The role of lipids during sponge cake making and their impact on sponge cake quality

Sponge cake (SC) has an airy and soft texture. Examples of such cakes include Swiss rolls and layer cakes. They contain toppings (e.g. whipped cream) and/or fillings (e.g. fruit jam). SCs belong to the class of foam-type cakes. The latter differ from batter-type cakes (e.g. pound cake) because they do not contain lipids such as butter, shortening or margarine. The main SC batter (SCB) ingredients are wheat flour, eggs, sucrose and leavening agents. At industrial scale, exogenous lipids (ExLS, e.g. monoacylglycerols, diacylglycerols and polyglycerol esters of fatty acids) are included in the recipe to improve gas incorporation and gas cell stability.

Cakes are prepared by mixing the ingredients into a batter, baking the batter to form cake and cooling the cake to room temperature. Especially for foam-type cakes, gas cell incorporation during mixing and gas cell stability after mixing and during early baking are essential to obtain cakes with soft texture, fine crumb and high volume. Indeed, the above determine leavening and crumb structure since no additional gas cells are formed during baking. Gas cell incorporation and stability depend on the presence of surface-active components that can adsorb at the air-liquid (AL) interface surrounding the gas cells. These components are typically low molecular weight surfactants, which are mostly lipids or lipid-like molecules, or proteins.

Most available research on cake making deals with batter-type cakes and the role of their constituents during the different stages of cake making. As a result, the role of most batter constituents during baking and cooling and their impact on cake quality is well understood. However, before this work was carried out, studies investigating the role of surface-active batter constituents at AL and/or oil-liquid interfaces in cake batters were lacking. Especially in the case of foam-type cake systems, this is an interesting research topic as AL interface stability highly determines their quality. In the specific case of SCs, only a limited amount of publications have dealt with the role of batter constituents during its production in a fundamental way. Recent research reports mainly dealt with reducing/replacing ingredients (e.g. sucrose) or examining the impact of ExLS on SC(B) properties (e.g. batter density and cake volume). As today’s consumers perceive the latter as unnatural and unhealthy, there is a clear demand for food products with clean(er) labels [i.e. products containing no (or less) additives].

Against this background, this dissertation aims at unraveling lipid functionality during SC making. Lipids of interest are ExLS and those originating from flour and eggs (i.e. endogenous lipids).

In a first part, a multi-stage mixing method and recipe suitable for analyzing SC making from recipes either containing ExLS or not was developed. Application of this mixing method led to SCs of (i) acceptable quality when prepared from recipes not containing ExLS, and (ii) significantly higher quality when prepared from recipes containing ExLS. The latter were 37% softer, 28% larger and had a finer crumb than the former. Hence, SCs containing ExLS were of superior quality. Why ExLS have such positive impact on SC quality was investigated in the next part of this work.

The role of lipids at (i) the AL interface in SCB and in (ii) structure setting during baking was investigated by applying a novel (combination of) experimental approach(es). A foaming protocol was developed to study the molecular population at the AL interface in SCB. Furthermore, to investigate the impact of lipids on key phenomena responsible for structure setting during baking (i.e. starch gelatinization and protein polymerization) three experimental methods were combined. Differential scanning calorimetry (DSC) and size-exclusion high performance liquid chromatography (SE-HPLC) were used to monitor starch crystal melting and the extent of protein network formation during SC baking, respectively. Finally, time domain proton nuclear magnetic resonance spectroscopy (TD 1H NMR) was used in a temperature-controlled mode to in situ study changes in proton mobility during cake making. The TD 1H NMR results obtained were related to the rigidity of cake matrix. The role of lipids during each stage of SC making were eventually related to their impact on SC quality.

In a second part, the above approach was applied for SC recipes either containing ExLS or not. In absence of ExLS, the AL interface in SCB is stabilized by a mixed molecular population consisting of both proteins and lipids. Furthermore, it was shown that proteins, especially ovalbumin and also α- and γ-gliadins, have a more prominent role than the lipids at the AL interface in SCB. Ovalbumin molecules formed intermolecular disulfide bonds and α- and γ-gliadins had a higher affinity for the AL interface than other batter proteins. Surprisingly, flour lipids showed a higher affinity for the AL interface than egg lipids. Starch gelatinization and protein polymerization started at temperatures exceeding 67 °C and the setting of the cake structure continued during the isothermal phase of the heating process.

When ExLS were used in SC making, they dominated the AL interface. Interactions with endogenous surface-active components at the AL interface could however not be ruled out. While during baking differences were noted in matrix rigidity and the extent of protein network formation as a result of including ExLS in the recipe, these differences were no longer observed in cooled cakes.

Because endogenous surface-active lipids may play an important role at AL interfaces in SCB not containing ExLS, its recipe was used to investigate the role of flour and egg lipids during SC making in the remainder of the work. Moreover, these constituents do not have to be mentioned on the label and thus satisfy recent consumer’s demands. Flour and egg lipid functionality were investigated by lowering SCB lipid content or by relocating/releasing lipids prior to batter preparation. For the latter, flour or egg yolk (EY) lipids were removed from their native location by solvent treatment and afterwards added back to the flour or to the batter during mixing, respectively. Hence, lipid content of batter prepared with (i) flour with relocated/released lipids or (ii) EY with relocated/released lipids equaled that of the reference batter (i.e. SCB not containing ExLS). The role of flour and egg lipids at (i) the AL interfaces in SCB and in (ii) structure setting during baking was then examined as described above.

Wheat flour lipids only make up 1.2% of batter dry matter. In a third part, their impact on the quality of SCs not containing ExLS was investigated. Analyses of batter and cake properties prepared with flour in which lipids had been relocated or from which lipids had been removed revealed that free flour lipids negatively impact AL interface stability in SCB during mixing and early baking. They most likely disturb protein-protein interactions at the AL interface and inhibit the formation of a strong viscoelastic layer. Bound flour lipids are believed to remain associated with starch or gluten in SCB and therefore do not diffuse to the AL interface. Partial removal or relocation/release of flour lipids prior to SC making did not significantly impact cake structure setting nor further stiffening of the cake matrix during baking (and cooling).

The fourth part of this work dealt with the role of egg lipids during SC making from recipes not containing ExLS. The results showed that EY low density lipoproteins are very important for gas cell incorporation during mixing and gas cell stability during early baking. Indeed, when EY lipids were removed prior to SCB making, cakes with very low volume and firm texture were obtained. Removing or relocating/releasing EY lipids prior to SC making did not significantly impact key phenomena responsible for cake structure setting during baking.

In summary, the present doctoral work provides an in-depth overview on the role of surface-active components in SCB. It has not only generated insights on the role of exogenous and endogenous lipids during SC making, it has also increased knowledge on the role of surface-active proteins in SCB. It is hoped that the generated insights can serve as an excellent basis for developing SCs with cleaner labels.