according to:
i . GðiÞ ¼ VðiÞ Vðf Þ 100 ½Vð0Þ VðfÞ
(5)
where i is the dilatation or erosion step, V(0) and V(f) are the sum of grey level for the original image and after the last dilatation or erosion step, respectively. The erosionedilation curve was obtained by assessing the derivative of G for erosion and dilatation step and by collating the two resulting curves from ef to 1 for dilatation and from 1 to f for erosion, f being the largest transformation applied (n f). In the present work, f ¼ 60. These curves were finally analysed by Principal Component Analysis (PCA) in order to map the crumb images. For each recipe, ten vertical and central slices (11 mm-thickness) were selected and 3 replicates were achieved after removing the crust area (thickness z 2 mm) from the image. Morphological analysis was developed using procedure of the rouville-Saint-Clair). CalculaAphelion software (ADCIS, F14, He tions of the granulometric curves and principal components analysis were performed with Matlab 7.1 (The Math Works, F92vres). Se 2.5.3. Texture by multi-indentation The multi-indentation test consisted in measuring the resistance of crust and crumb against compression, as described in detail by Chaunier et al. (2014). About 16 h after cooling, bread was submitted to a compression by a specific device consisting of 2 5 iron made cylindrical pins (indents with Ø ¼ 3 mm and conical tip of 40 ), set on a universal testing machine (Instron type #1122, Instron Corporation, Canton, MA, USA). The length of the bread piece was 7 cm and its maximum height (h0) was measured. The indents were adjusted to the crust surface of the bread. The crosshead was moved down at a speed of 50 mm/min and stopped after a distance equal to 2/3 h0. After compression, the bread was relaxed for 180 s. The apparent modulus, Ebread, derived from the initial slope of the force-signal was mainly attributed to crust stiffness whereas sr, the apparent residual stress at the end of the relaxation step, could be more impacted by crumb firmness. Three replicates were performed for the 18 breads.
3. Results 3.1. Dough behaviour during mixing The evolutions of mechanical power during mixing were very close for the pre-mixing step (240 s), but were quite different for the texturing step (420 s), depending on dough composition, as illustrated in Fig. 1a, for dough #2, #3 and #17. Images of dough at the end of mixing, also presented in Fig. 1a, showed that dough #2 appeared smoother and more homogeneous than those containing brans (#3 and #17), which had a less uniform appearance. Relative errors in determining specific mechanical energy (Es), the initial temperatures (Ti), and at the end of pre-mixing (Tf) and texturing (Tt) steps were less than 5%. A low temperature change was observed during the pre-mixing step (TfTi) for all dough (1.2 ± 0.4 C) whereas larger increase was observed, from 2.7 to 6 C, during texturing step. Largest values of specific mechanical energy (23.6 and 29.3 kJ/kg) were obtained for dough without bran (#1 and #2, respectively), whereas Es varied between 14.9 and 21 kJ/kg (#17 and #7, respectively) for bran-enriched dough, without dry gluten addition (Table 1). The increase of dough temperature during texturing (TtTf) was correlated to Es (Fig. 1b; R2 ¼ 0.91); this result showed that the dough was heated by viscous dissipation. By mixing wheat flour dough under various time and speed conditions, Shehzad et al. (2012) represented this phenomenon by writing:
. Es ¼ h:g_ 2 :Dt rp ¼ Cp :ðTt Tf Þ þ “losses”
(6)
where h, rp, Dt, Cp, were respectively the viscosity, the density, the texturing time and the heat capacity of the dough, and, g_ the average shear rate in the mixer, whereas the term “losses” meant heat losses with environment. The results of ANOVA on Es values (Table 2) confirmed the trends observed in Fig. 1b: adding coarse or fine brans induced a significant decrease of Es; conversely, Es increased with the addition of vital gluten. The addition of brans or gluten significantly modified the dough behaviour during mixing, but the interpretation of these effects was not straightforward. These changes reflected changes of dough rheological behaviour,
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
171
completely immersed and could shear it efficiently. Conversely, for flours added with bran, without gluten supplementation (#9 and #17), the dough had a more heterogeneous appearance, with separated pieces which the mixing arm pushed by distinct blocks, with poor shearing (Fig. 1a). So, these observations suggested that average shear rate during mixing was not constant and it was lower in the latter case (#9 and #17) than in the former (#2 and #10), which, in line with Eq. (6), would explain why lower specific mechanical energies were measured. These effects reflected the influence of bran particles on the gluten network formation; they suggested that gluten network would be less organized when adding bran solid particles. 3.2. Elongational properties The biaxial extension measurements performed at different strain values (0.1 < ε_ b < 1.25) for the 18 formulas, showed that the variations of elongational viscosity (hE) as a function of strain rate ε_ b could be adjusted by a power law, with r2 > 0.97:
hE ¼ K : ε_ n1 b
ðεb ¼ constantÞ
(7)
For strain values εb ¼ 1, the values of K (consistency index) varied between 13 kPa.sn (#2) and 21 kPa.sn (#11) (Table 1) whereas the flow index values (n) varied in a narrower interval (0.36e0.47). These values were in accordance with those reported in the literature (Launay and Michon, 2008). The addition of bran increased significantly K, with a more significant effect of coarse bran (p < 0.001) compared to fine bran (p < 0.05) (Table 2). Indeed, the formation of gluten network could have been hindered because the bran particles reduced flour proteins aggregation. But the physical action of bran, as solid particles acting like a charge in a suspension, seemed to have predominated and increased dough elongational viscosity, as reported by Cavella et al. (2008), and also by BonnandDucasse et al. (2010) for shear viscosity. Moreover, the water binding capacity of the bran could have affected dough viscosity more significantly with coarse brans, which had higher water absorption capacity than fine brans. Finally, the addition of gluten to bran-enriched doughs showed no significant influence on elongational properties (p > 0.05) (Table 2). These results illustrated that the mechanisms underlying viscosity increase related to bran addition involved largely the presence of solid particles, like a charge in a suspension.
▫
Fig. 1. Evolutions of (a) spiral mixer power ( ) and dough temperature (B) during mixing for #2 ( , O), #3 ( , ), #17 (-, C) and their images at the end of the texturing time, and (b), temperature rise as a function of specific energy during texturing from doughs containing different levels of wheat bran (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation.
linked to the presence of wheat brans and gluten. Indeed, if the average shear rate remained constant with bran addition, Eq. (6) suggested that the addition of bran would lead to a smaller dough viscosity, compared to a dough without brans; this result was in contradiction with Bonnand-Ducasse et al. (2010) and Cavella et al. (2008) who reported that viscosity increased with bran addition. This apparent contradiction might be raised by the observation during mixing (Fig. 1a). Comparison of the images of dough at the end of mixing suggested that the agitation of the dough in the mixer was different, depending on its composition. For flours without bran (#2) or added with brans and supplemented in gluten (#10), dough appeared smooth and homogeneous; it remained a continuous mass in which the mixing arm was
3.3. Dough behaviour during proofing The kinetics of porosity and stability of dough during fermentation were determined at macroscopic scale by 2D image followup. The porosity curves had a sigmoid shape with an initial value (t ¼ 0 min) between 0 and 0.25 and a final value (t ¼ 240 min) of 0.7e0.8 (Fig. 2). These curves were well fitted by the Gompertz model (Eq. (3), R2 > 0.97), for all formulas. The relative standard deviation of porosity parameters (a, b, c, d) was less than 7%. The time for inflection point (parameter c) did not vary much, between about 60 min for three doughs with 5% bran added (#3, #7, #8) and
Table 2 Results of three-way ANOVA on specific mechanical energy (Es), consistency index (K), porosity increase (a), loss of stability (a0 c0 ), apparent modulus (Ebread), residual stress (sr) and bread density (r*). Factor Fine bran Coarse bran Gluten
F F F
Es
K
72.43 131.52 144.66þ þ þ
3.86 þ 12.75 þ 2.64ns
a0 c0
a þ þ
18.93 7.88 3.40ns
8.54þ þ 19.44þ þ 3.82ns
Ebread þ
25.73þ þ 7.34þ þ 0.09ns
þ
sr
r*
8.39þ þ þ 14.85þ þ þ 5.89
161.28þ þþ 81.19þ þ þ 52.11
F values and the associated significance of the effect (þor , þþ or and þþþ or indicate p < 0.05, p < 0.01 and p < 0.001, respectively; ns: no-significant) with þ and meaning a positive or negative effect, respectively).
172
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
an average value of 49 ± 2 min for other doughs (Table 1). For t/þ ∞, porosity values (¼ a þ d) were all close to 0.77 (±0.04). The porosity increase a, varied between 0.53 (#11) and 0.73 (#2). The values were close to those encountered by Romano et al. (2007) and Shehzad et al. (2010). They showed an overall decreasing trend with the level of added bran (Fig. 2a). This trend confirmed the one observed by Cavella et al. (2008) showing that higher bran addition (>7%) led to decreased expansion of dough during proofing. An ANOVA on parameter a confirmed that bran contents had a significant negative effect on dough expansion (Table 2). Fine bran caused a larger effect (p < 0.001) compared to coarse bran (p < 0.01), especially between 0 and 5% of bran added. Gluten addition had no significant effect (p > 0.05) on dough expansion. As shown in Fig. 2a, the porosity increase was negatively correlated with the consistency index (R2 ¼ 0.70). This result could be explained by the mechanisms of bubble growth in a viscous medium, applied in the case of dough proofing by Della Valle et al. (2014): a high elongational viscosity decreases the bubble growth, hence limiting dough expansion. During proofing, dough exhibited a continuous decrease of the shape ratio S(t), reflecting a loss of stability. For t 90 min, the curves S(t) were well fitted by an exponential decay (Eq. (3),
R2 > 0.9). The starting time of the stationary phase (b0 ) varied between 19.6 min (#2) to 155 min (#17), with values larger than 60 min for doughs (#11, 15, 16, 17 and 18), i.e. when bran addition 15% (Table 1). This result suggested that a large addition of bran favoured a continuous loss of stability of the dough during proofing. The overall loss of dough stability, evaluated by (a0 c0 ) varied between 0.47 (#17) to 0.09 (#8), with (a0 c0 ) 0.23 when bran addition 15%. This result showed that bran addition increased the loss of stability of the dough during proofing. Results of ANOVA on (a0 -c0 ) (Table 2) confirmed that the bran content had a strong effect (p < 0.001) on the loss of stability and indicated that the addition of coarse bran was more significant (p < 0.001) than addition of fine bran (p < 0.01); gluten addition did not improve significantly the dough stability (p > 0.05). Finally, the loss of stability was only fairly correlated to the consistency index K (R2 ¼ 0.59; not shown), in a positive way. However, elongational viscosity was expected to have a positive effect on the stability of the medium. So, our results rather suggested that the loss of dough stability would be due to the destabilization of the gas bubbles interface by bran particles during dough proofing. Indeed solid particles could pierce the films separating bubbles (Gan et al., 1995), and this effect would be more important with particles of larger size. 3.4. Bread texture 3.4.1. Density As expected, the addition of bran increased significantly density, from r* ¼ 0.16 (#2) to 0.36 g cm3, for 20% bran added (#17) (Tables 1 and 2). Gluten addition decreased significantly (p < 0.05) density of bread with the coarse and/or fine brans (Table 2). The effect of particles size was less dramatic than that of bran content. From the breads enriched with bran, only #4, #8 and #10 samples reached a density close to the one of control bread (r* ¼ 0.18 g cm3), which suggested that a low bran addition (5%), and intermediate (10%) with the two particle sizes, could be balanced by addition of vital gluten. Furthermore, there was a significant correlation between the expansion of dough, or porosity increase, during proofing and final bread density (Fig. 2b, R2 ¼ 0.71): the higher the porosity at the end of proofing, the lower the bread density. This result suggested that baking did not much influence the density of bread enriched with bran. In other words, wheat bran particles had more effect during mixing and proofing than during baking.
Fig. 2. Porosity increase (parameter a of the Gompertz model, Eq. (2)) determined as illustrated in the insert (experimental curves of porosity during fermentation for #2 ( ), #5 ( ) and #17 (:)) (a) as a function of dough consistency index K, and (b) as a function of bread density for different levels of wheat bran (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation.
▫
3.4.2. Crumb cellular structure Different cellular structures were obtained as shown by images in Fig. 3. For example, crumb #1 had a uniform grain and smooth texture. By contrast, sample #17, containing 20% bran, gave a denser and closer crumb with smaller bread volume. Other samples with brans at different content exhibited heterogeneous texture with variations in cell sizes as well as cell wall thickness. In order to ascertain these qualitative observations, image analysis was carried out. Principal component analysis (PCA) was applied in order to compare the resulting granulometric curves and to represent a similarity map of the breads (Fig. 3). The first two components accounted for 95% (PC1 ¼ 78%; PC2 ¼ 17%). Each sample represented the mean coordinates of thirty images. Principal component 1 mainly described differences between sample #1 and other breads, stressing that #1 had a finest crumb and smallest cells size, and thus confirming first observation. Principal component 2 completed the crumb description by providing an indicator of cellular heterogeneity. Indeed, crumbs at the bottom had more regular crumb, although coarse, with cells of similar size and thick cell walls (#17 for instance); conversely, crumbs with irregular structure, containing both large and small cells, with thinner cell
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
Fig. 3. Similarity map for crumb morphology of bran-enriched breads (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation, obtained by principal analysis component (PCA) and images of bread slice (#1, #2, #8, #17). Dotted arrow indicated the direction of bread density increase. Each point included the standard deviation of PC1 and PC2 (30 replications per formula).
walls (like #2 and #8, for instance), were rather located at the top. Finally, the distribution of crumb cellular structures on the similarity map owed much to the density value, as illustrated by the direction of increasing density, and increasing bran content (Fig. 3). Breads with larger bran content had larger density and, consequently, exhibited coarser, but more homogeneous, crumb cellular structures. 3.4.3. Mechanical properties Results from multi-indentation showed that all breads displayed similar texture profiles, with three different sequences (Fig. 4), as described by Chaunier et al. (2014) for breads without fibre: (1) a stress increase up to a value sm, attributed to crust debonding; (2) an enhanced increase of stress till s2/3, reflecting the densification of crumb before the compression was stopped; (3) a decrease in stress during relaxation of the crumb to reach a stable value, sr, ranging between 1.5.105 Pa (#4) and 2.2.105 Pa (#17) (Table 1). The
Fig. 4. Apparent modulus of breads as a function of breads density for various contents of bran added (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation. The insert shows two experimental curves obtained during multi-indentation test and illustrates the determination of main mechanical properties.
173
apparent modulus of bread, related to crust stiffness, Ebread, varied from 6.9.105 Pa for the control bread, without bran (#1), and 38.2.105 Pa for bread with 20% bran added (#17) (Table 1). The relative standard deviation of these four variables was ±16% for all breads. These results underlined the influence of bran on the whole texture of breads, including both crust and crumb, and they were confirmed by an ANOVA on these mechanical variables (Ebread and sr) showing that bran content strongly influenced (p < 0.05) the texture of breads (Table 2), leading to stiffer crust and firmer crumb. The mechanical variables Ebread and sr were found to be correlated (R2 > 0.80). This result was in agreement with those from Chaunier et al. (2014) for bread of the same format. In our case, this result suggested that the texture contrast of breads, defined by the ratio Ebread/sr, was not modified by the addition of bran. The high correlation of Ebread with r* (R2 ¼ 0.83; Fig. 4) confirmed that the increase of mechanical properties, and consequently of crust stiffness and crumb firmness, was mainly due to an increase of density, induced by bran addition. This fitting was in agreement with the GibsoneAshby's model for cellular solids or solid foams (1997). It led to a value of the power law index (n ¼ 2.1) close to the one found by Poutanen et al. (2014) for crumbs of pan breads enriched with bran (n ¼ 2.3). This result confirmed that the density increase was the main factor involved in the texture changes of breads enriched with wheat bran. 4. About the mechanisms of dough densification by bran addition The last results of the present work (3.4) have confirmed that the texture changes of bread enriched with wheat bran were mainly governed by the density increase, and that this increase was already expected from a decrease of dough expansion during proofing (3.3). So we wanted to focus on the mechanisms of this decrease, and especially how it involved the rheological properties of dough. In this study, different particle size and level of wheat bran have been employed with or without vital gluten supplementation, and the three factors experimental design led to a large variation of rheological properties and bread texture. The influence of bran addition on the dough rheological properties has first been reflected by the behaviour of dough during mixing and the decrease of Es with bran addition; they were commonly attributed to a deficient formation of gluten network (Lai et al., 1989; Gomez et al., 2003), which was confirmed here by the significant effect of gluten addition. The detrimental effects of bran on gluten network reviewed by Noort et al. (2010) were: (1) particles steric hindrance, or physical barrier, (2) preferential water binding, (3) release of ferulic acid that cross-links arabinoxylans with gluten proteins. This third effect could appear in contradiction with findings of Wang et al. (2002, 2003) who reported that the gluten network became more resistant to extension by the action of ferulic acid. The first effect, already suggested by Courtin and Delcour (2002) and Katina et al. (2006), did not seem so important in our case, since bran particle size had no significant effect on Es (Table 2). The influence of the second effect has been reduced by adjusting the level of added water (Table 1), so any deficiency of gluten network formation could be attributed mostly to the third effect, i.e. bran chemical reactivity. However, negative effects on gluten formation would not explain the increase of dough elongational viscosity with bran addition, reported by Cavella et al. (2008), and also by BonnandDucasse et al. (2010) for shear viscosity, since a weaker network should lead to lower viscosity. Indeed, dough containing bran could be considered as a suspension of solid particles, bran acting as a filler in a viscoelastic medium, and its viscosity could be predicted by a suspension model, as suggested by Poutanen et al. (2014).
174
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
Nevertheless, according to such a model, the viscosity of a suspension containing 20% of solid particles should be increased by a factor 2, in comparison to the viscosity of the viscoelastic matrix, here the dough. In our case, elongational viscosity of dough with 20% bran, was increased by a factor 1.5, when comparing samples #1, 2 with #17, 18, respectively, which was less than expected by the suspension model. So, we might conclude that the changes of elongational properties of dough with wheat bran mainly resulted from a filler effect, due to solid particles addition, which were partly balanced by the negative effects, induced by a deficiency of gluten network formation. The increase of viscosity well explained the lower increase of porosity during fermentation (Fig. 2a), but did not explain the increased loss of stability, since more viscous dough should be more stable. Courtin and Delcour (2002) also attributed the default of dough development to gas cells piercing by bran particles. Indeed, for higher dough porosity (0.5), during fermentation, gas bubbles become connected to each other and would be separated by thin liquid films (Sroan and MacRitchie, 2009; Turbin-Orger et al., 2012) of lower thickness (10 mm) than bran particles. Hence, bran particles could destabilize gas bubbles interface in fermented dough favouring bubbles coalescence and causing dough collapse. This interpretation is consistent with observations obtained on the dough microstructure by X ray microtomography (Turbin-Orger et al., 2012), which showed that the gas cells contained in bran-enriched doughs had very irregular, polyhedral shapes, but clearly these observations would need quantification to confirm this interpretation. Finally, it could also explain why the loss of stability was more significant for bran of larger particle size. From a practical point of view, the proofing time could be reduced to limit the loss of dough stability. However, this action would lead to a decrease of porosity, and therefore a lower dough expansion, and, consequently a higher bread density. 5. Conclusion By adding up to 20% of bran with large and small particle sizes, we have covered a wide range of dough behaviour and rheological properties. Our results confirmed that the main effect of bran addition was to increase bread density, which induced an increase of the mechanical properties of bread, i.e. stiffer crust and firmer crumb. The changes of dough behaviour during fermentation with bran addition drew essential consequences on the formation of the cellular structure and the density of the bread. The decreased expansion and increased loss of stability during proofing could largely be explained by the changes of extensional properties of the dough at the end of mixing, and to a less extent by the direct effect of bran particles on the films separating gas bubbles. Finally, the negative effect of bran addition on gluten network formation could be partly balanced by its role as a filler like solid particles suspended in the dough. Acknowledgements This work was carried out within Pan&Sens program granted by the Region Pays de Loire among which all partners are gratefully guerre for her acknowledged. The authors are thankful to A.L. Re assistance in image analysis. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jcs.2015.06.014.
References Anderson, J.W., Baird, P., Davis, R.H., Ferreri, S., Knudtson, M., Koraym, A., Waters, V., Williams, C.L., 2009. Health benefits of dietary fiber. Nutr. Rev. 67 (4), 188e205. Bloksma, A.H., 1990. Dough structure, dough rheology, and baking quality. Cereal Foods World 35 (2), 237e244. Bonnand-Ducasse, M., Della Valle, G., Lefebvre, J., Saulnier, L., 2010. Effect of wheat dietary fibres on bread dough development and rheological properties. J. Cereal Sci. 52 (2), 200e206. Campbell, G.M., Ross, M., Motoi, L., 2008. Bran in bread: effects of particle size and level of wheat and oat bran on mixing, proving and baking. In: Campbell, G.M., Scanlon, M.G., Pyle, D.L. (Eds.), Bubbles in Food 2-Novelty, Health and Luxury. Eagan Press, St Paul, USA, pp. 337e354. Cavella, S., Romano, A., Giancone, T., Masi, P., 2008. The influence of dietary fibres on bubble development during bread making. In: Campbell, G.M., Scanlon, M.G., Pyle, D.L. (Eds.), Bubbles in Food 2-Novelty, Health and Luxury. Eagan Press, St Paul, USA, pp. 311e322. Chaunier, L., Chiron, H., Della Valle, G., Rouaud, O., Rzigue, A., Shehzad, A., 2014. Assessment of French bread texture by a multi-indentation test. J. Food Eng. 122, 92e98. Courtin, C.M., Delcour, J.A., 2002. Arabinoxylans and endoxylanases in wheat flour bread-making. J. Cereal Sci. 35, 225e243. Della Valle, G., Chiron, H., Cicerelli, L., Kansou, K., Katina, K., Ndiaye, A., Whitworth, M., Poutanen, K., 2014. Basic knowledge models for the design of bread texture. Trends Food Sci. Technol. 36, 5e14. Galliard, T., Gallagher, D.M., 1988. The effects of wheat bran particle size and storage period on bran flavour and baking quality of bran/flour blends. J. Cereal Sci. 8, 147e154. Gan, Z., Ellis, P.R., Schfield, J.D., 1995. Mini review: gas cell stabilisation and gas retention in wheat bread dough. J. Cereal Sci. 21 (3), 215e230. Gibson, L.J., Ashby, M.F., 1997. Cellular Solids, Structure and Properties. Cambridge Press University. Gomez, M., Ronda, F., Blanco, C.A., Caballero, P.A., Apesteguia, A., 2003. Effect of dietary fibre on dough rheology and bread quality. Eur. Food Res. Technol. 216, 51e56. Katina, K., Salmenkallio-Marttila, M., Partanen, R., Forssell, P., Autio, K., 2006. Effects of sourdough and enzymes on staling of high-fibre wheat bread. LWT e Food Sci. Technol. 39 (5), 479e491. Lai, C.S., Hoseney, R.C., Davis, A.B., 1989. Effects of wheat bran in breadmaking. Cereal Chem. 66, 217e219. guerre, A.L., 2007. GranLassoued, N., Babin, P., Della Valle, G., Devaux, M.F., Re ulometry of bread crumb grain: contributions of 2D and 3D image analysis at different scale. Food Res. Int. 40 (8), 1087e1097. Launay, B., Michon, C., 2008. Biaxial extension of wheat flour dough: lubricated squeezing flow and stress relaxation properties. J. Texture Stud. 39, 496e529. Noort, M.W.J., van Haaster, D., Hemery, Y., Schols, H.A., Hamer, R.J., 2010. The effect of particle size of wheat bran fractions on bread quality e evidence for fibreprotein interactions. J. Cereal Sci. 52 (1), 59e64. Pomeranz, Y., Shogren, M.D., Finney, K.F., Bechtel, D.B., 1977. Fiber in breadmaking e effects on functional properties. Cereal Chem. 54 (1), 25e41. Poutanen, K., Sozer, N., Della Valle, G., 2014. How can technology help to deliver more of grain in cereal foods for a healthy diet? J. Cereal Sci. 59 (3), 327e336. Romano, A., Toraldo, G., Cavella, S., Masi, P., 2007. Description of leavening of bread dough with mathematical modelling. J. Food Eng. 83, 142e148. Salmenkallio-Marttila, M., Katina, K., Autio, K., 2001. Effects of bran fermentation on quality and microstructure of high-fiber wheat bread. Cereal Chem. 78, 429e435. Shehzad, A., Chiron, H., Della Valle, G., Kansou, K., Lamrini, B., Lourdin, D., 2012. Energetical and rheological approaches of wheat flour during mixing with a spiral mixer. J. Food Eng. 110 (1), 60e70. guerre, A.L., 2010. Shehzad, A., Chiron, H., Della Valle, G., Kansou, K., Ndiaye, A., Re Porosity and stability of bread dough during proofing determined by video image analysis for different compositions and mixing conditions. Food Res. Int. 43 (8), 1995e2005. Shogren, M.D., Pomeranz, Y., Finney, K.F., 1980. Counteracting the deleterious effects of fiber in breadmaking. Cereal Chem. 58, 142e144. Sroan, B.S., MacRitchie, F., 2009. Mechanism of gas cell stabilization in breadmaking. II. The secondary liquid lamellae. J. Cereal Sci. 49, 41e46. guerre, A.L., Turbin-Orger, A., Boller, E., Chaunier, L., Chiron, H., Della Valle, G., Re 2012. Kinetics of bubble growth in wheat flour dough during proofing studied by computed X-ray micro-tomography. J. Cereal Sci. 56 (3), 676e683. van Vliet, T., 2008. Strain hardening as an indicator of bread-making performance: a review with discussion. J. Cereal Sci. 48 (1), 1e9. Wang, M., Hamer, R.J., van Vliet, T., Oudgenoeg, G., 2002. Interaction of water extractable pentosans with gluten protein: effect on dough properties and gluten quality. J. Cereal Sci. 36 (1), 25e37. Wang, M., Hamer, R.J., van Vliet, T., Gruppen, H., Marseille, H., Weegels, P.L., 2003. Effect of water unextractable solids on gluten formation and properties: mechanistic considerations. J. Cereal Sci. 37 (1), 55e64. Zhang, D., Moore, W.R., 1999. Wheat bran particle size effects on bread baking performance and quality. J. Sci. Food Agric. 79 (6), 805e809.
Contents lists available at ScienceDirect
Journal of Cereal Science journal homepage: www.elsevier.com/locate/jcs
Rheological properties of wheat flour dough and French bread enriched with wheat bran F. Le Bleis a, b, L. Chaunier a, H. Chiron a, G. Della Valle a, *, L. Saulnier a a b
INRA, UR 1268 Biopolym eres Interactions Assemblages (BIA), 44 316 Nantes, France Food Development, Conseils en Innovation Alimentaire, 44 316 Nantes, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 November 2014 Received in revised form 3 June 2015 Accepted 27 June 2015 Available online 16 July 2015
Wheat flour doughs were elaborated with wheat-bran in various contents, up to 20%, and particle sizes of fractions, in order to study the specific role of rheological properties in processing high fibre breads. The addition of wheat bran, especially more than 10%, decreased the specific mechanical energy developed by the mixer, which was attributed to a deficient formation of the gluten network. It increased the elongational viscosity of the dough, measured by biaxial extension tests, likely through a solid particles effect. These changes explained the lower increase of porosity during proofing, assessed by digital camera and 2D image analysis. The loss of dough stability was rather attributed to the destabilizing effect of bran particles on the films separating gas bubbles. The resulting changes of bread texture, determined by image analysis and mechanical testing of breads, including crust and crumb, were governed by bread density, which was established at the end of proofing. These results help to understand the impact of wheat bran on dough rheological properties in order to design French breads with increased fibre content. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Density Extensional viscosity Gluten Particle size Texture
1. Introduction High dietary fibre diets and whole grain consumption have been associated with a lower risk of various diet related diseases such as cardiovascular disease, diabetes, hypertension, obesity, and gastrointestinal disorders (Anderson et al., 2009). For a healthy diet the recommended daily dose of dietary fibre (DF) is at least 25 g/ day that is far higher than the consumption observed in most western countries. Cereals products and especially bread are staple food that provides opportunities to deliver health benefits to large populations and therefore bread is a priority target for enrichment in dietary fibre. Among the numerous sources of DF, the most prevalent insoluble DF source is the wheat bran of the cereal kernel (Galliard and Gallagher, 1988). However, the addition of wheat bran causes severe problems on dough rheology, texture and sensory quality of bread (Pomeranz et al., 1977). Addition of bran in bread production has been widely studied, but its effect on dough rheological behaviour, proofing and baking has not been fully explored. The breadmaking process is based on three main operations: (1) mixing, in which ingredients are transformed into a macroscopic
* Corresponding author. E-mail address: [email protected] (G. Della Valle). http://dx.doi.org/10.1016/j.jcs.2015.06.014 0733-5210/© 2015 Elsevier Ltd. All rights reserved.
homogeneous medium with viscoelastic properties, mainly due to gluten network formation; (2) proofing step over which dough expands due to gas production by yeast activity, until porosity reaches about 70%; gas retention capacity of the dough depends on the rheological properties of the gluten matrix built during mixing (van Vliet, 2008); (3) baking does not increase much the porosity but sets the cellular structure by dough/crumb transition and crust formation (Bloksma, 1990). Density, or inversely, specific volume, is a common technological target. Numerous negative effects of bran addition on dough processing and breads properties have been reported such as: dough stickiness and loaf weight increase, mixing and fermentation tolerances decrease, specific volume reduction, coarser crumb texture, darker crumb colour and reduced crumb softness (Poutanen et al., 2014). Loss of crust crispiness is also inferred in the case of French bread (Chaunier et al., 2014). But the main problems are the major increase of density and crumb elasticity (Pomeranz et al., 1977), which seem to explain the lower consumer's acceptability of dietary fibre-enriched breads (Zhang and Moore, 1999). Different mechanisms have been inferred to explain the effects of bran addition: bran particles would weaken gluten network (Pomeranz et al., 1977) rather than dilute it (Galliard and Gallagher, 1988), by preventing protein aggregation during mixing (Noort et al., 2010). This could possibly be due to ferulic acid release, size depletion
168
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
structure and the mechanical properties of the crumb. Nomenclature 2. Materials and methods a, b, c, d a0 , b0 , c0 Ebread Es K M n P(t) Pm(t) S(t) Tf, Tt
DT εb ε_ b
hE rs r* sm, s2/3,
g_
Parameters of the Gompertz model for porosity Parameters of the exponential decay for stability Apparent modulus of bread (Pa) Specific mechanical energy (kJ/kg) Consistency index (Pa.sn) Total dough mass during mixing (kg) Flow index () Dough porosity () Value of mechanical power during mixing (kW) Dough shape ratio or stability () Dough temperature after pre-mixing and texturing steps, respectively ( C) Temperature rise during mixing ¼ (TtTf) ( C) Elongational strain (biaxial) () Elongational strain rate (biaxial) (s1) Elongational viscosity (Pa s) Density of gas-free dough Density (g cm3) sr Apparent stress at the end of relaxation (Pa) Shear rate in the mixer (s1)
effects, and competition with gluten for water due to their high water binding capacities (Wang et al., 2002, 2003); in addition, during proofing, bran particles could also restrict gas cells expansion (Gan et al., 1995), destabilize the interface between gas bubbles in the fermented dough, thus limiting dough expansion (Cavella et al., 2008), while larger particles may also pierce gas cells (Courtin and Delcour, 2002). As solid particles act like a charge in a suspension, bran addition also increases the extensional and shear viscosities (Cavella et al., 2008; Bonnand-Ducasse et al., 2010), which is unfavourable to the growth of the gas cells. Campbell et al. (2008) also suggested that bran acted during baking rather than during proofing by releasing extra water available for starch gelatinisation, and thereby lowering the final bread volume. Various means can help to reduce these drawbacks. Reducing particle size has been claimed to increase bread volume (Lai et al., 1989), but opposite effects have been observed by Noort et al. (2010) and Zhang and Moore (1999). Noort et al. (2010) explained this negative impact by an increase of surface interaction and water absorption rate, and also liberation of reactive compounds diminishing aggregation of gluten proteins. Pre-treatments of bran, such as pre-fermentation or heat treatments (Zhang and Moore, 1999; Salmenkallio-Marttila et al., 2001), and addition of vital gluten, or the use of surfactants and enzymes (Shogren et al., 1980), can also minimize the negative effect of bran addition on dough rheology and bread volume. Finally, literature survey reveals that the effects of the incorporation of dietary fibres on dough and bread properties depend on the type of bran, the addition level and the breadmaking process. Meanwhile, very few studies addressed the effect of bran addition on the dough properties relevant at different steps of the breadmaking process. In this context, the aim of the study was to determine the influence of wheat bran addition on wheat flour dough rheological properties, in order to ascertain its consequence on bread texture. In this purpose, wheat bran fractions having different particle size were added at various levels to wheat flour. Gluten addition was also tested, as a possible mean to balance the effect of bran addition. The effects on the rheological properties of the dough were, investigated at the mixing and proofing steps. Subsequently, the effects on bread texture were assessed by studying the cellular
2.1. Raw materials Commercial wheat bran, a coarse fraction (CB) and a fine fraction (FB), was supplied by Moulins Soufflet (F91-Corbeil-Essonnes). FB fraction was obtained by grinding CB. Median particle size was 1.8 mm (CB) and 18 mm (FB). DF content, according to AOAC 985-2 method, was 47% (CB) and 40% (FB); protein content was 18% for both fractions. Water-binding capacity was 6.1 g/g for CB, 4.0 g/g for FB and both fractions were mainly constituted of insoluble fibres (>90%). Standard commercial bread flour was supplied by Moulins Soufflet (F44-Pornic). Its protein content was 10.5% and initial moisture content 15% (total wet basis). Dry vital gluten (Moulins Soufflet, F91-Corbeil-Essonnes), fresh yeast (Lesaffre, F94-Maisons Alfort), salt (K þ S, Hanovre) and tap water were used in the breadmaking experiments. 2.2. Breadmaking procedure Control dough was obtained by mixing 2000 g wheat flour, 1240 g tap water, 28 g fresh yeast, 28 g salt and 0.04 g acid ascorbic. Bran samples were substituted for flour at levels of 5%, 10%, 15% or 20% on flour weight basis (Table 1). In case of recipe containing dry vital gluten, 4% of the flour was replaced. To take into account water absorption increase due to bran addition, our expert baker adjusted the amount of water added, at the end of pre-mixing stage in view of dough behaviour and mechanical power measurements, in agreement with the French breadmaking procedure (AFNOR standard V03-716). This led to add an amount of water of about half the level of bran to the control water amount (Campbell et al., 2008). The water temperature was adjusted to give a final dough temperature of 25 ± 2 C. Mixing involved two stages: pre-mixing at 80 rpm for 4 min, and texturing at 160 rpm during 7 min, in a spiral mixer (Diosna SP12, Osnabrück, Germany) and the mechanical power supplied to the dough and of the temperature of the dough were recorded. As detailed by Shehzad et al. (2012), mixing conditions were assessed by computing the final temperature increase from the initial value during texturing (DT, C), and the specific mechanical energy (Es, kJ/kg) from the mechanical power Pm:
Es ¼
X
. Pm ðtÞ:Dt M
(1)
where Pm(t) is the average value of the mechanical power at time t, during interval Dt, with SDt ¼ t, total texturing time, and M the total dough mass. Three replicates were carried out for all 18 formulas (Table 1). After a first bulk fermentation of 20 min at 27 C and the relative humidity (rh) of 75% in proofing cabinet (Panem, che), the dough was divided into 350 g pieces. These F79-La Cre pieces were hand-rounded and mechanically moulded (Tregor, F35-Noyal sur Vilaine) and finally proofed for 100 min at 27 C and 75% (rh). After scarification, the dough pieces were baked in a traditional Bongard deck oven for 25 min at 245 C (hearth and vault temperatures), with 300 mL of steam just before and after loading. Breads were cooled at room temperature for 2 h before characterization. 2.3. Dough rheological properties Rheological measurements were performed at large biextensional deformations by Lubricated Squeezing Flow test. At
31.6 19,6 46.4 27.9 53.7 32.9 25.6 20.2 43.4 28.4 78.4 41.2 34.7 23.3 63.7 59.8 155.3 83.9 ± 1.4 ± 2.1 ± 0.3 ± 2.4 ± 0.1 ± 0.8 ± 3.4 ± 0.0 ± 2.6 ± 1.8 ± 7.1 ± 2.2 ± 3.5 ± 3.9 ± 0.7 ± 0.4 ± 1.3 ± 4.1 45.5 50.4 63.7 53.6 47.6 51.8 64.0 65.3 45.2 48.9 48.4 47.7 47.8 47.5 44.9 47.9 51.6 48.4 0.65 0.75 0.65 0.68 0.61 0.72 0.67 0.64 0.67 0.64 0.53 0.59 0.60 0.61 0.58 0.57 0.54 0.54 ± 0.3 ± 0.8 ± 1.9 ± 0.5 ± 0.9 ± 1.8 ± 1.4 ± 1.1 ± 1.2 ± 1.5 ± 1.2 ± 0.2 ± 0.9 ± 0.9 ± 0.2 ± 2.8 ± 0.2 ± 0.1 16.6 13.0 18.0 14.8 16.7 15.3 14.1 14.8 14.9 17.3 21.0 18.7 15.6 15.3 17.2 19.0 20.3 18.3 ± 1.1 ± 0.6 ± 0.1 ± 0.9 ± 0.1 ± 0.5 ± 0.6 ± 1.4 ± 0.2 ± 0.2 ± 0.3 ± 0.3 ± 0.9 ± 0.8 ± 0.2 ± 0.1 ± 0.1 ± 0.2 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 100 96 95 91 90 86 95 91 90 86 85 81 90 86 85 81 80 76 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
62 65 67 70 70 72 68 71 71 73 73 73 71 73 73 74 75 76
0 0 0 0 0 0 5 5 5 5 5 5 10 10 10 10 10 10
0 0 5 5 10 10 0 0 5 5 10 10 0 0 5 5 10 10
kPa.s kJ/kg
23.6 29.2 20.1 24.6 17.1 20.6 21.0 25.6 17.0 22.0 15.7 18.1 18.4 22.7 16.0 19.0 14.9 16.4
()
± 0.01 ± 0.02 ± 0.08 ± 0.00 ± 0.02 ± 0.05 ± 0.02 ± 0.00 ± 0.05 ± 0.02 ± 0.02 ± 0.03 ± 0.06 ± 0.04 ± 0.01 ± 0.01 ± 0.01 ± 0.00
min
0.0067 0.0068 0.0055 0.0061 0.0071 0.007 0.0059 0.0053 0.0074 0.0064 0.0055 0.0057 0.0068 0.0064 0.0063 0.0055 0.0057 0.0057
min
c
1
b a
n
K Es Vital gluten Coarse bran Fine bran Water Flour Formula
Table 1 Dough composition and results of dough and bread properties.
Ingredients are expressed as mass on flour/bran/gluten basis. The increase of bran contents is represented in grayscale with two symbols (- with or C without vital gluten supplementation). Es: specific mechanical energy; K: consistency index; a: porosity increase; b: maximum volume expansion growth rate (SD < 6 104); c: time for inflection point; b0 : starting time of the stationary phase; c0 : the stability at t/þ ∞; Ebread: bread apparent modulus; sr: residual stress; r*: bread density (SD < 0.01). The highest and lowest values of each variable are represented in bold. All measurements were duplicated, except bread density which was measured five times. Minima and maxima values of the variables are indicated in bold italics.
0.18 0.16 0.21 0.18 0.21 0.20 0.21 0.18 0.24 0.19 0.29 0.23 0.26 0.21 0.32 0.28 0.36 0.33 ± 0.4 ± 0.0 ± 0.1 ± 0.0 ± 0.1 ± 0.2 ± 0.0 ± 0.2 ± 0.1 ± 0.2 ± 0.1 ± 0.1 ± 0.1 ± 0.0 ± 0.1 ± 0.1 ± 0.1 ± 0.1 1.76 1.60 2.15 1.50 1.56 1.72 2.05 1.92 2.03 1.63 1.93 1.91 1.81 1.85 2.08 1.93 2.23 1.86 ± 1.0 ± 0.4 ± 2.6 ± 0.2 ± 0.4 ± 0.6 ± 1.7 ± 0.6 ± 0.9 ± 0.2 ± 2.4 ± 1.2 ± 1.1 ± 0.6 ± 3.6 ± 1.9 ± 2.2 ± 1.2 0.13 0.16 0.19 0.19 0.18 0.15 0.12 0.09 0.17 0.13 0.27 0.24 0.13 0.10 0.28 0.23 0.47 0.30
() min
± 1.0 ± 1,1 ± 2.3 ± 1.3 ± 0.4 ± 4.9 ± 3.4 ± 0.0 ± 5.6 ± 3.6 ± 6.0 ± 2.5 ± 1.7 ± 1.4 ± 10 ± 4.9 ±1 ± 8.9
a0 c0 b0
± 0.01 ± 0.01 ± 0.01 ± 0.00 ± 0.02 ± 0.00 ± 0.01 ± 0.00 ± 0.01 ± 0.01 ± 0.04 ± 0.00 ± 0.02 ± 0.01 ± 0.02 ± 0.01 ± 0.02 ± 0.02
6.9 8.0 11.8 6.9 7.8 9.2 15.4 11.5 16.7 11.2 29.3 13.6 10.9 10.6 32.6 18.8 38.2 28.7
g/cm3 10 Pa 10 Pa
5
Ebread
sr
5
r*
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
169
the end of mixing, unyeasted dough samples (z5 g) were loaded in Teflon cylinders (h ¼ 14 mm, Ø ¼ 20 mm), lubricated with paraffin oil (110e230 mPa s at 20 C) and kept at room temperature for 30 min to reach temperature equilibrium (Launay and Michon, 2008). The homogeneous samples were then removed from the cylinders and placed between two parallel plates of Teflon (Ø ¼ 20 mm) lubricated with paraffin oil. The upper plate was attached with movable crosshead of a traction/compression machine (Instron type #1122) equipped with force sensor in the range [0; 100 N] (Instron Corporation, Canton, MA, USA). The cylindrical samples were compressed until a final height of 1 mm, at a constant speed (5, 10 and 100 mm/min). Measurements were repeated 3 times at each speed and 2 replicates were performed for the 18 formulas. The force applied to samples was recorded as a function of displacement. Data were processed according procedure described by van Vliet (2008), in order to represent stressestrain curves for bi-extensional flow and derive the variations of biextensional viscosity, hE, with strain rate, ε_ b , which were fitted by the power law equation (Eq. (2)):
hЕ ¼ K: ð_εb Þn1
for constant strain εb
(2)
where K (Pa.sn) is the consistency index and n the flow behaviour index, n < 1 reflecting a strain-thinning behaviour of the dough.
2.4. Proofing follow-up Images of a rounded dough piece (m ¼ 25 g) during proofing in a controlled ambience (T ¼ 27 C, rh ¼ 75%) were acquired every 5 min through digital camera for 240 min, according to the procedure described in detail by Shehzad et al. (2010). The resulting volume V was converted into porosity, P(t), according to Eq. (3), rs being the density of gas-free dough (rs z 1.23 g/cm3 according to Shehzad et al., 2010). The dough shape ratio was defined by S(t) ¼ H/Lmax, H and Lmax being, respectively, the height and the maximum width of the dough sample at a given time. The porosity curves can be modelled using a Gompertz function, as proposed by Romano et al. (2007) (Eq. (3)):
b e ðt cÞ þ d PðtÞ ¼ 1 m=ðrs :VÞ ¼ a:expð exp a (3) where a is an approximation of the final porosity increase from the initial value, b is the maximum volume expansion growth rate, i.e., the slope at inflection point, c is the time for inflection point, and d is such as (a þ d) ¼ P(t ¼ þ∞) with d<
0 0 0 0 SðtÞ ¼ a c :expt=b þ c
(4)
where a0 and c0 being the stability at t ¼ 0 et t/þ ∞, respectively, (a0 c0 ) is the overall loss of dough stability, and b0 is the starting time of the stationary phase. To avoid any bias due to hand rounding of the dough prior to measurement, all curves were homothetically shifted to the same value of a0 , here 0.6. The exponential model was applied on the first 90 min of the proofing, in agreement with proofing time during breadmaking. For the 18 formulas, two replicates were carried out.
170
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
2.5. Bread characterisation
2.6. Experimental design
2.5.1. Density Bread volume based on seed displacement in a volumeter (Chopin, F92-Villeneuve-la-Garenne) was determined after 2 h cooling. Bread density, r* (g/cm3), was calculated from the ratio mass/volume. This measure was performed five times for each bread recipe.
Three factors were tested: wheat bran content (0%, 5% and 10%) and particle size (1.8 mm and 18 mm), and the level of dry vital gluten addition (0% and 4%). The experimental design resulted in 18 different combinations, allowing the addition of both fine and coarse brans, up to an overall level of 20% (Table 1). Statistical analysis was performed using XLStat software (v2.04, Addinsoft, NY, USA) on several variables of doughs (Es, K, a and a0 c0 ) and breads (r* and Ebread). These variables are selected because they are representative of each operation or property of dough and bread processing: mixing, rheology, porosity and stability kinetics and texture, respectively. It was also checked, thanks to the Spearman correlation matrix (see sup. Material), that they were correlated to other variables, not further taken into account in the statistical analysis. The relative standard deviation for each variable was computed from the replicates of experiments. In order to check whether particle size and content of brans and gluten addition can have an impact on dough properties (energical and rheological) during the breadmaking process (mixing, proofing and baking), and on bread properties, a three-way analysis of variance (ANOVA) was achieved at the 5% significance level. When a significant effect was revealed, formulations were compared using the StudentNewman-Keuls test (SNK) at a significance level of 5%, 1% and 0.1%.
2.5.2. Image analysis The objective was to define crumb grain from visual texture of bread. Images of crumb were acquired in 2D at a macroscopic scale by using a flatbed scanner (Epson Perfection, V100) and covered by a black box. The resolution was 400 dpi. Then, grey level granulometric method from mathematical morphology was applied. As described in detail by Lassoued et al. (2007), mathematical morphology is a set of transformations based on probing images with masks of given size and shape, called “structuring elements”. The basic transformations were erosion and dilatation with squared structuring elements. After n erosion or dilatation, the size of the square is 2nþ1. By applying erosions of increasing size to the original image, bright objects progressively disappeared and a curve of the sum of grey level, V(i) according to the erosion step, i, can be build. Conversely, the value V increased when dilatations were applied and the dark objects progressively disappeared. A granulometric curve was obtained by normalizing the sum of grey level, separately for dilatations and erosions according to:
i . GðiÞ ¼ VðiÞ Vðf Þ 100 ½Vð0Þ VðfÞ
(5)
where i is the dilatation or erosion step, V(0) and V(f) are the sum of grey level for the original image and after the last dilatation or erosion step, respectively. The erosionedilation curve was obtained by assessing the derivative of G for erosion and dilatation step and by collating the two resulting curves from ef to 1 for dilatation and from 1 to f for erosion, f being the largest transformation applied (n f). In the present work, f ¼ 60. These curves were finally analysed by Principal Component Analysis (PCA) in order to map the crumb images. For each recipe, ten vertical and central slices (11 mm-thickness) were selected and 3 replicates were achieved after removing the crust area (thickness z 2 mm) from the image. Morphological analysis was developed using procedure of the rouville-Saint-Clair). CalculaAphelion software (ADCIS, F14, He tions of the granulometric curves and principal components analysis were performed with Matlab 7.1 (The Math Works, F92vres). Se 2.5.3. Texture by multi-indentation The multi-indentation test consisted in measuring the resistance of crust and crumb against compression, as described in detail by Chaunier et al. (2014). About 16 h after cooling, bread was submitted to a compression by a specific device consisting of 2 5 iron made cylindrical pins (indents with Ø ¼ 3 mm and conical tip of 40 ), set on a universal testing machine (Instron type #1122, Instron Corporation, Canton, MA, USA). The length of the bread piece was 7 cm and its maximum height (h0) was measured. The indents were adjusted to the crust surface of the bread. The crosshead was moved down at a speed of 50 mm/min and stopped after a distance equal to 2/3 h0. After compression, the bread was relaxed for 180 s. The apparent modulus, Ebread, derived from the initial slope of the force-signal was mainly attributed to crust stiffness whereas sr, the apparent residual stress at the end of the relaxation step, could be more impacted by crumb firmness. Three replicates were performed for the 18 breads.
3. Results 3.1. Dough behaviour during mixing The evolutions of mechanical power during mixing were very close for the pre-mixing step (240 s), but were quite different for the texturing step (420 s), depending on dough composition, as illustrated in Fig. 1a, for dough #2, #3 and #17. Images of dough at the end of mixing, also presented in Fig. 1a, showed that dough #2 appeared smoother and more homogeneous than those containing brans (#3 and #17), which had a less uniform appearance. Relative errors in determining specific mechanical energy (Es), the initial temperatures (Ti), and at the end of pre-mixing (Tf) and texturing (Tt) steps were less than 5%. A low temperature change was observed during the pre-mixing step (TfTi) for all dough (1.2 ± 0.4 C) whereas larger increase was observed, from 2.7 to 6 C, during texturing step. Largest values of specific mechanical energy (23.6 and 29.3 kJ/kg) were obtained for dough without bran (#1 and #2, respectively), whereas Es varied between 14.9 and 21 kJ/kg (#17 and #7, respectively) for bran-enriched dough, without dry gluten addition (Table 1). The increase of dough temperature during texturing (TtTf) was correlated to Es (Fig. 1b; R2 ¼ 0.91); this result showed that the dough was heated by viscous dissipation. By mixing wheat flour dough under various time and speed conditions, Shehzad et al. (2012) represented this phenomenon by writing:
. Es ¼ h:g_ 2 :Dt rp ¼ Cp :ðTt Tf Þ þ “losses”
(6)
where h, rp, Dt, Cp, were respectively the viscosity, the density, the texturing time and the heat capacity of the dough, and, g_ the average shear rate in the mixer, whereas the term “losses” meant heat losses with environment. The results of ANOVA on Es values (Table 2) confirmed the trends observed in Fig. 1b: adding coarse or fine brans induced a significant decrease of Es; conversely, Es increased with the addition of vital gluten. The addition of brans or gluten significantly modified the dough behaviour during mixing, but the interpretation of these effects was not straightforward. These changes reflected changes of dough rheological behaviour,
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
171
completely immersed and could shear it efficiently. Conversely, for flours added with bran, without gluten supplementation (#9 and #17), the dough had a more heterogeneous appearance, with separated pieces which the mixing arm pushed by distinct blocks, with poor shearing (Fig. 1a). So, these observations suggested that average shear rate during mixing was not constant and it was lower in the latter case (#9 and #17) than in the former (#2 and #10), which, in line with Eq. (6), would explain why lower specific mechanical energies were measured. These effects reflected the influence of bran particles on the gluten network formation; they suggested that gluten network would be less organized when adding bran solid particles. 3.2. Elongational properties The biaxial extension measurements performed at different strain values (0.1 < ε_ b < 1.25) for the 18 formulas, showed that the variations of elongational viscosity (hE) as a function of strain rate ε_ b could be adjusted by a power law, with r2 > 0.97:
hE ¼ K : ε_ n1 b
ðεb ¼ constantÞ
(7)
For strain values εb ¼ 1, the values of K (consistency index) varied between 13 kPa.sn (#2) and 21 kPa.sn (#11) (Table 1) whereas the flow index values (n) varied in a narrower interval (0.36e0.47). These values were in accordance with those reported in the literature (Launay and Michon, 2008). The addition of bran increased significantly K, with a more significant effect of coarse bran (p < 0.001) compared to fine bran (p < 0.05) (Table 2). Indeed, the formation of gluten network could have been hindered because the bran particles reduced flour proteins aggregation. But the physical action of bran, as solid particles acting like a charge in a suspension, seemed to have predominated and increased dough elongational viscosity, as reported by Cavella et al. (2008), and also by BonnandDucasse et al. (2010) for shear viscosity. Moreover, the water binding capacity of the bran could have affected dough viscosity more significantly with coarse brans, which had higher water absorption capacity than fine brans. Finally, the addition of gluten to bran-enriched doughs showed no significant influence on elongational properties (p > 0.05) (Table 2). These results illustrated that the mechanisms underlying viscosity increase related to bran addition involved largely the presence of solid particles, like a charge in a suspension.
▫
Fig. 1. Evolutions of (a) spiral mixer power ( ) and dough temperature (B) during mixing for #2 ( , O), #3 ( , ), #17 (-, C) and their images at the end of the texturing time, and (b), temperature rise as a function of specific energy during texturing from doughs containing different levels of wheat bran (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation.
linked to the presence of wheat brans and gluten. Indeed, if the average shear rate remained constant with bran addition, Eq. (6) suggested that the addition of bran would lead to a smaller dough viscosity, compared to a dough without brans; this result was in contradiction with Bonnand-Ducasse et al. (2010) and Cavella et al. (2008) who reported that viscosity increased with bran addition. This apparent contradiction might be raised by the observation during mixing (Fig. 1a). Comparison of the images of dough at the end of mixing suggested that the agitation of the dough in the mixer was different, depending on its composition. For flours without bran (#2) or added with brans and supplemented in gluten (#10), dough appeared smooth and homogeneous; it remained a continuous mass in which the mixing arm was
3.3. Dough behaviour during proofing The kinetics of porosity and stability of dough during fermentation were determined at macroscopic scale by 2D image followup. The porosity curves had a sigmoid shape with an initial value (t ¼ 0 min) between 0 and 0.25 and a final value (t ¼ 240 min) of 0.7e0.8 (Fig. 2). These curves were well fitted by the Gompertz model (Eq. (3), R2 > 0.97), for all formulas. The relative standard deviation of porosity parameters (a, b, c, d) was less than 7%. The time for inflection point (parameter c) did not vary much, between about 60 min for three doughs with 5% bran added (#3, #7, #8) and
Table 2 Results of three-way ANOVA on specific mechanical energy (Es), consistency index (K), porosity increase (a), loss of stability (a0 c0 ), apparent modulus (Ebread), residual stress (sr) and bread density (r*). Factor Fine bran Coarse bran Gluten
F F F
Es
K
72.43 131.52 144.66þ þ þ
3.86 þ 12.75 þ 2.64ns
a0 c0
a þ þ
18.93 7.88 3.40ns
8.54þ þ 19.44þ þ 3.82ns
Ebread þ
25.73þ þ 7.34þ þ 0.09ns
þ
sr
r*
8.39þ þ þ 14.85þ þ þ 5.89
161.28þ þþ 81.19þ þ þ 52.11
F values and the associated significance of the effect (þor , þþ or and þþþ or indicate p < 0.05, p < 0.01 and p < 0.001, respectively; ns: no-significant) with þ and meaning a positive or negative effect, respectively).
172
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
an average value of 49 ± 2 min for other doughs (Table 1). For t/þ ∞, porosity values (¼ a þ d) were all close to 0.77 (±0.04). The porosity increase a, varied between 0.53 (#11) and 0.73 (#2). The values were close to those encountered by Romano et al. (2007) and Shehzad et al. (2010). They showed an overall decreasing trend with the level of added bran (Fig. 2a). This trend confirmed the one observed by Cavella et al. (2008) showing that higher bran addition (>7%) led to decreased expansion of dough during proofing. An ANOVA on parameter a confirmed that bran contents had a significant negative effect on dough expansion (Table 2). Fine bran caused a larger effect (p < 0.001) compared to coarse bran (p < 0.01), especially between 0 and 5% of bran added. Gluten addition had no significant effect (p > 0.05) on dough expansion. As shown in Fig. 2a, the porosity increase was negatively correlated with the consistency index (R2 ¼ 0.70). This result could be explained by the mechanisms of bubble growth in a viscous medium, applied in the case of dough proofing by Della Valle et al. (2014): a high elongational viscosity decreases the bubble growth, hence limiting dough expansion. During proofing, dough exhibited a continuous decrease of the shape ratio S(t), reflecting a loss of stability. For t 90 min, the curves S(t) were well fitted by an exponential decay (Eq. (3),
R2 > 0.9). The starting time of the stationary phase (b0 ) varied between 19.6 min (#2) to 155 min (#17), with values larger than 60 min for doughs (#11, 15, 16, 17 and 18), i.e. when bran addition 15% (Table 1). This result suggested that a large addition of bran favoured a continuous loss of stability of the dough during proofing. The overall loss of dough stability, evaluated by (a0 c0 ) varied between 0.47 (#17) to 0.09 (#8), with (a0 c0 ) 0.23 when bran addition 15%. This result showed that bran addition increased the loss of stability of the dough during proofing. Results of ANOVA on (a0 -c0 ) (Table 2) confirmed that the bran content had a strong effect (p < 0.001) on the loss of stability and indicated that the addition of coarse bran was more significant (p < 0.001) than addition of fine bran (p < 0.01); gluten addition did not improve significantly the dough stability (p > 0.05). Finally, the loss of stability was only fairly correlated to the consistency index K (R2 ¼ 0.59; not shown), in a positive way. However, elongational viscosity was expected to have a positive effect on the stability of the medium. So, our results rather suggested that the loss of dough stability would be due to the destabilization of the gas bubbles interface by bran particles during dough proofing. Indeed solid particles could pierce the films separating bubbles (Gan et al., 1995), and this effect would be more important with particles of larger size. 3.4. Bread texture 3.4.1. Density As expected, the addition of bran increased significantly density, from r* ¼ 0.16 (#2) to 0.36 g cm3, for 20% bran added (#17) (Tables 1 and 2). Gluten addition decreased significantly (p < 0.05) density of bread with the coarse and/or fine brans (Table 2). The effect of particles size was less dramatic than that of bran content. From the breads enriched with bran, only #4, #8 and #10 samples reached a density close to the one of control bread (r* ¼ 0.18 g cm3), which suggested that a low bran addition (5%), and intermediate (10%) with the two particle sizes, could be balanced by addition of vital gluten. Furthermore, there was a significant correlation between the expansion of dough, or porosity increase, during proofing and final bread density (Fig. 2b, R2 ¼ 0.71): the higher the porosity at the end of proofing, the lower the bread density. This result suggested that baking did not much influence the density of bread enriched with bran. In other words, wheat bran particles had more effect during mixing and proofing than during baking.
Fig. 2. Porosity increase (parameter a of the Gompertz model, Eq. (2)) determined as illustrated in the insert (experimental curves of porosity during fermentation for #2 ( ), #5 ( ) and #17 (:)) (a) as a function of dough consistency index K, and (b) as a function of bread density for different levels of wheat bran (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation.
▫
3.4.2. Crumb cellular structure Different cellular structures were obtained as shown by images in Fig. 3. For example, crumb #1 had a uniform grain and smooth texture. By contrast, sample #17, containing 20% bran, gave a denser and closer crumb with smaller bread volume. Other samples with brans at different content exhibited heterogeneous texture with variations in cell sizes as well as cell wall thickness. In order to ascertain these qualitative observations, image analysis was carried out. Principal component analysis (PCA) was applied in order to compare the resulting granulometric curves and to represent a similarity map of the breads (Fig. 3). The first two components accounted for 95% (PC1 ¼ 78%; PC2 ¼ 17%). Each sample represented the mean coordinates of thirty images. Principal component 1 mainly described differences between sample #1 and other breads, stressing that #1 had a finest crumb and smallest cells size, and thus confirming first observation. Principal component 2 completed the crumb description by providing an indicator of cellular heterogeneity. Indeed, crumbs at the bottom had more regular crumb, although coarse, with cells of similar size and thick cell walls (#17 for instance); conversely, crumbs with irregular structure, containing both large and small cells, with thinner cell
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
Fig. 3. Similarity map for crumb morphology of bran-enriched breads (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation, obtained by principal analysis component (PCA) and images of bread slice (#1, #2, #8, #17). Dotted arrow indicated the direction of bread density increase. Each point included the standard deviation of PC1 and PC2 (30 replications per formula).
walls (like #2 and #8, for instance), were rather located at the top. Finally, the distribution of crumb cellular structures on the similarity map owed much to the density value, as illustrated by the direction of increasing density, and increasing bran content (Fig. 3). Breads with larger bran content had larger density and, consequently, exhibited coarser, but more homogeneous, crumb cellular structures. 3.4.3. Mechanical properties Results from multi-indentation showed that all breads displayed similar texture profiles, with three different sequences (Fig. 4), as described by Chaunier et al. (2014) for breads without fibre: (1) a stress increase up to a value sm, attributed to crust debonding; (2) an enhanced increase of stress till s2/3, reflecting the densification of crumb before the compression was stopped; (3) a decrease in stress during relaxation of the crumb to reach a stable value, sr, ranging between 1.5.105 Pa (#4) and 2.2.105 Pa (#17) (Table 1). The
Fig. 4. Apparent modulus of breads as a function of breads density for various contents of bran added (0, 5, 10, 15, 20%), with (-) and without (C) vital gluten supplementation. The insert shows two experimental curves obtained during multi-indentation test and illustrates the determination of main mechanical properties.
173
apparent modulus of bread, related to crust stiffness, Ebread, varied from 6.9.105 Pa for the control bread, without bran (#1), and 38.2.105 Pa for bread with 20% bran added (#17) (Table 1). The relative standard deviation of these four variables was ±16% for all breads. These results underlined the influence of bran on the whole texture of breads, including both crust and crumb, and they were confirmed by an ANOVA on these mechanical variables (Ebread and sr) showing that bran content strongly influenced (p < 0.05) the texture of breads (Table 2), leading to stiffer crust and firmer crumb. The mechanical variables Ebread and sr were found to be correlated (R2 > 0.80). This result was in agreement with those from Chaunier et al. (2014) for bread of the same format. In our case, this result suggested that the texture contrast of breads, defined by the ratio Ebread/sr, was not modified by the addition of bran. The high correlation of Ebread with r* (R2 ¼ 0.83; Fig. 4) confirmed that the increase of mechanical properties, and consequently of crust stiffness and crumb firmness, was mainly due to an increase of density, induced by bran addition. This fitting was in agreement with the GibsoneAshby's model for cellular solids or solid foams (1997). It led to a value of the power law index (n ¼ 2.1) close to the one found by Poutanen et al. (2014) for crumbs of pan breads enriched with bran (n ¼ 2.3). This result confirmed that the density increase was the main factor involved in the texture changes of breads enriched with wheat bran. 4. About the mechanisms of dough densification by bran addition The last results of the present work (3.4) have confirmed that the texture changes of bread enriched with wheat bran were mainly governed by the density increase, and that this increase was already expected from a decrease of dough expansion during proofing (3.3). So we wanted to focus on the mechanisms of this decrease, and especially how it involved the rheological properties of dough. In this study, different particle size and level of wheat bran have been employed with or without vital gluten supplementation, and the three factors experimental design led to a large variation of rheological properties and bread texture. The influence of bran addition on the dough rheological properties has first been reflected by the behaviour of dough during mixing and the decrease of Es with bran addition; they were commonly attributed to a deficient formation of gluten network (Lai et al., 1989; Gomez et al., 2003), which was confirmed here by the significant effect of gluten addition. The detrimental effects of bran on gluten network reviewed by Noort et al. (2010) were: (1) particles steric hindrance, or physical barrier, (2) preferential water binding, (3) release of ferulic acid that cross-links arabinoxylans with gluten proteins. This third effect could appear in contradiction with findings of Wang et al. (2002, 2003) who reported that the gluten network became more resistant to extension by the action of ferulic acid. The first effect, already suggested by Courtin and Delcour (2002) and Katina et al. (2006), did not seem so important in our case, since bran particle size had no significant effect on Es (Table 2). The influence of the second effect has been reduced by adjusting the level of added water (Table 1), so any deficiency of gluten network formation could be attributed mostly to the third effect, i.e. bran chemical reactivity. However, negative effects on gluten formation would not explain the increase of dough elongational viscosity with bran addition, reported by Cavella et al. (2008), and also by BonnandDucasse et al. (2010) for shear viscosity, since a weaker network should lead to lower viscosity. Indeed, dough containing bran could be considered as a suspension of solid particles, bran acting as a filler in a viscoelastic medium, and its viscosity could be predicted by a suspension model, as suggested by Poutanen et al. (2014).
174
F. Le Bleis et al. / Journal of Cereal Science 65 (2015) 167e174
Nevertheless, according to such a model, the viscosity of a suspension containing 20% of solid particles should be increased by a factor 2, in comparison to the viscosity of the viscoelastic matrix, here the dough. In our case, elongational viscosity of dough with 20% bran, was increased by a factor 1.5, when comparing samples #1, 2 with #17, 18, respectively, which was less than expected by the suspension model. So, we might conclude that the changes of elongational properties of dough with wheat bran mainly resulted from a filler effect, due to solid particles addition, which were partly balanced by the negative effects, induced by a deficiency of gluten network formation. The increase of viscosity well explained the lower increase of porosity during fermentation (Fig. 2a), but did not explain the increased loss of stability, since more viscous dough should be more stable. Courtin and Delcour (2002) also attributed the default of dough development to gas cells piercing by bran particles. Indeed, for higher dough porosity (0.5), during fermentation, gas bubbles become connected to each other and would be separated by thin liquid films (Sroan and MacRitchie, 2009; Turbin-Orger et al., 2012) of lower thickness (10 mm) than bran particles. Hence, bran particles could destabilize gas bubbles interface in fermented dough favouring bubbles coalescence and causing dough collapse. This interpretation is consistent with observations obtained on the dough microstructure by X ray microtomography (Turbin-Orger et al., 2012), which showed that the gas cells contained in bran-enriched doughs had very irregular, polyhedral shapes, but clearly these observations would need quantification to confirm this interpretation. Finally, it could also explain why the loss of stability was more significant for bran of larger particle size. From a practical point of view, the proofing time could be reduced to limit the loss of dough stability. However, this action would lead to a decrease of porosity, and therefore a lower dough expansion, and, consequently a higher bread density. 5. Conclusion By adding up to 20% of bran with large and small particle sizes, we have covered a wide range of dough behaviour and rheological properties. Our results confirmed that the main effect of bran addition was to increase bread density, which induced an increase of the mechanical properties of bread, i.e. stiffer crust and firmer crumb. The changes of dough behaviour during fermentation with bran addition drew essential consequences on the formation of the cellular structure and the density of the bread. The decreased expansion and increased loss of stability during proofing could largely be explained by the changes of extensional properties of the dough at the end of mixing, and to a less extent by the direct effect of bran particles on the films separating gas bubbles. Finally, the negative effect of bran addition on gluten network formation could be partly balanced by its role as a filler like solid particles suspended in the dough. Acknowledgements This work was carried out within Pan&Sens program granted by the Region Pays de Loire among which all partners are gratefully guerre for her acknowledged. The authors are thankful to A.L. Re assistance in image analysis. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jcs.2015.06.014.
References Anderson, J.W., Baird, P., Davis, R.H., Ferreri, S., Knudtson, M., Koraym, A., Waters, V., Williams, C.L., 2009. Health benefits of dietary fiber. Nutr. Rev. 67 (4), 188e205. Bloksma, A.H., 1990. Dough structure, dough rheology, and baking quality. Cereal Foods World 35 (2), 237e244. Bonnand-Ducasse, M., Della Valle, G., Lefebvre, J., Saulnier, L., 2010. Effect of wheat dietary fibres on bread dough development and rheological properties. J. Cereal Sci. 52 (2), 200e206. Campbell, G.M., Ross, M., Motoi, L., 2008. Bran in bread: effects of particle size and level of wheat and oat bran on mixing, proving and baking. In: Campbell, G.M., Scanlon, M.G., Pyle, D.L. (Eds.), Bubbles in Food 2-Novelty, Health and Luxury. Eagan Press, St Paul, USA, pp. 337e354. Cavella, S., Romano, A., Giancone, T., Masi, P., 2008. The influence of dietary fibres on bubble development during bread making. In: Campbell, G.M., Scanlon, M.G., Pyle, D.L. (Eds.), Bubbles in Food 2-Novelty, Health and Luxury. Eagan Press, St Paul, USA, pp. 311e322. Chaunier, L., Chiron, H., Della Valle, G., Rouaud, O., Rzigue, A., Shehzad, A., 2014. Assessment of French bread texture by a multi-indentation test. J. Food Eng. 122, 92e98. Courtin, C.M., Delcour, J.A., 2002. Arabinoxylans and endoxylanases in wheat flour bread-making. J. Cereal Sci. 35, 225e243. Della Valle, G., Chiron, H., Cicerelli, L., Kansou, K., Katina, K., Ndiaye, A., Whitworth, M., Poutanen, K., 2014. Basic knowledge models for the design of bread texture. Trends Food Sci. Technol. 36, 5e14. Galliard, T., Gallagher, D.M., 1988. The effects of wheat bran particle size and storage period on bran flavour and baking quality of bran/flour blends. J. Cereal Sci. 8, 147e154. Gan, Z., Ellis, P.R., Schfield, J.D., 1995. Mini review: gas cell stabilisation and gas retention in wheat bread dough. J. Cereal Sci. 21 (3), 215e230. Gibson, L.J., Ashby, M.F., 1997. Cellular Solids, Structure and Properties. Cambridge Press University. Gomez, M., Ronda, F., Blanco, C.A., Caballero, P.A., Apesteguia, A., 2003. Effect of dietary fibre on dough rheology and bread quality. Eur. Food Res. Technol. 216, 51e56. Katina, K., Salmenkallio-Marttila, M., Partanen, R., Forssell, P., Autio, K., 2006. Effects of sourdough and enzymes on staling of high-fibre wheat bread. LWT e Food Sci. Technol. 39 (5), 479e491. Lai, C.S., Hoseney, R.C., Davis, A.B., 1989. Effects of wheat bran in breadmaking. Cereal Chem. 66, 217e219. guerre, A.L., 2007. GranLassoued, N., Babin, P., Della Valle, G., Devaux, M.F., Re ulometry of bread crumb grain: contributions of 2D and 3D image analysis at different scale. Food Res. Int. 40 (8), 1087e1097. Launay, B., Michon, C., 2008. Biaxial extension of wheat flour dough: lubricated squeezing flow and stress relaxation properties. J. Texture Stud. 39, 496e529. Noort, M.W.J., van Haaster, D., Hemery, Y., Schols, H.A., Hamer, R.J., 2010. The effect of particle size of wheat bran fractions on bread quality e evidence for fibreprotein interactions. J. Cereal Sci. 52 (1), 59e64. Pomeranz, Y., Shogren, M.D., Finney, K.F., Bechtel, D.B., 1977. Fiber in breadmaking e effects on functional properties. Cereal Chem. 54 (1), 25e41. Poutanen, K., Sozer, N., Della Valle, G., 2014. How can technology help to deliver more of grain in cereal foods for a healthy diet? J. Cereal Sci. 59 (3), 327e336. Romano, A., Toraldo, G., Cavella, S., Masi, P., 2007. Description of leavening of bread dough with mathematical modelling. J. Food Eng. 83, 142e148. Salmenkallio-Marttila, M., Katina, K., Autio, K., 2001. Effects of bran fermentation on quality and microstructure of high-fiber wheat bread. Cereal Chem. 78, 429e435. Shehzad, A., Chiron, H., Della Valle, G., Kansou, K., Lamrini, B., Lourdin, D., 2012. Energetical and rheological approaches of wheat flour during mixing with a spiral mixer. J. Food Eng. 110 (1), 60e70. guerre, A.L., 2010. Shehzad, A., Chiron, H., Della Valle, G., Kansou, K., Ndiaye, A., Re Porosity and stability of bread dough during proofing determined by video image analysis for different compositions and mixing conditions. Food Res. Int. 43 (8), 1995e2005. Shogren, M.D., Pomeranz, Y., Finney, K.F., 1980. Counteracting the deleterious effects of fiber in breadmaking. Cereal Chem. 58, 142e144. Sroan, B.S., MacRitchie, F., 2009. Mechanism of gas cell stabilization in breadmaking. II. The secondary liquid lamellae. J. Cereal Sci. 49, 41e46. guerre, A.L., Turbin-Orger, A., Boller, E., Chaunier, L., Chiron, H., Della Valle, G., Re 2012. Kinetics of bubble growth in wheat flour dough during proofing studied by computed X-ray micro-tomography. J. Cereal Sci. 56 (3), 676e683. van Vliet, T., 2008. Strain hardening as an indicator of bread-making performance: a review with discussion. J. Cereal Sci. 48 (1), 1e9. Wang, M., Hamer, R.J., van Vliet, T., Oudgenoeg, G., 2002. Interaction of water extractable pentosans with gluten protein: effect on dough properties and gluten quality. J. Cereal Sci. 36 (1), 25e37. Wang, M., Hamer, R.J., van Vliet, T., Gruppen, H., Marseille, H., Weegels, P.L., 2003. Effect of water unextractable solids on gluten formation and properties: mechanistic considerations. J. Cereal Sci. 37 (1), 55e64. Zhang, D., Moore, W.R., 1999. Wheat bran particle size effects on bread baking performance and quality. J. Sci. Food Agric. 79 (6), 805e809.