3.1. Gluten-Free Dough Properties
Dough consistency (water addition).
Dough consistency is a crucial parameter in breadmaking. Generally, with regard to wheat bread doughs, a 500 BU consistency is considered as “optimal” for the rheological properties of doughs and the final quality of breads. On the contrary, with regard to GF breads, such specific and robust indications are not yet available, and various consistency levels can be found in the scientific literature to produce an acceptable GF dough, essentially as a function of the GF recipe adopted. When using the same GF recipe, a lower consistency (i.e., higher water levels in the dough) has been found as preferable to assure good dough performances during leavening, in particular when ingredients having a high water affinity are included into the recipe [33
]. A recent review revealed that textural parameters of GF crumbs are strongly related to dough consistency, and that a low dough consistency corresponds to a softer crumb [30
]. The use of low GF bread dough consistency was also adopted in a recent study on the use of high pressure treated raw materials in GF baking [32
]. However, the consistency cannot be too low. Finding a proper balance is therefore the crucial point in GF.
In some previous studies [27
], the authors investigated a 230 BU GF bread dough consistency. In this research, 180 ± 10 BU was chosen as a reference for all the reasons stated above. The sample containing sourdough, which is usually the most critical in terms of leavening properties and dough development, was replicated at a 230 BU level consistency for comparison. The amounts of water required to reach the desired consistencies were the following (expressed as water absorption; WA, %): SD230, 57.8%; SD180, 64.5%; CY180, 82.0%; SDCY180, 64.5%. As already observed in the previous study [27
], being SD and SDCY recipes with 20% of a high moisture ingredient, they required less water to reach the desired consistency.
Dough leavening properties
. The development of the various doughs during leavening, as well as their CO2
production and retention, as measured by the rheofermentographic test, were used to set up the proofing times to be adopted for each formulation during the real breadmaking process. Therefore, based on the previous experience [27
], the test was carried out for different lengths in relation to the leavening agent used (SD230, 16 h 30 min; SD180, 15 h; CY180, 3 h; SDCY180, 16 h). In particular, for those doughs containing the SD, the test was stopped when a pH of 4.0 ± 0.2 was reached. The rheofermentographic profiles of the various doughs are reported in Figure 1
As expected, SD-doughs required longer periods of time to actively start the proofing, in comparison to the other dough systems. However, after that lag-period, SD microorganisms started their fermentation process, as well. In particular, the higher presence of water in SD180 probably guided the quicker and more intense height development of the dough, with respect to SD230 (Hm = 30.1 mm, SD180 vs. Hm = 14.5 mm, SD230). The higher amount of water in the dough could have been, at the same time, the cause of the larger decrease in SD180 dough height at the end of the test in comparison to SD230 (Hf = 21.1 mm, SD230 vs. Hf = 12.1 mm, SD180). Interesting dough height developments were reached in a very short period of time for CY180 (Hm = 28.2 mm; Hf = 26.2 mm), as expected. The recipe SDCY180 turned out to be very effective: the microorganisms, in fact, began the fermentation in a very short time and with a rapid growth (Hm = 38.1 mm), suggesting a synergistic effect among those microbial populations. To test this hypothesis, the development rates of these last three samples (SD180, CY180, and SDCY180) were quantified via linear interpolation (R > 0.997) of the development curves between 5 and 20 mm of dough height. The following dough height development kinetics were obtained: SD180, 13.33 mm/h; CY180, 24.13 mm/h; SDCY180, 29.70 mm/h. The positive effect of combining SD and CY in the same system is thus clear. Even if their proofing rates were largely different, the maximum dough height developments were similar for all of these samples, suggesting that all of the leavening agents taken into consideration were suitable for the processing conditions adopted in this study.
Dough heights are generally a reflection of CO2
production and retention in the system. As can be appreciated from Figure 1
, dough heights generally increase as long as the doughs are able to completely retain inside all of the produced CO2
. As soon as Tx appears (time when the dough begins to give off CO2
), a stability or a decrease in dough height is attained, in relation to the system’s strength [1
]. Tx was equal to 5 h, 3 h 43 min, 1 h 52 min, and 1 h 43 min for SD230, SD180, CY180, and SDCV180, respectively. All the observations previously reported are congruent with these findings.
As regards the doughs retention coefficients (Rc, %), the highest value was observed for CY180 (Rc = 95.2%), while the others were more close and similar to each other (Rc = 87.6%, SD230; Rc = 89.0%, SD180; Rc = 88.3%, SDCY180). However, when considering data related to CO2 production and retention, the influence of the leavening agent, the water level, the dough’s strength, and the test duration should be taken into account contemporarily.
On the basis of the rheofermentographic results, and particularly considering the Tx values, the following leavening times for bread production were defined: SD230, 3 h; SD180, 3 h; CY180, 1 h 30 min; SDCY180, 1 h. These times were slightly shorter than their corresponding Tx, since during the first period of baking, a further dough volume increase generally takes place. GF dough systems are usually very weak, and prolonging the proofing phase too much could result in a collapse (instead of a further development) of the structure during baking.
Dough pH evolution during proofing. In parallel to the rheofermentographic test, pH evolution during proofing was recorded; for those doughs containing the SD, the test was stopped when a pH of 4.0 ± 0.2 was reached. The pH values of the various samples at the beginning of the rheofermentographic test were the following: SD230, 5.28; SD180, 5.30; CY180, 5.51; SDCY180, 5.13. During the test, pH decreased (curves not reported) by 1.35 and 1.36 units for SD230 and SD180, of 0.51 for CY180 and of 1.02 for SDCY180.
3.2. Gluten-Free Bread Properties
Features of the fresh products
. The main characteristics of the GF fresh breads 1 h after baking are reported in Table 2
. Significant differences (p
< 0.05) were evidenced among the samples after baking: loaves weights varied at 123.1 ± 1.3 g (SDCY180-bread) and 135.0 ± 1.0 g (SD180-bread), and baking losses were equal to 13.0% for SD230-bread, 10.0% for SD180-bread, 10.1% for CY180-bread, and 17.9% for SDCY180-bread.
Significant differences (p
< 0.05) were also evidenced in terms of maximum height, minimum height, volume, and specific volume. For all these parameters, samples could be ordered as follows (from the lowest to the highest value): SD230-bread < SD180-bread < CY180-bread < SDCY180-bread. The best performances, therefore, were reached when a 180 BU dough consistency was adopted. More specifically, SDCY180-bread exhibited the highest values in terms of geometrical features (Table 2
; Figure 2
). As reported in other studies, the addition of sourdough to GF bread containing yeast as a leavening agent generally does not exert a remarkable influence on its specific volume [7
]. On the contrary, in the present study, the contemporary presence of CY and SD determined a 19.5% increase in bread specific volume (with respect to CY180-bread specific volume), suggesting that the metabolites produced by the lactic acid bacteria were effective in modifying the rheological properties of SDCY180 dough, improving its deformation capability during proofing and baking.
Another important parameter for consumer acceptability is bread softness, which is related to both the crumb porosity and the characteristics of the alveolar walls. GF bread texture was evaluated by a compression test during which the resistance offered by the crumb to the compression was continuously recorded. Young’s modulus was calculated from the resulting stress versus strain curve: the higher the value, the higher the loaf hardness. The two SD-loaves were significantly different from the other samples, and between them: as expected, the SD230-loaf exhibited the highest Young’s Modulus. As can be determined from Figure 2
, these bread loaves were characterized by a denser and heavier crumb structure, in comparison to the others. The two breads containing CY, on the contrary, were much softer and not significantly different between them, independently of the amount of water required to reach the 180BU dough consistency (82.0% for CY180; 64.5% for SDCY180) and from their slice and crumb moisture values (Table 2
These results were aligned with the crumb porosity features (Table 3
; Figure 2
): SD230-bread and SD180-bread crumbs were very dense, with a high percentage of holes (>90%) and a mean diameter of 0.6 mm, whereas CY180-bread and SDCY180-bread crumbs also exhibited a high amount (17%) of pores with an intermediate size (1 < x
≤ 3 mm). The presence of big holes (mean diameter > 2.2 mm) was very restricted in all the samples, indicating a limited structure breakdown during leavening and baking. Overall, SD230-bread and SD180-bread were characterized by a minor total mean alveolate area (14.0% and 16.5%, respectively), whereas CY/SD180-bread and CY180-bread showed values of 24.8% and 21.0%, respectively. All of these features are critical points in building up the final GF breads quality. As reported in [27
], variations in bread crumb cellular structures (in addition to moisture variations)—which are themselves strictly dependent from formulation and processing conditions—are key contributors to changes in the final bread eating quality.
Features of the products during storage
. Samples were also monitored under fixed times (23 h, 46 h, and 69 h) during storage under controlled conditions (Figure 3
). Breads were packed in hand-folded paper bags to mimic a domestic shelf life and to enhance samples performances.
Breads weight losses were more limited for SD230-bread and SD180-bread, in comparison to those containing CY (Figure 3
a), whereas the reductions in slice moisture values were quite evident (18%–27%) for all the samples (Figure 3
As already evidenced for the fresh products, S230-bread and SD180-bread were also characterized during the whole storage period by the highest values of Young’s modulus and load at 25% deformation, and—in contrast with the expectations—by faster hardening kinetics as well (Figure 3
c,d). These results did not seem to be directly related to the moisture levels of the various products; on the contrary, bread crumb structure (denser and heavier for SD230- and SD180-breads) seemed to be the main cause. At the opposite, CY180-bread and SDCY-bread were characterized by the lowest (and very close) values of Young’s modulus and load at 25% deformation during the whole storage period, as well as by the slowest hardening kinetics (Figure 3
c,d). However, CY180-bread exhibited a crumbly behavior just after 30 h of storage. This undesirable characteristic could be related to the lower tendency of CY180-bread to bind water during storage, also evidenced by its weight losses (Figure 3
a), thus confirming the findings of the previous work [27
]. In that study, in fact, what was thought to be “softness” was on the contrary a higher “fracturability” of the sample, which indicated a higher tendency to staling. However, when the two leavening agents were combined in the SDCY180-bread, a synergistic effect was observed, and the resulting bread—despite its intermediate weight losses—did not show a crumbly behavior. In this case, therefore, the softness was a “real” softness and not a consequence of a higher fracturability of the crumb.