3.1 Dough characteristics
3.1.1 Texture: sphere penetration measurement
Results from the sphere penetration test are shown in Table 1. All doughs were significantly different (p < 0.05) to one another. Some of the main factors to consider when evaluating the functionality of the fat in a biscuit dough are the ratio of the solid phase to the liquid phase and the crystal structure of the solid fats 5; the effect of different fats in the overall dough texture is determined by the consistency of the fat and its content in solid fats 21. Among the fat systems used in this study, EVOO had the lowest solid fat content thus producing the softest (p < 0.05) dough. It has been observed previously that the use of highly unsaturated fats in biscuits gave place to doughs with softer texture 18,42.
The replacement of butter with the CNE gave place to dough pieces (D-CNE) with significantly higher (p < 0.05) hardness values than the D-control. The reduction of the fat content in the system may have hindered some of the main functionalities of fat in a dough system, lubrication and aeration 5. During mixing fat surrounds sugar and flour particles, breaking the continuity of the protein-starch matrix and reducing gluten development 5,43. When less fat is present in the system water has easier access to flour proteins, hydrating them and creating a cohesive and extensive matrix 5, thus giving place to harder doughs. When the ingredients of the emulsion were added individually during the mixing process the hardest (p < 0.05) dough was obtained (D-INE). When adding the HMPC as a dry ingredient it may have not formed an entangled polymer solution within the dough matrix due to limited time and shearing forces during mixing. The water added for the formation of the HMPC solution was freely available to interact with other ingredients, such as flour particles, thus higher dough hydration over dough shortening effect could have taken place in D-INE.
Table 1
Hardness and oil migration of biscuit doughs (D-). Control is the dough made with butter; in D-EVOO, 33% of the butter was replaced with Extra Virgin Olive Oil; D-CNE has 30% less saturated fat by replacing butter with a Complex Nano-Emulsion; INE has 30% less saturated fat by replacing butter with the individual ingredients of the complex nanoemulsion.
Formulations | Hardness (N) | Oil migration (%) |
18°C | 30°C |
D-Control | 2.42 c (0.1) | 5.35b (0.66) | 9.11c (0.73) |
D-EVOO | 0.74 d (0.05) | 12.69a (1.18) | 19.19a (1.53) |
D-CNE | 2.88 b (0.15) | 6.51b (1.58) | 10.99b (1.45) |
D-INE | 3.44 a(0.52) | 5.88b (2.54) | 11.43b (2.17) |
Values are reported as means. Values in parentheses are the standard deviation. Means in the same column without a common letter are significantly different (p < 0.05). |
3.1.2 Oil migration
When the fat consistency is very soft or liquid at typical dough manufacturing temperatures, it can cause problems of oiling out 44. Oil migration indicates that there is not strong interactions between the fat phase and all the other ingredients 45. The migration of the fat phase outside the surface of the dough was examined and results are shown in Table 1. D- Control, D-CNE and D-INE showed no significant differences (p > 0.05) in oil migration at 18°C. However, D-EVOO showed the highest oil migration from its surface (p < 0.05); more than double the quantity of oil migrated from D-Control. These results were expected due to the use of liquid fat to replace of butter without using any other structuring strategy that could give more stability to the liquid fat in the dough. In fact, the addition of HPMC and lecithin in the dough formulation, as a pre-prepared emulsion or as individual ingredients, stabilised successfully the oil in the biscuit doughs. When adding the CNE, the oil was emulsified by the combination of lecithin and HMPC in nanodrops in a continuous aqueous phase of entangle HMPC, which confirmed by the confocal images as shown in Fig. 2. Confocal micrographs of biscuit doughs emulsions made of sunflower oil and HMPC as shortening replacer showed that the oil was dispersed homogeneously in a continuous phase of proteins, carbohydrates and dispersed starch granules 23. However, the efficacy of the emulsifier mix in the stabilisation of the oil in the dough matrix was significantly (p < 0.05) decreased at higher storage temperature (30°C). This could be because there was a decrease in the viscoelasticity of the HMPC solutions or emulsions due to a sort of a thermal softening effect in the inter and intra-molecular hydrogen bonds between HPMC molecules 46. It has also been reported that at temperatures from 30°C to 90°C the hydrophobic headgroup of lecithin molecules changed its shape inducing oil droplet coalescence 47. These destabilisation mechanisms, together with the increase in volume and movement of the fat phase at higher temperatures could have led to oil droplet aggregation and diffusion to the surface of the biscuit dough.
3.2 Biscuit characteristics
3.2.1 Weight loss during baking (WL), moisture and water activity (aw)
Weight loss, moisture and water activity values of the biscuits are presented in Table 2. B-EVOO lost significantly more weight during baking (p < 0.05) giving place to a final biscuit with lower moisture and aw values. During baking water is available to solubilise sugars and gives place to a rich fatty sugary liquid viscous enough to contain the water vapour at initial stages of baking 3. Then, the viscosity of the continuous network increases and the protein film formed is set, defining the width of the biscuit; then gases are released (carbon dioxide and water vapor) and towards the end of the baking the height of the structure collapses 3,5. The effects of replacing butter by EVOO in WL and moisture could be explained by the distribution of the oil in the dough and its effects on water distribution and availability during baking. During mixing oil gets dispersed as fine droplets that are significantly less effective in imparting the shortening functionality than plastic fats do 21. When fat is badly distributed in the dough, flour particles are more readily available to be hydrated 21. It is then hypothesised that when water is interacting with flour particles in the dough instead of forming the syrupy phase the rate of evaporation during baking is higher, as it has been shown for B-EVOO.
The incorporation of emulsifiers, such as lecithin and HMPC helped in the retention of water during baking, giving place to final biscuits with similar WL, moisture and aw than B-Control. When HPMC is dispersed in water hydrogen bonding between the hydroxyl groups in HPMC and water are formed 48,49. Then during baking, there is an increase of strong hydrophobic interactions between HPMC chains that gave place to a sol-gel transition 50,51, which will have retained more water within the biscuit systems (B-CNE and B-INE).
Table 2
Weight loss during baking (WL), moisture, water activity (aw), spreadability index (SI), hardness and fracturability of the biscuits (B-). B-Control is the biscuit made with butter; in B-EVOO, 33% of the butter was replaced with Extra Virgin Olive Oil; B-CNE has 30% less saturated fat by replacing butter with a Complex Nano-Emulsion; B-INE has 30% less saturated fat by replacing butter with the individual ingredients of the complex nanoemulsion.
Biscuits | WL (%) | Moisture (%) | aw | SI (mm) | Fracture strength (N) | Fracturability (mm) |
B-Control | 13.07b (0.58) | 4.14ab (0.38) | 0.49a (0.04) | 3.32b (0.04) | 27.31c (8.83) | 38.00a (0.21) |
B-EVOO | 15.33a (0.77) | 3.48b (0.53) | 0.36b (0.03) | 3.73a (0.16) | 18.37d (5.53) | 37.24b (0.32) |
B-CNE | 13.20b (1.19) | 4.78a (1.01) | 0.51a (0.08) | 3.31 b (0.14) | 43.33b (10.38) | 38.32a (0.31) |
B-INE | 13.23b (0.97) | 4.99a (0.89) | 0.50a (0.06) | 3.21 b (0.14) | 50.86a (16.47) | 36.57c (1.07) |
Values are reported as means. Values in parentheses are the standard deviation. Means in the same column without a common letter are significantly different (p < 0.05). |
3.2.2 Spreadability Index (SI)
Biscuit quality can be summarised in two general aspects: the size and the bite 5. Pareyt and Delcour 5 and Manley 3 explained that final biscuit dimensions depend on several factors; the dough spread onset which can be explained by an increase in the mobility of gluten proteins as temperature increases and is influence by the level of plasticiser (water) or anti-plasticiser (syrup); the dough spread rate during baking, which is defined by the levels of dissolved sugar and melted fat, or in other words it is defined by dough viscosity; and the set time which is defined by an ‘apparent’ glass transition of the protein network, increasing its viscosity and stopping cooking spreading. The height of the biscuit continues to increase due to gases formation until the structure collapses at the end of the baking process. The appearance of four formulation of biscuits was showed in Fig. 1.
Results of the SI of biscuits are presented in Table 2. B-Control, B-CNE and B-INE showed no statistical differences (p > 0.05) in terms of spreadability. On one hand, it could have been hypothesised that the fat reduction would have decreased the spread rate and the addition of water would have decreased the anti-plasticiser effect giving place to smaller biscuits. Previous work where the shortening was replaced by hydrocolloids (polydextrose, maltodextrin, inulin or whey proteins) in biscuits showed that when replacing fat (and sugars) in biscuits the diameter of the final products decreases due to elastic shrinkage tendency of these samples during baking 41,52. On the other hand, the results from this study showed that the functionality of the CNE and its ingredients had a different effect on biscuit SI than other emulsions with HPMC. Previous works in which oil in water emulsion with HPMC or ethyl cellulose were used to replace shortening in short dough biscuits observed a later set time for biscuits made with the emulsions 53; and reported higher biscuit diameter, thickness and SI values due to a softer and more fluid dough behaviour during baking 24. However, the emulsions from the aforementioned studies, although using HMPC with similar degree of substitutions than the HPMC used in this study, contained higher amounts of oil (47%-52%). The lower oil content (10%) and the combination of emulsifiers (HMPC and lecithin) used in this study to formulate the emulsions gave place to similar spreadability properties as butter, and as a consequence the strategy used in this study has given place to a similar spreadability behaviour in biscuits made with the nanoemulsion (B-CNE) than the ones with butter (B-control).
As expected, B-EVOO biscuit showed the highest (p < 0.05) SI. The higher content of liquid fat gave place to a softer dough with higher spreadability rate, higher final diameter and lower height. These results were in agreement with other studies, which presented that higher fraction of unsaturated fats lead to a higher spreadability index 21,54; the degree of saturation affects physical and thermal properties in fatty acids, specifically unsaturations decrease the fat melting point. This characteristic of the oil decreases the viscosity of the dough, increasing the spreadability on set and rate.
3.2.3 Texture
During the baking process the viscoelastic dough changes into a solid with an aerated cellular structure and a characteristic texture 55. Fracture strength and fracturability values of the biscuits are presented in Table 2. The results suggested that dough mechanical properties (Table 1) defined biscuits’ fracturability strength. B-CNE and B-INE showed significantly higher (p < 0.05) fracture strength than B-Control; and B-INE showed the highest fracturability (p < 0.05) among biscuit samples. These results could be explained by the fact that when less fat is present in a dough formulation less shortening effect is achieved, so flour particles are more accessible to water, giving place to harder and brittle biscuits 3. When comparing B-CNE showed similar (p > 0.05) fracturability to B-control, suggesting that this fat replacer offers the potential to be used as butter replacer for specific bakery applications. Biscuits in which 33% of butter was replaced by EVOO presented the lowest (p < 0.05) fracture strength values due to higher content of total fat and the lowest consistency of the oil at ambient temperature in comparison to butter or the emulsion.
3.4 Sensory profile of biscuits
The trained panel developed a consensus vocabulary of 28 attributes in 5 main modalities including appearance, aroma, flavour, mouthfeel and aftereffect as shown in (Table S.1). The results showed that 21 out of 28 attributes were found to differ significantly (p > 0.05) between samples. The sensory profiles of biscuits are presented in Fig. 4. As expected, due to their similar formulations, QDA evaluation identified that B-CNE and B-INE had more similar sensory attributes to B-Control than B-EVOO. There were no significant (p > 0.05) differences among the samples in terms of floury (aroma and taste), savoury (taste), dryness (mouthfeel), sweet (taste), greasy (aftereffect) and numbing/ cooling (aftereffect).
In terms of the appearance (Fig. 4A), B-EVOO presented a significantly lower crumb density that the other three formulations and a significantly higher surface smoothness than B-Control. On the other hand, B-CNE and B-INE showed a similarly surface smooth, dark specks and crumb density as B-Control. The presence of HPMC in the formular could be a factor to improve crumb density of B-CNE and B-INE because HPMC could develop a more stable system by keeping air bubbles during dough preparation and baking. This is in agreement with Bousquières, et al. 56, who stated that HPMC controlled the viscosity of cake batter and limited bubble loss during batter preparation. This paper also reported that during baking, methylcellulose (MC) and HPMC governed the sol-gel transition by forming hydrophobic bonds and at the same time there was a starch gelatinization, which affected on a wide range of cellular homogeneities in crumb. Moreover, B-CNE exhibited a significantly higher shiny appearance than B-Control and B-EVOO. This could be due to of the formation of a shiny HPMC layer coating the biscuit. HPMC is a widely used edible coating component to enhancethe glossy appearance of fruit skins 57,58.
When replacing butter in baked goods, it is important to consider several sensory properties of the final product: the aroma and flavour of butter; Maillard derived compounds and their sensory properties; the mouthfeel sensations from both the butter and structural changes in the baked good; the flavour release when changing the food matrix; and any new attributes introduced with the replacement ingredient.
With regards to the aroma (Fig. 4B), a decrease in buttery, sugary and baked aroma was found in B-EVOO compared to B-Control, which is to be expected due to the reduction aroma compounds associated with butter and those derived from the Maillard reaction linked to baking and sugary aromas. However, there were no significant differences across aroma attributes between B-Control and B-CNE, indicating that aroma perception of B-CNE matched the same levels of B-Control. Although there was a reduction of butter in biscuits, there were no significant differences found for buttery, sugary, floury and baked aromas in B-CNE when compared to B-Control.
However, when assessing the taste and flavour (Fig. 4C), restuls suggested that a reduction of butter in biscuits by replacing with EVOO and CNE leads to a lower perception of buttery taste and flavour. While B-CNE and B-INE contained the same ingredients, the processing of those ingredients into a complex nanoemulsion did have an effect on some of the sensory properties of the biscuits. For example, where B-CNE did not vary from the control in sugary aroma, B-INE had a significantly lower intensity of sugary aroma (Fig. 4B).
In addition, B-EVOO had significantly higher fatty/oily flavour than the other three formulations, expected due to. higher amounts of oil in the composition leading to more aroma compounds associated with fatty/oily. B-Control was also found to be significantly more salty than B-EVOO. This result agrees with previous findings of Paneras, et al. 59, Romeih, et al. 60 and Chabanet, et al. 61, who observed greater saltiness intensity with a lower fat content in food products such as sausages, frankfurters and cheese, due to matrix effects.
Significant differences between samples were found across almost all mouthfeel attributes (Fig. 4D). Many of these differences found B-CNE and B-INE to be similar, with largest differences to B-EVOO. A significant decrease (p < 0.05) in firmness at first bite was found in the B- EVOO, which was in accordance with the instrumental texture results on fracture strength (Table 2). B-EVOO scored significantly higher for crumbly than all other samples; B-CNE and B-INE both scored significantly lower in crumbly than B-EVOO and B-Control. This indicates that B-CNE and B-INE were more compact and with a stronger continuous phase breaking into fewer crumbs during first bite. The prescence of HPMC in the formula of B-CNE and B-INE could contribute to the similar bite to B-control and the lower crumbliness. This is supported by previous literature 23, which reported that the microstructure of biscuits made with cellulose showed a dense matrix of stable fat globules, corresponding to the dispersed phase or oil globules, immersed in a continuous phase made up of water hydrated cellulose (a three-dimensional network of HPMC chains). Moreover, pastiness of biscuits significantly increased in B-CNE and B-INE, compared to B-Control. This increase in pastinesscould be due to the tendency of HPMC to interact with water; when breaking the biscuits into smaller pieces during mastication, the continuous HPMC network is increasingly exposed to saliva, leading to a higher water absorption rate on the crumb pieces. This hypothesis could also explain the significantly higher tooth packing aftereffect (Fig. 4E) of B-CNE samples in comparison to other biscuits.
Overall, the sensory data showed that both B-CNE and B-INE were similar to B-control in many attributes. However, there were significant differences in shiny appearance, butter taste and flavour, and tooth packing between B-CNE and B-control, and significant differences in sugary aroma and gritty mouthfeel between B-INE and B-Control. Moreover, the finding of this study suggests that HPMC plays an important role on improving crumb density (appearance and mouthfeel) and bite of the biscuits. However, pasty and tooth packing of biscuits could be affected by HPMC, whereas EVOO could be a cause of fatty/ oils in biscuits.