Sucrose and leavening agent functionality in cake systems

Cakes are widely consumed bakery products which unfortunately contain ingredients which are associated with health-related issues, i.e. sucrose and, in many instances, phosphate-containing leavening agents. Based on the cake type, the ingredient bill further contains additional components. Sponge cake is made from flour, egg, sucrose, and leavening agent only, while cream and pound cake recipes contain either oil or margarine on top.

The functionality of sucrose in cake systems is relatively well documented. During batter mixing it dissolves in the aqueous phase [i.e. batter liquor (BL)]. It thereby increases batter viscosity and batter stability. During baking, it regulates cake crumb structure setting by impacting the temperatures at which starch gelatinizes and protein denatures. This then leads to cakes with optimal volume and crumb texture.

Similarly, the functionality of phosphate-containing leavening agent is also reasonably well understood. A leavening agent consists of a leavening acid and sodium bicarbonate (NaHCO3), which together provide carbon dioxide (CO2) during both the mixing and baking phase to benefit cake quality. The most common-place HXs all contain phosphate.

The aim of this doctoral work was to provide beyond the state of the art understanding of sucrose and leavening agent functionality during cake making in order to provide a science based basis for reducing or even replacing said components in cake recipes.

In a first experimental part, the focus was on sucrose functionality. The effects of sucrose, polyols (i.e. maltitol and mannitol), and dietary fibers (i.e. oligofructose and inulin)] on batter, BL, cake baking, and cake properties were compared. It was shown that, much as sucrose, the more optimal sucrose substitutes fully dissolve in the BL, resulting in similar BL quantities, BL viscosities, and proton mobility in the BL [as measured with time-domain proton nuclear magnetic resonance (TD 1H NMR)]. Optimal substitutes like sucrose result in appropriately delayed starch gelatinization and protein denaturation. While maltitol, oligofructose, and sucrose have similar effects on BL properties, oligofructose use results in too much delayed structure setting during baking. Mannitol and inulin have severe negative effects on BL properties due to their low solubility. Interestingly, the combination of oligofructose and mannitol is an optimal sucrose substitute in sponge, cream, and pound cake recipes, even if these components on their own are suboptimal sucrose substitutes.

A different approach to alleviate the negative consequences of reducing the sucrose content in cake systems was to alter the mixing atmosphere from air to pure nitrogen gas (N2) or CO2. The effect of different mixing atmospheres was found to depend on the solubility of the used gas in the aqueous (and lipid) phase(s). N2 is slightly less soluble in BL than air. As a result, its use causes the batter to be slightly more stable. Be as it may, the quality of sucrose-reduced sponge cake is slightly better when mixed under N2 atmosphere, while no such effect is detected for cream cake.

CO2 is significantly more soluble in BL than air. Its use results in batter which is more dense. However, because during baking the dissolved CO2 is released into the gas cells due to its solubility decreasing at increasing temperatures, more intense leavening occurs during baking. In sponge cake making, the negative effect of its use on foam density overshadows the above positive effect on leavening during baking, resulting in significantly worse cake quality. For cream cakes, its use has a limited negative effect on batter density, and the more intense leavening during baking overrules the negative effect during batter mixing. As a result, sucrose-reduced cream cakes the batter of which are mixed under CO2 atmosphere have significantly higher quality. Additionally, in the cake types studied, the use of CO2 atmosphere lowers batter pH, which facilitates protein network formation during baking and leads to improved cake texture.

In the second experimental part, the focus was on leavening agent functionality. In order to replace the inorganic phosphate-containing leavening acid sodium acid pyrophosphate (SAPP), organic replacers (i.e. adipic, fumaric, citric, and α-ketoglutaric acid) were tested. Their functionality depends on their solubility at room temperature and to what extent they dissolve during baking. Citric and α-ketoglutaric acid are very soluble at room temperature. They dissolve mostly during the batter mixing phase which qualifies them as early-acting HXs. As a result, only low amounts of CO2 can be formed and released during baking, resulting in low volume cakes. However, in this case, the small amount of CO2 release during baking favorably occurs before the crumb structure sets. Adipic and fumaric acid, just like SAPP, have low solubility at room temperature which qualifies them as late-acting HXs. As they release most of the CO2 from NaHCO3 during baking, their use leads to cakes of high volume and quality. However, the use of adipic acid causes part of the CO2 to be formed and released only after structure setting, resulting in less than optimal leavening during baking. Also, that some acidic hotspots are present in cakes prepared from adipic and fumaric acid containing recipes proves that they are not fully dissolved at the end of baking.

A different approach to remove inorganic phosphate-containing SAPP is to use an entirely different way of leavening; i.e. enzymatic leavening. The potential of using glutamic acid decarboxylase (GAD ST) enzyme alongside its substrate monosodium glutamate (MSG) and its cofactor pyridoxal 5’‑phosphate was explored in both pancakes and cream cakes. This part of the experimental work focused on testing whether this leavening system can be effective and, if so, on finding optimal conditions for using it. While the quantity of MSG affects the total amount of CO2 produced, the concentration of the enzyme GAD ST and its cofactor determine the rate of CO2 release. Additionally, the efficiency of the enzymatic system heavily depends on the batter pH and temperature. A pH (about 5.5) lower than the usual (pan)cake pH, a temperature around 55 °C, and a heating rate lower than typical heating rates during baking are needed for the enzyme to work optimally. Under such conditions, enzymatic leavening is at least equally effective as chemical leavening. It was further found that the enzyme activity does not depend on the sucrose concentration. Since standard cake making is not done under the conditions at which this enzymatic system performs optimally, more testing needs to be done before it can be readily used in cake systems.

Thus, based on this manuscript, we conclude that the future regarding both sucrose replacement and phosphate-containing leavening agent replacement looks promising. However, some more work needs to be done before they can be fully removed from cake recipe bills.