Rheological, microstructural and sensorial properties of durum wheat bread as affected by dough water content

Food Research International 51 (2013) 458–466

Contents lists available at SciVerse ScienceDirect

Food Research International journal homepage: www.elsevier.com/locate/foodres

Rheological, microstructural and sensorial properties of durum wheat bread as affected by dough water content Marcella Mastromatteo a, Mariangela Guida a, Alessandra Danza b, Janine Laverse a, Pierangelo Frisullo a, Vincenzo Lampignano a, Matteo Alessandro Del Nobile a,⁎ a b

Department of Food Science, University of Foggia, via Napoli, 25, 71100 Foggia, Italy Department of Agro-Environmental Sciences, Chemistry and Plant Protection, University of Foggia, via Napoli, 25, 71100 Foggia, Italy

a r t i c l e

i n f o

Article history: Received 12 November 2012 Accepted 5 January 2013 Keywords: Durum wheat bread Water content Textural characteristics Sensorial properties

a b s t r a c t In this paper the influence of water content on the rheological, microstructural and sensorial properties of durum wheat bread was evaluated. In order to evaluate bread quality, oscillation measurements, stress relaxation test and creep–recovery measurements were performed on dough samples, whereas tomographic and sensorial analyses were performed on baked bread samples. Results of the rheological analysis highlighted that both the storage and loss moduli (G′, G″) showed a descending trend with the increase of the water content. This is also confirmed by stress relaxation tests. Creep–recovery tests for strong doughs (with low water content), recorded greater resistance to deformation, therefore a smaller creep strain than the softer doughs. These results were reflected in the microstructural properties of the bread; an increase in water content caused an increase in the percentage volume of pores. Regarding the sensorial properties, the overall acceptability of the investigated bread samples was low for both the lowest and the highest water contents, and this was due primarily to the compact crumb with small bubbles and high crust firmness for the former and to the loaf volume collapsed with irregular distribution of very large bubbles for the latter. Therefore, the bread samples with intermediate water content were preferred by the panelists. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Bread is considered to be of global importance in nutrition being a source of proteins, dietary fibers, vitamins, micronutrients and antioxidants. Unlike other areas of the world, in southern Italy bread is commonly produced using durum wheat flour (Triticum durum). This type of bread represents a traditional product characterized by a higher crumb firmness, a lower loaf volume and a longer shelf-life compared to wheat bread (Boyacioglu & D'Appolonia, 1994). It is well known that wheat flour dough is a heterogeneous system in which starch granules are included in a gluten network. The wheat gluten proteins correspond to the major storage proteins that are deposited in the starchy endosperm cells of the developing grain. These form a continuous proteinaceous matrix in the cells of the mature dry grain and are brought together to form a continuous viscoelastic network when flour is mixed with water to form dough (Angioloni & Collar, 2009). In making dough, water is an essential ingredient; in fact, it is needed to form the gluten and give the dough consistency. In particular, the consistency depends clearly on the amount of water used in making it. The water added to the flour fulfills four functions: it dissolves soluble molecules, activates enzymes, brings about the formation of new bonds between the macromolecules in the flour, and ⁎ Corresponding author. Tel./fax: +39 0881 589 242. E-mail address: [email protected] (M.A. Del Nobile). 0963-9969/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodres.2013.01.004

alters the rheological properties of the dough. The large amount of water that is added to the flour has to be absorbed by the flour polymers. The majority of the water added to make up the dough is absorbed by hydrophilic groups on the protein molecules. If the water is insufficient for the hydration of all dough ingredients, the gluten does not become fully hydrated and the elastic nature of the dough does not become fully developed. Conversely, an excessive level of free water in the dough results in the domination of the viscous component of dough, with a decreased resistance to extension, increased extensibility and the development of sticky dough (Spies, 1997). The potential role of an aqueous liquid phase in doughs is to stabilize the surface active materials at the gas–liquid interface, to maintain the integrity of gas bubbles and to promote gas retention. Moreover, the amount of free water is also likely to determine the type and quantity of material that may become solubilized during mixing and dough development. Varying the amount of water can modify the microstructure of the dough. Water is considered to play the most important role in the viscoelastic properties of dough due to its influence on the development of the gluten protein network (Skendi, Papageorgiou, & Biliaderis, 2010). Crumb cellular structure (or its grain) is an important quality criterion used in commercial baking and research laboratories to judge bread quality alongside taste, crumb color and crumb physical texture. Umbach, Davis, Gordon, and Callaghan (1992) found a higher diffusion coefficient of water molecules in starch–water mixtures than in gluten–

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

water mixtures before heating. It was concluded that small amounts of water in dry starch were very tightly associated, but additional water did not interact with starch, thus remaining quite mobile. Once further hydrated, water in gluten led to more water–protein interaction, thus becoming more immobile. Experimental studies using the aqueous phase of dough to furnish the required water in breadmaking gave beneficial effects on loaf volume and crust color (Baker, Parker, & Mize, 1946). It was shown that a lower loaf volume was obtained when the dough was deprived of its aqueous phase, and this affected gas retention (MacRitchie, 1976). Letang, Piau, and Verdier (1999) have already shown that the microstructure is essential to compare the evolution of different doughs based on soft wheat flour. It can be seen as the linkage between the ingredients and the apparent macroscopic properties of the final product. However, the majority of the studies reported in literature based on the water–dough relation referred to soft wheat flour mixed with water. There are no studies dealing with the effect of dough water content on durum wheat flour bread. In addition, results from the literature are qualitatively interesting, but each flour is different and general quantitative interpretations are difficult. This work proposes to study the influence of water content on the process parameters of durum wheat flour doughs and bread by relating the microstructure to the rheological and sensorial properties. This approach allows giving an interpretation of the evolution of dough properties in terms of microstructural changes. The aim of this work was to evaluate the influence of dough water content on the dough's rheological, microstructural and sensorial properties of durum wheat bread. Oscillation measurements, stress relaxation test and creep–recovery analysis were performed in order to evaluate the rheological behavior of the different doughs. Moreover, tomographic analysis was carried out to evaluate the texture properties of the manufactured bread. Finally, the sensorial quality of the final product was also evaluated and correlated with some of the above properties. 2. Materials and methods 2.1. Raw materials Durum wheat flour was supplied from Tandoi mill (Corato, Bari, Italy), fresh compressed yeast and salt were purchased from a local market, and dried sourdough was supplied from Bongiovanni mill (Villanova Mondovi', Cuneo, Italy). The amount of tap water used to make bread varied according to Table 1. 2.2. Breadmaking process A basic bread formula consisting of durum wheat flour (4500 g), fresh compressed yeast (100 g), salt (100 g), dried sourdough (100 g) and tap water (Table 1) was used. Dough mixing, processing and baking were performed on laboratory-scale equipment. A straight dough process was used. All dry ingredients were mixed thoroughly with half of water in a mixer (Bernardi Impastatrici, Cuneo, Italy) at high speed

459

(120 rpm) for 10 min, and then the rest of the water was added slowly and mixed for 15 min (4 rpm). After complete mixing, the dough rested in bulk for a period of 15 min, then portions of 1500 g were made, manually rounded and put into canvas for the final fermentation. The portions are put in the incubator (Thermogel, Varese, Italy) for 60 min, at 26 °C of temperature and 65% of relative humidity. Following fermentation, the samples were baked at 270 °C for 55 min in an electric oven (Europa Forni, Vicenza, Italy). 2.3. Dough rheological properties 2.3.1. Oscillation measurements Dough samples for the rheological tests were prepared without adding any yeast to the formulation to avoid bubble interference. The rheological measurements were conducted using a controlledstrain rotational rheometer (ARES model, TA Instruments, New Castle, DE) equipped with a force rebalance transducer (model 1K-FRTN1, 1–1000 g cm, 200 rad/s, 2–2000 gmf) and parallel plates (superior plate diameter of 50 mm). A steady temperature was ensured with an accuracy of ±0.1 °C by means of a controlled fluid bath unit and an external thermostatic bath. Before starting the measurement, a sample taken from the inside of the dough, was rested between the plates for 5 min, so that the residual stresses would relax (Amemiya & Menjivar, 1992; Letang et al., 1999). Each type of dough was placed onto the surface of the lower plate and the upper plate was lowered until it reached a 2 mm gap distance as to avoid sample disruption and the excess sample was trimmed. Slippage was prevented by using sandpaper. To prevent water evaporation, a suitable cover tool sealing the top of the superior plate was used during testing. In fact, the dehydration leads to crusting, which affects the results significantly (Szczesniak, Loh, & Wesley, 1983). Storage modulus (G′), loss modulus (G″) and loss tangent (tanδ =G″/ G′) were determined in a frequency range of 0.05 to 10 Hz in the linear viscoelastic region. As an example, the storage and loss moduli, G′ and G″, as functions of the frequency taken in the linear viscoelastic region are shown in Fig. 1A. The strain value was obtained by preliminary strain sweep oscillatory trials to determine the linear viscoelastic region. The strain sweep oscillatory tests were carried out at a frequency of 1 Hz and in a range of shear strain of 0.01 to 300%. The linear domain was found to be very small, the linear strain limit being around 0.02–0.065%. All experiments were carried out at 25 °C. Three repetitions of the dynamic mechanical experiments were performed for each dough sample. To compare the G′ and G″ values between the investigated dough samples an oscillatory frequency of 10 Hz was chosen as a reference (Dimitreli & Thomareis, 2008). 2.3.2. Stress relaxation tests Mechanical transient tests were performed to evaluate the spectrum of the relaxation times from relaxation curves. Fig. 1B highlights an example of stress plotted as a function of decay time in a stress

Table 1 Levels of the experimental design used for bread-making process, and parameters of the stress relaxation and creep analysis for dough samples.

Br-w54 Br-w57 Br-w61 Br-w64 Br-w67 Br-w71 Br-w74

Water content (%)

tanδ

1/σ0 (1/MPa)

m

λ (s)

K (1/MPa)

Maximum creep strain (%)

Maximum recovery strain⁎ (%)

Relative recovery strain⁎⁎ (%)

54.44 57.77 61.11 64.44 67.77 71.11 74.44

0.31a ± 0.005 0.48b ± 0.05 0.43b ± 0.01 0.46b,c ± 0.05 0.46b,c ± 0.01 0.46b,c ± 0.02 0.5c ± 0.02

17.65a 9.48b 12.93c 4.18d 5.07d 2.43e 1.23f

1.01a 1.01a 1.01a 1.02a 0.13b 0.67c 0.81c

1.16e-2a 1.2e-2a 1.18e-2a 2.00e-2b 9.94e-3a 3.13e-2c 3.76e-2c

0.85a 0.85a 0.85a 0.73b 0.42c 0.30d 0.15e

7.83a ± 0.27 9.71a ± 2.9 9.74a ± 1.9 44.88a ± 20.1 106.86b ± 3.15 214.46c ± 26.5 679.92d ± 80.8

3.49 3.22 3.35 26.00 40.43 68.78 109.88

224.35 301.55 290.74 172.61 264.31 311.81 618.78

a–f Mean in the same column followed by different superscript letters differs significantly (p b 0.05). ⁎Calculated as: [maximum creep strain − steady recovery]. ⁎⁎Calculated as: [maximum creep strain/ maximum recovery strain ∗ 100].

460

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

tests were prepared as those used in breadmaking but without added yeast. The flour–water dough samples were prepared with five different water contents (Table 1). The sample preparation before testing was performed as in the previous tests. After loading, the dough sample was left to rest for 5 min (Amemiya & Menjivar, 1992). Trimming of excess dough was performed just before the measurement to avoid moisture loss during the resting period. In a creep–recovery experiment, a shear stress is imposed on a dough sample and the sample's deformation or strain is recorded in function of the creep time. For a creep–recovery experiment, three variables have to be chosen, more precisely the length of the creep phase, the shear stress applied during creep and the length of the recovery phase (Van Bockstaele, De Leyn, Eeckhout, & Dewettinck, 2011). In particular, creep tests by applying a constant stress of 50 Pa for 60 s on the sample and allowing strain recovery by the sample for 100 s after removal of load were performed. Each test was performed on different dough samples, at least in triplicates. 2.4. Tomographic analysis

Fig. 1. Rheological tests: A) G′ and G″ values as a function of the frequency; B) stress plotted as a function of decay time in a stress relaxation test.

relaxation test for the investigated samples. The relaxation data were obtained using a controlled-strain rotational rheometer (ARES model, TA Instruments, New Castle, DE) by setting the following experimental conditions: imposed strain 10%, displace time of 2 min. The sample preparation before testing was performed as in the previous test, in particular without adding any yeast to the formulation to avoid bubble interference. The relaxation behavior of the investigated samples was described by using a modified version of the expression proposed by Xianzhong and Jinping (2011): 1 σ m−1

¼

  t þK ⋅ 1 þ ð m−1 Þ⋅ λ σ m−1 0 1

ð1Þ

where σ and σ0 are the decaying stress and initial stress expressed in MPa, t is the decaying time expressed in s, λ is the characteristic relaxation time expressed in s, m is the steady-state creep rate exponent, K is a constant value regarded as a fitting parameter. The relaxation spectrum is a fundamental quality in the linear theory of viscoelastic materials that does not depend on the experimental conditions but only on the physical nature of the specimen. At least three replicates were performed for each different dough sample. 2.3.3. Creep–recovery measurements Creep–recovery measurements of flour–water doughs were made on a controlled-strain rotational rheometer (ARES model, TA Instruments, New Castle, DE), by using a parallel plate geometry (50 mm diameter and 2 mm gap). The test was performed in a controlled temperature environment (25 °C) by means of a controlled fluid bath unit and an external thermostatic bath with an accuracy of ±0.1 °C. To prevent water evaporation, a suitable cover tool sealing the top of the superior plate was used during testing. Dough samples for the rheological

For X-ray microtomographical analysis (μCT) the bread and dough samples were imaged under the same conditions, using the SkyScan 1172 high-resolution desktop X-ray microtomography system (SkyScan, Belgium). The dough samples were analyzed after 105 min of leavening. Before analyzing, the dough samples were put in cooling cells for 20 min to stop the dough from continuing to leaven during scanning. The dough and bread samples for the tomographic analysis were prepared as those used in breadmaking (they contained the amount of water specified in Table 1), and were placed on a round plate; the source and the detector were fixed, while the sample was rotated during measurement. Power settings of 100 kVp and 100 μA were used. A CCD camera with 2000 × 1048 pixels was used to record the transmission of the conical X-ray beam through all samples. The distance source–object–camera was adjusted to produce images with a pixel size of 2 μm. Four-frame averaging, a rotation step of 0.40° and an exposure time of 1767 ms were chosen to minimize the noise, covering a view of 180°. Smoothing and beam-hardening correction steps were applied to suppress noise and beam hardening artifacts, respectively. Beam hardening correction was only moderately applied, set to 25% within NRecon and a fast ring artifact reduction (set to 51 within NRecon) was also applied. Once initial parameters were set, the acquisition step was completely automated and did not require operator assistance. Scan time, on average, required 30 min. A set of flat cross section images, was obtained for each sample after tomographical reconstruction by the reconstruction software NRecon (SkyScan). Three-dimensional reconstructions of the samples were created by effectively stacking all 2D tomographs, a total of 125 slice images with a slice spacing of 0.069 mm. For image processing and analysis, the SkyScan software, CTAnalyser (CTAn) was used. Image segmentation was firstly carried out on the smoothed 8-bit gray-scale images obtained from the reconstruction step, using CTAn (SkyScan) software. Segmentation is the process of converting the gray-scale images into black and white images by assigning the value 1 to all pixels whose intensity was below a given gray tone value and 0 to all the others. For this, an automatic threshold based on the entropy of the histogram was calculated for each image. The lower gray threshold (8) and upper gray threshold (110) values were identified; each sample was processed under the same conditions. For data analysis, prior to 3D reconstruction, a componentlabeling algorithm, available within CTAn, was used to isolate the largest 3D connected structures. All reconstructions were created in CTAn (SkyScan) using an adaptive rendering (locality 10 and tolerance 0.25) algorithm and saved as P3G surface model (SkyScan model format). P3G models were then imported into CT vol software (SkyScan) for visualization. The following four geometric parameters were measured using the CTAn software (SkyScan): percent object volume (POV), object

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

461

surface/volume ratio (OSVR), fragmentation index (FI), and structure modeling index (SMI), where, i) percentage object volume is the proportion of the VOI (volume of interest) occupied by void areas, i.e. holes; ii) object surface/volume ratio is the basic parameter in characterizing the complexity of the structures and is also the basis of model-dependent estimates of thickness i.e. size and distribution of holes present in each sample; iii) fragmentation index is an index of connectivity of structure, it calculates the index of relative convexity or concavity of the total surface of the sample and finally, iv) SMI estimates the characteristic form of which the structure is composed, i.e. whether it is more plate-like, rod-like or even sphere-like (0 = ideal plate, 3 = cylinder and 4 = sphere). 2.5. Sensory analysis Bread samples were submitted to a panel of 10 trained tasters in order to evaluate the sensorial attributes. The panelists were selected on the basis of their sensory skills (ability to accurately determine and communicate the sensory attributes such as appearance, odor, flavor and texture of a product). Prior to testing, the panelists were however trained in sensory vocabulary and identification of particular attributes, by evaluating commercial bread. The panelists were asked to indicate color, appearance, crust and crumb firmness, large bubbles and overall acceptability of the bread. With this aim, a nine-point scale, where 1 corresponded to extremely unpleasant, 9 to extremely pleasant and 5 to satisfactory was used to quantify each attribute (Chillo, Laverse, Falcone, & Del Nobile, 2007). 2.6. Statistical analysis The rheological, microstructural and sensorial properties of bread with different contents of water were evaluated in this study. The water content of the bread was reported in Table 1. The lowest and highest levels of independent variable studied were chosen from results of preliminary laboratory tests. The results of the rheological, microstructural and sensorial analyses were compared by a one-way variance analysis (ANOVA). A Duncan's multiple range test, with the option of homogeneous groups (pb 0.05), was carried out to determine significant differences between bread samples. STATISTICA 7.1 for Windows (StatSoft, Inc., Tulsa, OK, USA) was used for this aim. Moreover, the interactions among the rheological properties of the dough samples and the microstructural and sensorial parameters of manufactured bread were evaluated using a correlation matrix. 3. Results and discussion 3.1. Dough rheological properties 3.1.1. Oscillation measurements The viscoelastic behavior of durum wheat dough samples with different water contents was investigated by means of oscillation frequency sweep experiments conducted in the linear viscoelastic range. The storage modulus (G′) gives an indication about the elastic nature of the sample and the loss modulus (G″) is related to its viscous behavior. The loss tangent tanδ is often used to measure the ratio of viscous and elastic responses of the material being tested (Letang et al., 1999). As can be seen in Fig. 1A, for all samples the storage modulus G′ was higher than the loss modulus G″ through all the frequency range; this result showed that the investigated doughs were more elastic than viscous. Furthermore, it can be observed from Fig. 2A that both the parameters under observation showed a decreasing trend with the water content, indicating that less force was needed to deform doughs containing high amount of water (Hibberd, 1970a, 1970b). In fact, the sample with the lowest amount of water (Br-w54) showed the highest value of both G′ and G″ moduli with respect to the other samples. In contrast, the Br-w74 sample, which has the highest quantity of water, recorded

Fig. 2. Rheological tests: A) Illustration of the G′ and G″ values at different concentrations of water for the dough samples; B) creep–recovery curves for some dough samples.

the lowest value of both parameters. All the other investigated samples showed a behavior that is bracketed between that of the sample Br-w54 and that of the sample Br-w74 according to dough water content. Letang et al. (1999) also found that an increase of water content had softening effects. In literature, the influence of water on the dynamic viscoelastic behavior of flour doughs has been interpreted in two different ways. It has been advanced that water above a given limiting value does not interfere with dough structure but acts as simple inert filler. Therefore, varying moisture content of the dough proportionally changes dynamic properties, but water can behave as a lubricant enhancing the relaxation phenomena. On the other hand, it has been suggested that water molecules act as a plasticizer lowering the Tg of the doughs. In this case, increasing the moisture content of the doughs changes dynamic response because the relaxation dynamics accelerates in a manner similar to that of amorphous synthetic polymers when the temperature rises (Masi, Cavella, & Sepe, 1998). Some authors (Edwards, Dexter, Scanlon, & Cenkowski, 1999) observed that the difference in the levels of dough softening obtained with different water amount used in the breadmaking process was a phenomenon well documented for wheat flour doughs that was simply attributed to constituent concentration effects. The same considerations can be extended to durum wheat flour doughs. Regarding the tanδ values, Table 1 shows that this parameter is independent of the water content; in fact, the values obtained for all doughs showed a negligible variation (Hibberd, 1970a, 1970b). Moreover, as reported by Letang et al. (1999), the loss tangent tanδ can be taken as an indicator of the structure's organization (molecular interactions) in the material: highly structured materials generally give low tanδ. Letang et al. (1999) found that long mixing times (more than 150 s) act like long rest times giving an increase of tanδ (up to 0.4–0.5)for the dough sample based on soft wheat flour and 50% of water. The authors

462

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

stated that these results are quite characteristic of a weakening of the material structure. Moreover, they found that a mixing time between 120 and 150 s corresponded to optimally mixed doughs (tanδ of about 0.35). Experiences have been conducted in order to see the influence of the water content (Hibberd, 1970a, 1970b; Navickis, Anderson, Bagley, & Jasberg, 1982). By studying gluten/starch blends (synthetic doughs), it has been possible to observe the moduli dependence against the protein/starch ratio (Hibberd, 1970b; Petrofsky & Hoseney, 1995; Smith, Smith, & Tschoegl, 1970). However, studies about the influence of mixing time (Bohlin & Carlson, 1980; Dreese, Faubion, & Hoseney, 1988) do not provide general conclusions, probably because the effect of overmixing depends strongly on the type of flour used.

3.1.2. Stress relaxation tests The curves showing the experimental and theoretical results of stress relaxation for dough samples with different water contents were obtained by fitting Eq. (1) to the experimental data (Fig. 1B). The theoretical model satisfactorily fitted the experimental results. As expected there is first a sharp decrease in the stress after which the stress levels off to constant value. This behavior was in accordance with that found by Li, Dobraszczyk, and Schofield (2003) for wheat flour doughs. In Table 1, the values of steady-state creep exponent, m, the characteristic relaxation time, λ, the fitting parameter, K, and 1/σ0 evaluated by means of Eq. (1) are listed. The first term on the right of Eq. (1) illustrates the stress level of relaxation at the beginning and the second shows the balance relation between the elastic behavior and the viscous behavior of the material at a certain relaxation time. From the data in Table 1, it can be inferred that the steady state creep exponent, m, and the characteristic relaxation time, λ, of the investigated doughs increased with decreasing water content (Maache-Rezzoug, Bouvier, Allaf, & Patras, 1998). In particular, the m value equal to one for the samples Br-w54, Br-w57, Br-w61, Br-w64 indicate a Newtonian behavior, whereas the m value less than one indicate a pseudoplastic behavior for the samples Br-w67, Br-w71, Br-w74 with high amount of water; in fact, free water molecules become to interact with the matrix system leading to a departure from Newtonian behavior. During mixing, the flour is progressively hydrated and the amount of free water is decreased so that the cohesion of the system increases and a macromolecular gluten network progressively develops. In fact, it is only for specific flour/water ratios (intermediate moisture levels) that a cohesive, macroscopically smooth and homogeneous mass can be created. On the other hand, in highly hydrated doughs the starch granules are surrounded not only by gluten but also by an aqueous phase due to excess free water (Blanshard, Frazier, & Galliard, 1986). Therefore, the particles of starch appear little embedded in the protein matrix compared to the samples obtained with the optimal water content. Moreover, the increasing water content can be considered as an addition of plasticizer to the polymeric matrix under investigation. Water is indeed the most common plasticizer for hydrophilic food components and this influences its mechanical properties. In the specific case, the samples with low and intermediate water content (Br-w54, Br-w57, Br-w61, Br-w67) showed shorter relaxation times so they appeared as elastic doughs; on the other hand, the samples with high water content (Br-w71, Br-w74) had longer relaxation times; therefore, they were less elastic compared to the previous samples. This last finding was due to the excessive hydration of the doughs in which, as said above, the particles of starch appear little embedded in the protein matrix but surrounded by the aqueous phase; this caused a major relaxation of the macromolecules chains (i.e. gluten proteins) that leads to a less-structured material thus less elastic. In their work Del Nobile, Chillo, Mentana, and Baiano (2007) recorded similar results; in particular, they found that in spongy foods (that is, white pan bread and “mozzarella” cheese) the relaxation time

is shorter with respect the bulky ones (that is, agar, meat and ripened cheese). Masi et al. (1998), in their work on wheat flour doughs, suggested that water molecules enhance relaxation behavior of mobile units, which evolve in less time than the experimental period, and the dough behaved as a liquid flow material. By lowering the moisture content, the relaxation dynamics slows down and the percentage of mobile units increases. They become cooperatively immobilized during the experimental period. In other words, when a dough is diluted, the local friction of its structural components drops, reducing all relaxation times. This situation is very similar to what occurs when the temperature of an amorphous synthetic polymer is increased (Masi et al., 1998). It is worth noting that the doughs prepared by Masi et al. (1998) had water content between 43 and 58%; thus the high levels investigated in this study (up to 74%) were not taken into account. Regarding the 1/σ0, an increase in the value of this parameter means more or less elastic samples; in fact, this value increased for samples with low and intermediate water contents. On the contrary, for the samples with excessive water content this parameter decreased which means less elastic samples. This latter finding could be due to the change in the microstructure of the doughs with high water level as discussed in the next sections. In fact, as demonstrated by Masi et al. (1998) not only the different moisture contents vary the viscoelastic response of doughs, but also the water molecules interfere with the dynamics by which relaxation phenomena take place. 3.1.3. Creep–recovery measurements Creep–recovery curves of dough samples with different water contents (samples Br-w54, Br-w64, Br-w74) are shown in Fig. 2B. The above figure shows that during the creep stage the strain increased with time in response to the applied constant force so that the dough was rapidly deformed. After a sufficient time is elapsed, the strain approached a steady state by reaching an equilibrium deformation. In the recovery stage the constant force is removed, the dough strain partially recovered with time from equilibrium deformation to a constant value. Moreover, the creep–recovery curves of doughs exhibited a typical viscoelastic behavior combining both viscous fluid and elastic components (Steffe, 1992). Table 1 shows significant differences in maximum creep strain and maximum recovery strain recorded for all dough samples. The maximum creep strain (strain at the end of creep phase) was used to describe dough rigidity (Wang & Sun, 2002). As can be inferred from Table 1, the stronger doughs, that is the samples with the lower water content and with greater resistance to deformation (samples Br-w54, Br-w57, Br-w61), had smaller creep strain than the softer doughs with higher water content (samples Br-w64, Br-w67, Br-w71, Br-w74). Lazaridou, Duta, Papageorgiou, BeIc, and Biliaderis (2007) reported that the maximum creep strain increased with rising water content in the gluten-free doughs. Moreover, this is in agreement with the findings of Edwards et al. (1999) and Van Bockstaele, De Leyn, Eeckhout, and Dewettinck (2008), who reported an increase in maximum strain with increasing water absorption of durum wheat doughs for pasta. The maximum recovery strain can be determined by subtracting the unrecoverable strain caused by viscous components from the total deformation (the maximum creep strain). The recovery strain is a useful tool for determining dough springiness or resilience and can be used to describe the elastic property of dough (Wang & Sun, 2002). Creep–recovery experiments may give some insight into the microstructure of dough. For example, a small recovery may indicate the presence of small structures in dough, whereas a large recovery may indicate the presence of larger structures (Weegels et al., 1995). The recovery also is an important factor for dough film stability. The higher the recovery strain, the better the stability against rupture of dough films between gas cells. The elasticity or springiness should be sufficiently high to prevent the ascent and spreading of

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

463

gas cells under the influence of gravity (Bloksma & Bushuk, 1988). Table 1 shows that the samples Br-w54, Br-w57, Br-w61 with lower water content had little elasticity and, consequently, had little substantial recovery when force was removed. A much greater recovery was found with the increase of the water content (samples Br-w64, Br-w67, Br-w71, Br-w74). For a better explanation of the results, the relative recovery strain was also evaluated (Table 1). Regarding this latter parameter, only the Br-w74 sample recorded the highest value of relative recovery strain (about 600%) by showing the highest recovery of this sample when the water content increased. These results are in agreement with what is found for the oscillation measurements; in fact, the samples with lowest amount of water showed the highest values of both elastic and loss moduli with respect to the samples with highest water content, showing a firmer structure. Therefore, the rheological properties of flour–water doughs determined by creep–recovery test can give an indication on the bread volume potential of the flour and on how the dough will behave during processing (Van Bockstaele et al., 2011).

triangulated surface of the structure under examination. The SMI assumes integer values of 0, 3 and 4 for ideal plates, cylinder and spheres, respectively (Hildebrand & Ruegsegger, 1997). It can be noted from the results in Table 2 that for the dough samples, the SMI values range from 0.5 to 2.3 therefore indicating that the characteristic shape of the pores present in the dough is anything in between that of a plate and cylinder. Whereas for the bread samples, the SMI values calculated for all the samples range from almost 2 to 3, these values are fairly close to 3, therefore suggesting that the average characteristic shape of the pores present in all the bread is more or less cylindrical in shape and does not change with change in water content. It is worth noting that the commercial bread sample named as Br-Altamura has microstructural parameters near to that of the samples with intermediate water content that is the Br-w64, Br-w67, Br-w71 samples.

3.2. Microstructure analysis

Table 3 shows the sensorial properties of durum wheat bread with different water contents. Results highlighted that the maximum score of overall quality (up to 8.1) was obtained by the Br-w64 and Br-w61 samples. Samples with the lowest water content (Br-w54 and Br-w57 samples) showed an overall quality score very positive (above 6) but significantly different from the Br-w64 sample. This result was due principally to the compact crumb with presence of smaller bubbles and to a higher crust firmness related to the lower amount of water added. Regarding the investigated samples with the highest water content, only Br-w67 sample showed acceptable characteristics, whereas Br-w71 and Br-w74 samples showed the lowest values of overall quality. In particular, with the increase of the water added to the dough the only parameters that remained acceptable were color and appearance. In fact, the excessive water content in the dough led to a collapse of the loaf volume that appeared flat with irregular distribution of very large bubbles. It is well known that the potential role of an aqueous liquid phase in the dough was to maintain the integrity of gas bubbles and promote gas retention (Sahi, 2003), so a more or less addition of water causes an irregular crumb structure and modification on crust firmness. This is in agreement with what reported by Sahi (2003), which found that an increase of water content induces the reduction in the surface tension and negatively influenced the crumb of the baked wheat flour bread. The same results can be applied to durum wheat flour doughs. In fact, there are no studies in literature dealing with the effect of water content on the durum wheat flour dough.

Fig. 3 shows examples of the set of flat cross sections that were obtained for each sample after binarization of the images using the reconstruction software NRecon (SkyScan); the void spaces in the images, represented by the white areas, are clearly visible. From these images the three-dimensional reconstructions were obtained from which the geometrical parameters were calculated using the CTAn software (SkyScan). For the microstructural analysis the pore size structure of each sample was studied. Table 2 shows the average values obtained for the four tomographical parameters using the CTAn software (SkyScan), POV, OSVR, FI and SMI and the results of the statistical analysis carried out as reported above for the dough and baked bread samples. It can be observed from this table that percentage object volume, i.e. the geometric parameter POV, was calculated for each image as a representation of the percentage total pore content within the sample. It can be noted that for the dough samples, the sample Br-w54 has the lowest POV value and that initially POV increases with increase in water content for the first three samples after which the POV value stabilizes with minimum fluctuations in the values. Whereas for the bread samples, the statistical analysis performed on the results of this table confirmed that samples Br-w71 and Br-w74 have the highest POV, whereas samples Br-w54 and Br-w57 have the lowest POV values. It can also be noted from these results that an increase in water content causes also an increase in the percentage volume of pores for all the samples. OSVR indicates the ratio of the surface of the cell walls to the total volume of the object (i.e. dough or bread). It can be observed from the table that the OSVR value for all samples (bread and dough samples) is very low, therefore indicating that the pores present in all the samples are largely distributed, i.e. indicating high porosity. With regard to FI parameter, i.e. the index of connectivity and therefore a measure of relative convexity or concavity of the total pore surface, based on the principle that concavity indicates connectivity, and convexity indicates isolated disconnected structures (Lim & Barigou, 2004). A lower FI signifies better-connected solid lattices and has a negative index, on the other hand a higher FI indicates a more disconnected solid structure and has a positive index. As it can be noted from the table, the FI is positive for all samples except for samples Br-w71 and Br-w74 that contain the highest water content therefore suggesting that the water content after a certain level does have an effect on the pore structure, i.e. increasing the water content will start to produce pores with less disconnected solid lattices and therefore convex in structure. As stated above the SMI parameter, that is a topological index, gives an estimate of the characteristic shape of a structure in terms of plates and cylinders composing the 3D structure (Hildebrand & Ruegsegger, 1997) and is calculated using a differential analysis of

3.3. Sensory analysis of durum wheat bread with different water contents

3.4. Correlations In order to evaluate the correlations among the bread's microstructural, rheological and sensorial parameters a principal component analysis was performed. The results obtained are shown in Table 4. Significant correlations among the investigated parameters were observed. In fact, the decrease of the water content led to stronger dough and then to the increase of the resistance to deformation. This means that both the moduli G′ and G″ increased with the decrease of the water content and this also corresponded to a decrease of both the maximum creep and recovery strain. Moreover, Table 4 shows that the increase of the maximum recovery strain, due to the increase of the dough water content, led to a decrease of the sensorial parameters such as crumb firmness, large bubbles and overall quality. In fact, the bread shown a low loaf volume with a crumb characterized by irregular distribution of very large bubbles that negatively affected the overall quality of the bread. The rheological parameters were also strongly correlated with the microstructural parameters. In fact, the increase of the moduli G′ and

464

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

Baked Bread ‘Br-w54’

Baked Bread ‘Br-w57’

Baked Bread ‘Br-w61’

Baked Bread ‘Br-w64’

Baked Bread ‘Br-w67’

Baked Bread ‘Br-w71’

Baked Bread ‘Br-w74’ Fig. 3. Flat cross sections of bread samples where the white areas are the void spaces.

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

465

Table 2 Values of the geometric parameters for the bread and dough samples. POV

OSVR

SMI

FI

Bread samples Br-w54 Br-w57 Br-w61 Br-w64 Br-w67 Br-w71 Br-w74 Br-Altamura

39.51a ± 2.2 44.68a,b ± 4.8 48.81b,c ± 7.6 52.67b,c,d ± 1.9 56.26c,d,e ± 5.1 62.05e ± 6.2 70.48f ± 3.7 58.88d,e ± 2.8

0.0102a ± 0.0007 0.0092a ± 0.0007 0.0076b ± 0.001 0.0068b,c ± 0.0008 0.0056d ± 0.0006 0.006c ± 0.0007 0.0042d ± 0.0003 0.0065c,b ± 0.0005

2.85a,b ± 0.3 2.63a,b ± 0.4 2.95a,b ± 0.6 2.56a,b ± 0.3 3.28b ± 0.6 1.98a ± 0.7 1.91a ± 0.9 1.97a ± 0.4

0.0020a ± 0.0002 0.0009a,b ± 0.0011 0.0012a,b ± 0.0016 0.0007a,b ± 0.0009 0.0011a,b,c ± 0.0008 −0.0028c ± 0.0035 −0.0021a,b,c ± 0.0014 −0.0028c,b ± 0.0008

Dough samples Br-w54 Br-w57 Br-w61 Br-w64 Br-w67 Br-w71 Br-w74

12.61c ± 5.5 26.41b ± 7.5 26.41b ± 8.5 43.39a ± 8.7 32.17a,b ± 3.6 42.88a ± 5.3 41.14a ± 3.0

2.447b ± 0.2058 1.5469a ± 0.3632 1.6411a ± 0.2722 1.3271a ± 0.2407 1.5734a ± 0.2376 1.2195a ± 1246 1.2178a ± 0.1534

2.28c ± 0.3 2.17c ± 0.1 1.21a ± 0.3 0.57b ± 0.5 1.26a ± 0.1 0.86a,b ± 0.4 0.72a,b ± 0.1

0.4271c ± 0.3387 0.0595b ± 0.1526 −0.3757a ± 0.2216 −0.6414a ± 0.2338 −0.3368a ± 0.0860 −0.5047a ± 0.1512 −0.4899a ± 0.0418

a–e

Mean in the same column followed by different superscript letters differs significantly (p b 0.05).

Table 3 Sensory characteristics of durum wheat bread with different water contents.

Br-w54 Br-w57 Br-w61 Br-w64 Br-w67 Br-w71 Br-w74

Color

Appearance

Crust firmness

Crumb firmness

Large bubbles

Overall quality

6.50a ± 0.47 7.60c ± 0.39 7.61c ± 0.55 8.05c ± 0.37 7.30b,c ± 0.42 6.65a,b ± 0.41 6.25a ± 0.35

6.35b ± 0.41 6.70b,c ± 0.35 7.25c,d ± 0.35 7.90d ± 0.39 6.60b,c ± 0.39 6.45b ± 0.37 5.61a ± 0.42

6.40a ± 0.39 6.55a,b ± 0.37 7.55c ± 0.44 7.95c ± 0.37 7.20b,c ± 0.54 6.35a ± 0.41 5.85a ± 0.47

5.45b ± 0.44 6.15b ± 0.58 7.15c ± 0.58 7.80c ± 0.54 5.80b ± 0.42 5.45b ± 0.44 3.60a ± 0.52

5.15b ± 0.47 6.40c ± 0.39 7.25d ± 0.26 7.70d ± 0.35 6.35c ± 0.41 4.50a,b ± 0.41 4.00a ± 0.53

6.20b,c ± 0.35 6.55c ± 0.37 7.45d ± 0.44 8.10d ± 0.39 6.40b,c ± 0.39 5.60b ± 0.39 4.25a ± 0.49

a–d

Mean in the same column followed by different superscript letters differs significantly (p b 0.05).

the dough microstructure, the rheological and sensorial analyses (data not shown). In this case only the microstructural parameters were not significantly correlated with the sensory attributes.

G″ led to a decrease of POV and an increase of both OSVR and FI parameters. Therefore, the decrease of the water content caused a decrease in the percentage volume of pores and an increase of the porosity with pores more disconnected. As it can also be noted there is an inverse correlation among the microstructural parameters, POV and OSVR. This is as expected as in general an increase in POV leads to a decrease in OSVR, as during baking the rupture of the cell walls leads to the formation of larger pores therefore a decrease in surface to volume ratio and an increase in percentage volume of the pores (Campbell, 2003). On the other hand, the decrease of the POV parameter and the increase of the SMI and FI parameters caused an increase of the sensorial parameters and then an improving overall quality. Moreover, the increase of the large bubbles also led to an increase of the overall quality of the manufactured bread. Same results were also found about the correlations among

4. Conclusions In this work the influence of different water concentrations on the rheological, microstructural and sensorial properties of durum wheat bread was evaluated. Results of the sensorial properties highlighted that the overall quality of the investigated bread samples has a maximum at the dough water content equal to that of Br-w64 sample. However, the bread samples with intermediate water content (Br-w61 and Br-w64) were preferred by the panelists. In fact, these samples showed a soft crumb with presence of large bubbles. The rheological and microstructural analyses also highlight that dough and bread structure are

Table 4 The correlation results for the microstructural, rheological and sensorial parameters of the durum wheat bread samples.

Maximum creep strain (%) Maximum recovery strain (%) G′ (Pa) G″ (Pa) POV OSVR SMI FI Crumb firmness Large bubbles Overall quality a

Maximum creep strain (%)

Maximum recovery strain (%)

G′ (Pa)

G″ (Pa)

POV

OSVR

SMI

FI

Crumb Firmness

Large bubbles

Overall quality

1 – – – – – – – – – –

0.9424a 1 – – – – – – – – –

−0.6070a −0.7553a 1 – – – – – – – –

−0.6206 −0.781a 0.9933a 1 – – – – – – –

0.8691a 0.9581a −0.8862a −0.9097a 1 – – – – – –

−0.7566a −0.8679a 0.9035a 0.9350a −0.9592a 1 – – – – –

−0.7070a −0.7267a 0.4645 0.4567 −0.649a 0.4275 1 – – – –

−0.7433a −0.8617a 0.7406a 0.7441a −0.8561a 0.7012a 0.8794a 1 – – –

−0.4742 −0.7267a 0.2528 0.2529 −0.8557a 0.3838 0.5056 0.5458 1 – –

−0.4513 −0.7038a 0.1924 0.2033 −0.5454 0.2995 0.6469a 0.6651a 0.9324a 1 –

−0.5451 −0.7749a 0.3382 0.3348 −0.6126a 0.4365 0.5953a 0.6549a 0.9888a 0.9516a 1

Highly and significantly correlated.

466

M. Mastromatteo et al. / Food Research International 51 (2013) 458–466

strongly affected by the water content. In fact, an increase in the water content caused an increase in the percentage volume of pores of the bread sample resulting in more soft dough. In conclusion, water addition in commercial practice needs to be controlled very carefully otherwise unsuitable water addition can compromise the bread quality and then the economic benefits. References Amemiya, J. I., & Menjivar, J. A. (1992). Comparison of small and large deformation measurements to characterize the rheology of wheat flour doughs. Journal of Food Engineering, 16, 91–108. Angioloni, A., & Collar, C. (2009). Significance of structuring/prebiotic blends on bread dough thermo-mechanical profile. European Food Research and Technology, 229, 603–610. Baker, J. C., Parker, H. K., & Mize, M. D. (1946). Supercentrifugates from dough. Cereal Chemistry, 23, 16–30. Blanshard, J. M. V., Frazier, P. J., & Galliard, T. (1986). Chemistry and physics of baking. London: The Royal Society of Chemistry. Bloksma, A. H., & Bushuk, W. (1988). Rheology and chemistry of dough. In Y. Pomeranz (Ed.), Wheat chemistry and technology. MN, St. Paul: American Association of Cereal Chemistry. Bohlin, L., & Carlson, T. L. G. (1980). Dynamic viscoelastic properties of wheat flour dough: Dependence on mixing time. Cereal Chemistry, 57(3), 174–177. Boyacioglu, M. H., & D'Appolonia, B. L. (1994). Durum wheat and bread products. Cereal Foods World, 39, 168–174. Campbell, G. M. (2003). Bread aeration. In S. P. Cauvain (Ed.), Breadmaking: Improving quality (pp. 352–374). Cambridge: Woodhead. Chillo, S., Laverse, J., Falcone, P. M., & Del Nobile, M. A. (2007). Effect of carboxymethylcellulose and pregelatinized corn starch on the quality of amaranthus spaghetti. Journal of Food Engineering, 83, 492–500. Del Nobile, M. A., Chillo, S., Mentana, A., & Baiano, A. (2007). Use of the generalized Maxwell model for describing the stress relaxation behavior of solid-like foods. Journal of Food Engineering, 78, 978–983. Dimitreli, G., & Thomareis, A. S. (2008). Effect of chemical composition on the linear viscoelastic properties of spreadable type processed cheese. Journal of Food Engineering, 84, 368–374. Dreese, P. C., Faubion, J. M., & Hoseney, R. C. (1988). Dynamic rheological properties of flour, gluten, and gluten-starch doughs. II — Effect of various processing and ingredient changes. Cereal Chemistry, 65(4), 354–359. Edwards, N. M., Dexter, J. E., Scanlon, M. G., & Cenkowski, S. (1999). Relationship of creep–recovery and synamic oscillatory measurements to durum wheat physical dough properties. Cereal Chemistry, 76, 638–645. Hibberd, G. E. (1970a). Dynamic viscoelastic behaviour of wheat flour doughs. Part 2: Effect of water content in the linear region. Rheologica Acta, 9, 497–500. Hibberd, G. E. (1970b). Dynamic viscoelastic behaviour of wheat flour doughs; Part III: The influence of starch granules. Rheologica Acta, 9, 501–505. Hildebrand, T., & Ruegsegger, P. (1997). A new method for the model independent assessment of thickness in three dimensional images. Journal of Microscopy, 185, 67–75. Lazaridou, A., Duta, D., Papageorgiou, M., BeIc, N., & Biliaderis, C. G. (2007). Effects of hydrocolloids on dough rheology and bread quality parameters in gluten-free formulations. Journal of Food Engineering, 79, 1033–1047.

Letang, C., Piau, M., & Verdier, C. (1999). Characterization of wheat flour–water doughs. Part I: Rheometry and microstructure. Journal of Food Engineering, 41, 121–132. Li, W., Dobraszczyk, B. J., & Schofield, J. D. (2003). Stress relaxation behavior of wheat dough, gluten, and gluten protein fractions. Cereal Chemistry, 80, 333–338. Lim, K. S., & Barigou, M. (2004). X-ray micro-tomography of cellular food products. Food Research International, 37, 1001–1012. Maache-Rezzoug, Z., Bouvier, J. M., Allaf, K., & Patras, C. (1998). Study of mixing in connection with the rheological properties of biscuit dough and dimensional characteristics of biscuits. Journal of Food Engineering, 35, 43–46. MacRitchie, F. (1976). The liquid phase of dough and its role in baking. Cereal Chemistry, 53, 318–326. Masi, P., Cavella, S., & Sepe, M. (1998). Characterization of dynamic viscoelastic behavior of wheat flour doughs at different moisture contents. Cereal Chemistry, 75, 428–432. Navickis, L. L., Anderson, R. A., Bagley, E. B., & Jasberg, B. K. (1982). Viscoelastic properties of wheat flour doughs: Variation of dynamic moduli with water and protein content. Journal of Texture Studies, 13, 249–264. Petrofsky, K. E., & Hoseney, R. C. (1995). Rheological properties of dough made with starch and gluten from several cereal sources. Cereal Chemistry, 72(1), 53–58. Sahi, S. S. (2003). The interfacial properties of the aqueous phases of full recipe bread doughs. Journal of Cereal Science, 37, 205–214. Skendi, A., Papageorgiou, M., & Biliaderis, C. G. (2010). Influence of water and barley [beta]-glucan addition on wheat dough viscoelasticity. Food Research International, 43, 57–65. Smith, J. R., Smith, T. L., & Tschoegl, N. W. (1970). Rheological properties of wheat flour doughs: III. Dynamic shear modulus and its dependence on amplitude, frequency, dough composition. Rheologica Acta, 9(2), 239–252. Spies, R. (1997). Application of rheology in the bread industry. In H. Faridi, & J. M. Faubion (Eds.), Dough rheology and baked product texture (pp. 343–361). USA, Houston: CBS Press. Steffe, J. F. (1992). Rheological methods in food process engineering. USA, Michigan: East Lansing, Freeman Press. Szczesniak, A. S., Loh, J., & Wesley, R. (1983). Effect of moisture transfer on dynamic viscoelastic parameters of wheat flour/water systems. Journal of Rheology, 27, 537–556. Umbach, S. L., Davis, E. A., Gordon, J., & Callaghan, P. T. (1992). Water self-diffusion coefficients and dielectric properties determined for starch–gluten–water mixtures heated by microwave and by conventional methods. Cereal Chemistry, 69, 637–642. Van Bockstaele, F., De Leyn, I., Eeckhout, M., & Dewettinck, K. (2008). Rheological properties of wheat flour dough and the relationship with bread volume. I. Creep– recovery measurements. Cereal Chemistry, 85, 753–761. Van Bockstaele, F., De Leyn, I., Eeckhout, M., & Dewettinck, K. (2011). Non-linear creep– recovery measurements as a tool for evaluating the viscoelastic properties of wheat flour dough. Journal of Food Engineering, 107, 50–59. Wang, F. C., & Sun, X. S. (2002). Creep–recovery of wheat flour doughs and relationship to other physical dough tests and breadmaking performance. Cereal Chemistry, 79, 567–571. Weegels, P. L., Orsel, R., van de Pijpekamp, A. M., Lichtendonk, W. J., Hamer, R. J., & Schofield, J. D. (1995). Functional properties of low Mr wheat proteins. II. Effects on dough properties. Journal of Cereal Science, 21, 117–126. Xianzhong, X., & Jinping, H. (2011). A stress relaxation model for the viscoelastic solids based on the steady-state creep equation. Mechanisms Time-Dependent Materials, 15, 29–39.