Effect of pentosan addition on dough rheological properties

Food Research International 43 (2010) 2315–2320 Contents lists available at ScienceDirect Food Research International j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / f o o d r e s Effect of pentosan addition on dough rheological properties Massimo Migliori, Domenico Gabriele ⁎ Department of Engineering Modelling, University of Calabria, Via P. Bucci, Cubo 39C, I 87030 Rende (CS), Italy a r t i c l e i n f o Article history: Received 13 May 2010 Accepted 6 August 2010 Keywords: Rheological models Oscillatory measurements Transient test Stress relaxation Nonlinear behaviour a b s t r a c t In this paper a complete rheological characterisation of bread dough added with water-solvable pentosans is showed. In the literature several works are available showing the chemical and physical effect of pentosan addition but it is still matter of discussion of their effect on the mechanical properties of dough. Therefore, the main objective is to further study this point, evaluating the effect of pentosans on the rheological properties of dough, using fundamental measurements and rheological modelling. Small amplitude oscillations at different temperatures were performed to evaluate material properties and stress relaxation tests, either within or out of the linear range, were used to investigate the effect of large deformations on material structure. Results showed that the effect of the addition is variable, depending on the amount, type of pentosans and deformation amplitude. The obtained results, together with rheological modelling, allow either to design dough having controlled properties during critical manufacturing steps (e.g. leavening or baking) or to reduce mechanical properties variability as effect of natural variation in flour characteristics. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction Flour based products are always requested by consumers and their demand is sharply increased during the last decade. This phenomenon, apart from normal market increase, is basically due to new trend of flour manufacturers that are increasing the offer of “special flours” particularly designed for specific application such as: home-made bread, pizza, spongy cakes, biscuits, and “self-leavening” flour. According to the craft-made preparation process and the specific recipe characteristics (e.g.: vegetable fats, sugar, raising agent addition) the flour should be able to guarantee in any condition the desired performance, even though in-home facilities are strongly different from the industrial one. In this concern researchers have to look into two different problems: either to design the specific flour composition in order to get the expected performance or to guarantee constant flour characteristics time to time, tackling natural flour composition changes. In this concern the addition of natural additives, obtained either by different flour fractions or by different vegetal sources, may play a crucial role helping in control flour characteristics (Fustier, Castaigne, Turgeon, & Biliaderis, 2008; Fustier, Castaigne, Turgeon, & Biliaderis, 2009; Paraskevopoulou, Provatidou, Tsotsiou, & Kiosseoglou, 2010; Skendi, Biliaderis, Papageorgiou, & Izydorczyk, 2010). Water soluble pentosans (WSP) and unextractable solids (WUS) are the major nonstarch component of flour, being mostly based on hemicellulose (Yin & Walker, 1990) and their role, as functional ingredients, is well known (D'Appolonia & Gilles, 1971; ⁎ Corresponding author. Tel.: + 39 0984 496687; fax + 39 0984 494009. E-mail addresses: [email protected] (M. Migliori), [email protected] (D. Gabriele). 0963-9969/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2010.08.008 Michniewicz, Biliaderis, & Bushuk, 1991). Very extensive chemical investigations of water soluble (WSP) and insoluble (WIP) pentosans are available, even though different explanations still hold about their ability in conditioning rheological properties of dough (Denli & Ercan, 2001; Wang, van Vliet, & Hamer, 2004a). A very accurate description of the status of the art about pentosan (and xylanase) effect on dough is proposed by Wang et al. (2004a) and the same authors also identified that water affinity change due to pentosan presence is driven both by chemical and physical effect (Wang, van Vliet, & Hamer, 2004b). In this series of paper, a rheological characterisation at room temperature (Wang et al., 2003), is applied to different mixtures of glutenin macro-polymer (GMP) and pentosans, in order to quantify mechanical impact of the addition, showing that WSP affect GMP rheological properties and the effect can be adapted to the desired value by changing water content and/or mechanical energy input. The authors also showed that WSP treated under freeze-drying conditions did not exhibit any significant difference when compared with nondried one. This represents valuable information because, in the view of industrial application, the effect of storage conditions and technique on pentosan activity has to be clearly understood. The effects of soluble and insoluble pentosans on breadmaking were examined measuring specific volume and staling of breads (Maeda & Norita, 2003) prepared by flours from polishing wheat grain. Positive results, in terms of increase in loaf volume and retarded firmness, were obtained for both types of pentosans, even though larger amount of WIP than WSP seems necessary to obtain good baking characteristics. The quality improvement caused by WSP was attributed by the authors to their viscous and gelling properties that improve the strength of gluten network and, as a consequence, the gas retention (Maeda & Norita, 2003). 2316 M. Migliori, D. Gabriele / Food Research International 43 (2010) 2315–2320 In this paper the characterisation of standard flour added with WSP is presented, aiming to investigate the effect of temperature and WSP amount on the rheological characteristics of dough by performing a more extensive fundamental characterisation also using transient experiment. In addition the effect of one WSP storage technique is investigated by analysing performance of WSP treated using a spray drying process, in order to increase the storage time. 2. Materials and methods Samples were prepared by using commercial flour for breadmaking and pentosans kindly supplied by IPALC (Italy). In order to check the effect of the dehydration process on the pentosan activity, the starting solution was also spray dried and the same type of pentosans were supplied by IPALC in two different ways: the original water solution and the powder obtained by spray drying the solution. Water content of either flour and pentosans was determined by using an halogen moisture analyzer (HB43, Mettler Toledo, Germany) set at 130 °C and recording the final moisture (on wet basis) when weight loss rate was lower than 1 mg/140 mg. A moisture of 12.6 ± 0.4% (w/w) was determined for flour, whilst a solid content of 6.8 ± 0.2% (w/w) was obtained for the pentosan solution and the powder moisture was found lower than 5%. 2.1. Sample preparation The reference sample (B sample), including only flour and water in a mass ratio of 2:1, was modified adding pentosans from both sources and keeping a constant weigh ratio (1:100 or 1.5:100) between pentosans and flour. When the solution was used, the proper amount was computed, on the basis of the determined solid content, to keep the desired pentosans to flour ratio. When pentosans in powder were tested, because of the small amount of powder added to the flour, the water coming from the powder was neglected when changing the recipe. Three samples were obtained and details on their composition are reported in Table 1. Dough was prepared by mixing ingredients in a lab scale mixer (Hobart, UK) equipped with a flat beater (B type, Hobart, UK) and keeping 5 min as constant mixing time. Temperature of the mixing bowl was controlled by circulating water at 30 °C in an external jacket. After mixing, dough slabs were prepared by using a commercial handroller machine (Imperia, Italy), the sheeting procedure was standardised and after six pass through rollers (decreasing the nip), a final thickness of approximately 2 mm was obtained. In order to relax sheeting-induced stresses, samples were stored in aluminium foil for 20 min at room temperature before starting the rheological characterisation. No water loss was measured during this storage period. 2.2. Rheological characterisation Rheological characterisation of dough has been performed using either a controlled stress rheometer (DSR200, Rheometric Scientific, USA) for small amplitude oscillation test or a controlling strain rheometer (ARES-RFS, TA Instruments, USA) for transient stress relaxation tests. Both instruments were equipped with parallel plate geometry (ϕ = 25 mm) and temperature control was ensured by a Table 1 Sample composition. Ingredient Flour (g) Water (g) Pentosan powder (g) Pentosan solution (g) Sample ID B 1%P 1%S 1.5%S 500 250 – – 500 250 5 500.0 181.5 – 73.5 500.0 147.2 – 110.3 Peltier system acting under the lower plate. Potential slippage in nonlinear viscoelastic conditions was eliminated by applying commercial adhesive-backed sandpaper (grit P40) on surfaces of either the Peltier plate or the upper plate tool. A constant gap of 2.2 ± 0.1 mm was kept during all tests. Oscillatory tests included stress sweep test at the fixed frequency of 1 Hz performed at different temperatures (30 °C, 50 °C, 70 °C and 90 °C), in order to found the linear viscoelastic region of the material. At the investigated temperatures, frequency sweep tests were carried out, in linear conditions, in the frequency range 0.1–10 Hz. Even though lower frequencies are often used in material characterization to yield information on long relaxation processes, according to the rheological model adopted in the present paper to describe dough behaviour, the 0.1–10 Hz frequency range was considered adequate to fully describe the rheological behaviour of tested samples. Dynamic characterization was completed by a temperature ramp test (time cure) at 1 Hz, in linear viscoelastic conditions, increasing temperature from 25 °C to 100 °C at 1 °C/min. Transient tests were performed either within or out of the linear viscoelastic range of strain (Keentok, Newberry, Gras, Bekes, & Tanner, 2002; Uthayakumaran, Newberry, Phan-Thien, & Tanner, 2002) imposing two different strain: 0.04% (typical value for linear viscoelastic range determined by strain sweep tests) and 4%, well out of the linear region, in order to study mechanical behaviour under large deformation. It was not possible to carry out transient tests in nonlinear conditions at high temperature because the increase in dough consistency, caused by the starch gelatinization, requires torque values above the rheometer upper limit (maximum torque 0.1 N m). Therefore, stress relaxations, both within and out of the linear conditions, were carried out only at 30 °C and 50 °C. During all tests water loss was reduced by covering the sample rim with a silicon oil (Silicon oil DC, Fluka, Italy) having a viscosity enough low (300 mPa s), if compared with dough consistency, to avoid any effect on the test results. All the tests were performed on three independent preparations and a good reproducibility was found; experimental data are shown in terms of average value and standard deviation (always lower than 5.0%). 2.3. Rheological data analysis According to literature (Phan-Thien, Safari-Ardi, & Morales-Patino, 1997; Uthayakumaran et al., 2002) dough exhibits mainly a solid-like behaviour as confirmed by the absence of a real steady state shear viscosity. Any attempt to measure viscosity, in a transient test at constant shear rate, ended up to a continuous decrease of the viscosity without any steady viscosity plateau. This problem was faced by some authors (Phan-Thien et al., 1997) that also recognised the exigency to use for dough a “solid-like” constitutive equation. A solid-like approach, recently proposed, is based on the observation that dough, such as many other foods, exhibits a critical gel behaviour (Gabriele, D'Antona, & de Cindio, 2001; Ng & McKinley, 2008; Ng, McKinley, & Padmanabhan, 2006). This approach, originally proposed to describe the linear viscoelastic behaviour in dynamic tests (Gabriele et al., 2001) has also been used to describe other types of deformation, including stress relaxation in linear and nonlinear conditions for gluten dough (Ng & McKinley, 2008). According to this literature, rheological properties of dough have been interpreted considering it as a solid behaving as a “weak gel” in the investigated frequency range (Gabriele et al., 2001). The proposed model includes the prediction of a linear behaviour in a double-log plot of the dynamic moduli vs the frequency and, referring to complex modulus G*, data can be fitted using a two parameter power law:  G ðωÞ = pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 G′2 + G″2 = Ad ω z ð1Þ M. Migliori, D. Gabriele / Food Research International 43 (2010) 2315–2320 where, according to the gel theory (Winter & Chambon, 1986), A can be interpreted as a measure of the network strength and z is related to the number of interacting rheological units forming the threedimensional network. It is worth reminding that dough rheological behaviour is described by this power law model in a limited frequency window, usually ranging between 0.1 and 100 Hz, therefore experimental frequency sweep tests were carried out only within this region. Stress relaxation tests in a defined time range (0.1–10 s) revealed a power law trend, typical of a critical gel behaviour, as already found in the literature (Keentok et al., 2002; Ng & McKinley, 2008) for other type of dough. At shorter times an additional Rouse-like response can be observed, whilst at long times, terminal relaxation processes can appear (Ng & McKinley, 2008). As a consequence the relaxation modulus, in the considered time window, can be fitted by a power law model: Gðt Þ = Sd t −n ð2Þ S can be correlated to A by Fourier-transforming Eq. (2) to obtain dynamic moduli (Winter & Chambon, 1986), yielding: S= A Γð1−nÞ ð3Þ Therefore S is related to the strength of the network whilst n is inversely related to the network extension, both parameters characterise the mechanical response under shear step strain experiment. It should be noticed that even though both dynamic and transient tests are able to characterise the viscoelastic behaviour of materials, different information can be obtained. In fact when oscillatory tests are performed, Deborah number (De) is kept at its unitary value and the (linear) viscoelastic structure is investigated at different frequency; on the contrary, during transient tests, observation time is changed and De changes from very high values to very low ones, separating the solid-like behaviour form the liquid-like one. Experimental data from frequency sweep and stress relaxation tests were fitted according to Eqs. (1) and (2), respectively, and obtained parameters are reported as mean value and standard error. 3. Results and discussion 3.1. Dynamic temperature ramp test A typical time cure for sample B is shown in Fig. 1, with the aim to enhance the effect of temperature on the dough characteristics. Starting from 25 °C, G* decreases sensibly up to onset of starch gelatinisation temperature and then increases up to a maximum value Fig. 1. Time cure test for sample B, G′ (●), G″(○), loss tangent (▲). 2317 indicative of the completion of the gelatinisation phenomena. After this peak in temperature, the effects of the gelatinised starch become dominant (Salvador, Sanz, & Fiszman, 2000) and no more structural changes take place. Any further increase in temperature acts only on the kinetic of molecules mobility, determining the observed complex modulus decrease. Another relevant parameter to be considered is the loss tangent tan δ defined as: tan δ = G″ G′ ð4Þ Loss tangent greater than one indicates a predominant liquid-like behaviour as the loss modulus is greater than the elastic one. On the contrary more solid-like material exhibits a tan δ lower than 1. Initially in the time cure test an almost constant value of the loss tangent is observed, up to a temperature value higher than the gelatinisation-onset one; afterwards a continuous decrease, showing the tendency of the material towards a more “solid like” behaviour, is found. This result evidenced that, at the beginning of the gelatinisation process when water penetrates the amorphous zone of the starch granules and swelling starts, an increase in the network strength is observed whilst no relevant change in dough structure is detected. As the temperature rises, the progress in water absorption has a destabilising effect leading to the disorganisation and disruption of the crystalline zones (Salvador et al., 2000) and, from a rheological point of view, this causes a significant structuring effect (drop in loss tangent). The analysis of time cure (Fig. 1) evidences some relevant temperatures related to rheological changes in material: the onset (T 0 ) and the completion (T e ) of gelatinisation phenomena corresponding, respectively, to the minimum and the maximum in G* plot; the onset of structural change corresponding to the loss tangent decrease (Tsl) when the material starts developing a more pronounced solid-like behaviour. It is interesting to notice that up to To temperature, even though moduli value decrease, the material behaviour keeps its initial characteristics because loss tangent did not change with temperature and no prevailing decay of any moduli is observed up to Tsl. The difference between the two temperatures can help in distinguishing between two different phenomena: the increase of the moduli, as effect of the gelatinisation in progress, and the more “solid like” behaviour that can be primarily related to the network extension variation. It is also noteworthy from Eq. (1) that complex modulus, measured at 1 Hz in linear viscoelastic conditions, is numerically the same than A parameter, therefore time cure yields the trend of the network strength as a function of temperature. Time cure results show a similar temperature dependence for all tested samples (Fig. 2a) and complex modulus data evidenced that 1% P exhibited the weakest structure along the whole temperature range, even if the G* gap is reduced at high temperatures at the end of the gelatinisation process. The comparison between samples 1%S and 1.5% S indicated that at temperature lower than T0 a larger amount of pentosans yields to a stronger network. On the contrary, after the gelatinisation onset, the opposite trend is found and sample 1%S seems to be stronger than 1.5%S up to Te where differences are not relevant anymore. When the loss tangent trend is considered (Fig. 2b), it can be seen that samples B, 1%S and 1.5%S are characterised by a sharp decrease of this parameter as evidence of a more clear turn-up of the solid-like behaviour when increasing temperature. On the contrary the sample 1%P shows a transition characterised either by an early onset (lower Tsl temperature) or by a smoother shape. This trend is confirmed when the characteristic temperatures are considered (Table 2), manually determined for all the investigated samples. It can be seen that the addition of any amount and type of 2318 M. Migliori, D. Gabriele / Food Research International 43 (2010) 2315–2320 Fig. 2. Time cure test for investigated samples, a) complex modulus, b) loss tangent. Fig. 3. Frequency sweep test for sample 1%P, a) complex modulus, b) loss tangent. pentosans did not lead to relevant effects on the onset and final gelatinisation temperature whilst the addition of pentosan powder strongly affect the Tsl and the sample 1%P is characterised by an earlier drop in the loss tangent, even if the value of the parameter is always greater than that measured for the other samples. 3.2. Isothermal oscillatory measurements More detailed information can be obtained by frequency sweep tests performed at different temperatures in linear conditions, the typical trend is shown in Fig. 3 for sample 1%P in terms of complex modulus and loss tangent evidencing the mechanical response of a “weak gel”: dynamic moduli increase in value when increasing frequency, moduli are parallel in the investigated frequency region and G′ is greater than G″ of more than one order of magnitude (loss tangent is nearly constant and lower than 1 at any frequency). The same mechanical behaviour was found for all the investigated samples and data treatment according to the model proposed in Eq. (1) was applied computing weak gel model parameter (Table 3). At room temperature no relevant effect on dough structure was given by the addition of 1% WSP in solution, in fact network strength and extension are equal to those of B sample, whilst when increasing the WSP content (sample 1.5%S) the increase of A was found. On the contrary the 1%P showed a decrease of the network strength, whilst no relevant effect was found for the network extension z, for any of the pentosan addition. When temperature increases below the gelatinisation temperature, the decrease of A was observed up to 50 °C for 1%P and 1%S whilst the other two samples did not show a significant variation. The further increase of temperature up to 70 °C, determines the starch gelatinisation for all samples as demonstrated either by the increase of the network strength A of about one order of magnitude or by the increase of the network extension z that nearly doubles. As a result, the gelatinised dough is characterised by a more structured network even if the addition of pentosans affect the structuration process. In fact at 70 °C 1%S sample exhibits the highest A value, even though at lower temperature this sample has a network strength lower than B and 1.5%S. This suggests that, in terms of dough network strength, the addition of a lower amount of solubilised pentosans acts at higher temperature whilst, if the structurising effect is requested at lower temperature, a higher amount of added pentosans is requested. On the other hand the trend of dried pentosans as less active, is confirmed Table 3 Weak gel model analysis of dynamic frequency sweep tests (value±mean error). Sample A (Pa s1/z) z (–) T = 30 °C Table 2 Characteristic temperatures of samples as determined from time cure test. Sample ID To (°C) Te (°C) Tsl (°C) B 1%S 1.5%S 1%P 48 ± 1 52 ± 1 52 ± 1 52 ± 1 76 ± 1 80 ± 1 80 ± 1 78 ± 1 60 ± 1 68 ± 1 64 ± 1 38 ± 1 B 1%S 1.5%S 1%P 33,500 ± 700 32,400 ± 400 41,700 ± 600 24,500 ± 400 B 1%S 1.5%S 1%P 249,000 ± 4000 270,000 ± 2000 229,000 ± 4000 185,000 ± 2000 A (Pa s1/z) z (–) T = 50 °C 4.6 ± 0.3 4.3 ± 0.2 4.5 ± 0.2 4.0 ± 0.2 T = 70 °C 32,900 ± 500 28,000 ± 500 31,800 ± 400 19,900 ± 500 5.0 ± 0.3 4.5 ± 0.2 5.0 ± 0.2 4.3 ± 0.3 T = 90 °C 8.3 ± 0.7 8.2 ± 0.4 7.7 ± 0.7 8.0 ± 0.5 132,000 ± 2000 170,000 ± 2000 169,000 ± 2000 132,000 ± 1000 16 ± 3 15 ± 2 15 ± 2 14 ± 1 2319 M. Migliori, D. Gabriele / Food Research International 43 (2010) 2315–2320 because of the increase in network strength lower than the other samples. When the temperature is further increased, an opposite trend of network strength and extension is observed. In fact at 90 °C the dough network strength decreased and the structure extension increased in all samples as can be seen if the z value is considered. As far as the A value is concerned, it is worthy to observe that, despite the differences in the mechanical behaviour of 1%S and 1.%5S up to 70 °C, at 90 °C both samples exhibit the same network strength well higher than those of the reference sample B. On the contrary the addition of spraydried pentosans did not show any significant effect onto the mechanical behaviour of dough at high temperatures. 3.3. Transient tests: Stress relaxation The typical relaxation modulus for sample 1%S is reported in Fig. 4 at 30 °C and 50 °C in linear and nonlinear conditions. It can be seen that in both cases the relaxation modulus shows a linear trend in a log–log plot, even though a more scattered trend was found for low strain test because of the closeness to the instrument minimum sensitivity. The same trend was found for all the investigated samples and the model proposed (Eq. (2)) was used to fit experimental data; the values of mechanical parameters are summarised in Table 4. It can be seen that temperature causes a decrease of the network strength (S decreases from 30°C to 50 °C) and this confirms, also for transient mechanical response, the trend already discussed for dynamic test. For sample B, 1%S and 1.5%S the network extension does not depend on temperature, as demonstrated by the nearly constant n values, whilst the network strength, S, slightly decreases, probably owing to the kinetic effects. These results are in agreement with either data obtained from frequency sweep tests, showing an almost constant network extension between 30 °C and 50 °C, or time cure exhibiting a plateau in loss tangent up to the Tsl limit, always higher than 50 °C for considered samples. A different trend was found for sample 1%P characterised by a slight change in n value, probably due to the early structuring phenomena caused by the pentosans, in agreement with the lower Tsl estimated by dynamic tests. The effects of nonlinear conditions, i.e. large deformations, on material behaviour are more evident if the ratios between linear and nonlinear parameters, according to the following definitions, are considered: S* = S in non linear regime S in linear regime ð4Þ Table 4 Weak gel model analysis of transient tests (value ± mean error). T (°C) Strain 0.04% Strain 4% S (Pa sn) n (–) B 30 50 19,400 ± 300 19,100 ± 400 0.267 ± 0.008 0.26 ± 0.01 7580 ± 90 7300 ± 100 0.239 ± 0.007 0.226 ± 0.009 1%S 30 50 7650 ± 50 8020 ± 90 0.290 ± 0.005 0.267 ± 0.009 4250 ± 20 4150 ± 80 0.244 ± 0.004 0.240 ± 0.004 13,400 ± 100 8860 ± 60 0.271 ± 0.009 0.275 ± 0.006 11,660 ± 80 4240 ± 20 0.227 ± 0.006 0.223 ± 0.004 8550 ± 60 695 ± 9 0.257 ± 0.006 0.30 ± 0.01 3470 ± 20 2317 ± 7 0.249 ± 0.005 0.221 ± 0.003 1.5%S 30 50 1%P 30 50 n* = n in non linear regime n in linear regime S (Pa sn) n (–) ð5Þ It is worthy noticing that both S* and n*, related respectively to ratio in the strength of interactions and the inverse of network extension, are always lower than unity (see Fig. 5a–b). Therefore it seems that the increase in deformation promotes a larger network extension characterised by weaker interactions. Computed parameters show that pentosan addition yields a structure more resistant to large deformations, as evidenced by values of S* always larger than that computed for sample B, even though solubilised pentosans seems more active than dried ones, in agreement with dynamic data (Fig. 5a). When network extension is considered (Fig. 5b), no relevant effect is found, as evidenced by the value of n*, approximately constant for all tested samples. Therefore it seems that pentosan addition is particularly positive when large deformations occur, this effect can be useful to stabilize the mechanical behaviour of dough during process operations (for example leavening or baking) when local deformations can be quite large, improving the final properties. These results seem in agreement with literature data (Maeda & Norita, 2003) showing an improvement in bread quality as a consequence of the increase in gluten network strength caused by the addition of WSP (up to 1% w/w). Fig. 4. Relaxation modulus for sample 1%P in linear and nonlinear conditions. 2320 M. Migliori, D. Gabriele / Food Research International 43 (2010) 2315–2320 pentosans and temperature). Moreover the dough mechanical behaviour was revealed to be also more complex, as some of these effects disappeared when the conditions were changed. Therefore, in order to obtain the desired final dough characteristics, pentosan concentration and temperature conditions should be properly selected. In these concerns, considering that cereal based products are characterised by a wide range of kinematic conditions and temperature, rheological analysis can result as a useful tool to optimise dough mechanical properties. In fact, it can support the addition of specific ingredients, such as pentosans, aiming to tailor dough properties for any specific production, also overcoming the problem of natural variability of flour characteristics. Acknowledgments The work was funded by Italian Ministry of University 355 and Research (MIUR) project “New wheat based products” File no. 6971. Acknowledgments are due to the Laboratory of Rheology and Food Engineering at University of Calabria, where the experimental tests were performed by Mr. Claudio G. Filippo, who is also acknowledged. References Fig. 5. Dimensionless weak gel parameters for tested samples, a) S* (Eq. (4)), b) n* (Eq. (5)). 4. Conclusions Even though they were extensively studied, the effects of pentosan addition in dough is not yet completely understood and different opinions, mainly on rheological effects, can be found in the literature (Labat, Rouau, & Morel, 2002; Uthayakumaran et al., 2002; Wang et al., 2004a). In the present work, dynamic and transient tests were adopted to evidence the potential contribution of pentosans as “rheological modifiers” useful to prepare flours for specific applications. It was found that the addition of solubilised pentosans decreases the network strength at low temperature (T b 50 °C) whilst increases it at higher temperature; this effects is slightly modified by increasing the added amount and a network strength reduction can be noticed at the investigated intermediate temperature (between 50 °C and 70 °C). On the contrary a slight reduction of network extension was observed at any condition, even though the effect is more evident at high temperatures. The effect of spray-dried pentosans appeared to be strongly modified by the drying process and, as a consequence, only a reduction in both network strength and extension was observed for all tested conditions. They could be useful to reduce the strength of some flours making them more suitable to be processed. Nonlinear transient tests evidenced that pentosans seem more active when a large deformation is used, strongly increasing the network strength, than in linear viscoelastic conditions. Dough rheological behaviour is the result of complex interactions between different components (mainly starch, gluten and water) and the addition of pentosans introduces a new variable to be considered when dealing with dough mechanical properties. Results of this paper demonstrated that modification of rheological behaviour was found when changing some operating parameters (the amount of added D'Appolonia, B. L., & Gilles, K. A. (1971). Pentosans associated with gluten. Cereal Chemistry, 48, 427−436. Denli, E., & Ercan, R. (2001). Effect of added pentosans isolated from wheat and rye grain on some properties of bread. European Food Research and Technology, 212, 374−376. Fustier, P., Castaigne, F., Turgeon, S. L., & Biliaderis, C. G. (2008). Flour constituent interactions and their influence on dough rheology and quality of semi-sweet biscuits: A mixture design approach with reconstituted blends of gluten, watersolubles and starch fractions. Journal of Cereal Science, 48, 144−158. Fustier, P., Castaigne, F., Turgeon, S. L., & Biliaderis, C. G. (2009). Impact of endogenous constituents from different flour milling streams on dough rheology and semisweet biscuit making potential by partial substitution of a commercial soft wheat flour. LWT Food Science and Technology, 42, 363−371. Gabriele, D., D'Antona, P., & de Cindio, B. (2001). A weak gel model for foods. Rheologica Acta, 40, 120−127. Keentok, M., Newberry, M. P., Gras, P., Bekes, F., & Tanner, R. I. (2002). The rheology of bread dough made from four commercial flours. Rheologica Acta, 41, 173−179. Maeda, T., & Norita, N. (2003). Flour quality and pentosan prepared by polishing wheat grain on breadmaking. Food Research International, 36, 603−610. Michniewicz, J., Biliaderis, C. G., & Bushuk, W. (1991). Effect of added pentosans on some physical and technological characteristics of dough and gluten. Cereal Chemistry, 68, 252−258. Ng, T. S. K., McKinley, G. H., & Padmanabhan, M. (2006). Linear to non-linear rheology of wheat flour dough. Applied Rheology, 16(5), 265−274. Ng, T. S. K., & McKinley, G. H. (2008). Power law gels at finite strains: The nonlinear rheology of gluten gels. Journal of Rheology, 52(2), 417−449. Paraskevopoulou, A., Provatidou, E., Tsotsiou, D., & Kiosseoglou, V. (2010). Dough rheology and baking performance of wheat flour–lupin protein isolate blends. Food Research International, 43(4), 1009−1016. Phan-Thien, N., Safari-Ardi, M., & Morales-Patino, A. (1997). Oscillatory and simple shear flows of a flour water dough: A constitutive model. Rheologica Acta, 36, 38−48. Salvador, A., Sanz, T., & Fiszman, S. M. (2000). Dynamic rheological characteristics of wheat flour–water doughs. Effect of adding NaCl, sucrose and yeast. Food Hydrocolloids, 20, 780−786. Skendi, A., Biliaderis, C. G., Papageorgiou, M., & Izydorczyk, M. S. (2010). Effects of two barley b-glucan isolates on wheat flour dough and bread properties. Food Chemistry, 119, 1159−1167. Uthayakumaran, S., Newberry, M. P., Phan-Thien, N., & Tanner, R. I. (2002). Small and large strain rheology of wheat gluten. Rheologica Acta, 41, 162−172. Labat, E., Rouau, X., & Morel, M. H. (2002). Effect of flour water-extractable pentosans on molecular associations in gluten during mixing. Lebensmittel-Wissenschaft und Technologie, 35, 185−189. Wang, M. W., Hamer, R. J., van Vliet, T., Gruppen, H., Marseille, J. P., & Weegels, P. L. (2003). Effect of water unextractable solids on gluten formation and properties: Mechanistic considerations. Journal of Cereal Science, 37, 55−64. Wang, M. W., van Vliet, T., & Hamer, R. J. (2004a). Evidence that pentosans and xylanase affect the re-agglomeration of the gluten network. Journal of Cereal Science, 39, 341−349. Wang, M. W., van Vliet, T., & Hamer, R. J. (2004b). How gluten properties are affected by pentosans. Journal of Cereal Science, 39, 395−402. Winter, H. H., & Chambon, F. (1986). Analysis of linear viscoelasticity of a cross-linking polymer at the gel point. Journal of Rheology, 30, 367−382. Yin, Y., & Walker, C. E. (1990). Pentosans from gluten-washing wastewater: Isolation, characterization and role in baking. Cereal Chemistry, 69, 592−596.