1. Introduction
The demand for eggs and products derived from eggs in Europe is expected to grow by 20% until 2030 [
1]. However, since the last decade, consumers’ behaviour has changed regarding health, well-being, and environmental concerns [
2]. Several issues are directly related to the conventional production of eggs. For instance, there has been a higher awareness of animal welfare standards since banning cage systems in 2012 [
3,
4]. Another point is correlated with eggs being produced from animals, hence the occurrence of food safety scandals such as the Fipronil scandal in the European Union in 2017 [
5] or the salmonella outbreak [
6] in the United States in 2015. Also, high fluctuations in the prices of eggs due to outbreaks of avian influenza, causing bird losses, pose a problem for the industry [
7]. Furthermore, environmental issues are posed by the egg industry, being responsible for about 10% of total livestock emissions [
8]. As a result, there is a growing demand for egg alternatives originating from plants that replicate similar functionalities [
9]. The global market for egg alternatives accounted for USD 1.5 billion in 2021, with an expected annual growth rate of 8.3% through 2031 [
10]. From 2016 to 2021, 102 egg replacement products were launched globally [
2]. There are various applications for egg alternatives, including mayonnaise, biscuits and cookies, noodles and pasta, products to simulate egg dishes (e.g., scrambled eggs), chocolates, and cakes [
2].
To simulate eggs’ physicochemical, techno-functional and sensory attributes, it is essential to understand their complex composition and structure [
11]. Eggs are large single cells that contain all important materials to contribute to the initial development of the embryo. Therefore, an egg is enriched with the necessary proteins, lipids, carbohydrates, vitamins, and minerals [
11,
12]. In summary, a whole hen’s egg contains 75% water, 12% protein, 12% lipids, and smaller amounts of carbohydrates, vitamins, and minerals [
13]. Pound cake is a complex food system containing mainly eggs in addition to wheat flour, sugar, margarine, or butter, added in equal portions [
14,
15]. Broken down, the cake batter is divided into a continuous aqueous phase with dissolved sugars and proteins, hydrated flour particles, and margarine or butter emulsified in the liquid phase. The incorporation of air creates foam, consisting of air bubbles and small water cells, which are stabilised by the fat phase of the system [
16]. Forming a stable protein network contributes to the required properties of pound cake. Egg protein is mainly responsible for the end product’s characteristics, such as microstructure, appearance, texture, and sensory attributes [
17]. However, the batter preparation does not contribute to the protein network formation significantly. One reason is the presence of fat and the high sugar contents that restrict the gluten network development during mixing [
18]. In contrast to the batter preparation, multiple covalent and SS bonds are formed during the baking process. Therefore, it is hypothesised that different proteins from the various ingredients in cake (flour and egg) interact with each other during baking, whereas ovalbumin is the most crucial factor regarding the development of SS and covalent bonds [
14,
19]. Furthermore, SS bonds are formed with protein from egg yolk and wheat flour through SH–SS exchange reactions between 78 and 84 °C [
20]. Following higher temperatures (between 84 and 90 °C), the denaturation of ovalbumin allows the incorporation of other proteins from wheat flour, such as α- and γ-gliadin [
21]. In addition, the LDL fraction of the egg yolk plasma is a strong emulsifier. The fraction contains proteins and phospholipids. These amphiphilic molecules contribute to surface activity [
12]. Adding eggs to pound cake is vital in forming a stable protein network between all available protein sources, guaranteeing adequate cake quality.
Multiple studies focused on the replacement of eggs in bakery products. The egg alternatives, based on soy, wheat, whey, or hydrocolloids, had an inferior effect on the products and were not comparable to the egg-containing control cakes [
22,
23]. Furthermore, the performance of soy milk as a potential egg replacer was evaluated [
24]. The addition of the emulsifier distilled glycerol monostearate led to promising results whereas adding lecithin resulted in a dark-coloured cake with a firm texture and low volume and sensory scores [
25,
26]. Other studies evaluated the influence of emulsifiers on whey protein as a possible egg replacement. Both could determine a positive effect of the added emulsifiers, although the final products did not match the quality of the control cakes [
27,
28]. Overall positive effects could be achieved by using lentil protein as a replacer in muffins and angel cake, excluding the hardness and chewiness parameter, which increased over storage [
29]. Agrahar-Murugkar et al. (2016) could identify banana as an excellent egg replacer whereas adding chia seeds was less accepted by consumers [
30]. However, chia seeds increase the protein, fat, and mineral content [
31]. Aslan and Ertas (2020) added chickpea aquafaba in different concentrations [
32]. The maximum addition of 50% aquafaba to the cakes had no effect on the cake quality.
The objective of this study was to evaluate a selection of egg replacers currently on the market in a pound cake system. Egg replacers were chosen based on their local availability while also selecting products that ranged in protein content in order to obtain samples representative of the egg replacer market. A variety of quality criteria were applied to determine their suitability in comparison to control cakes. It is hypothesised that replacing eggs with commercial egg replacers impacts the quality attributes of the baked good in comparison to the egg-based controls.
4. Discussion
This study revealed the decisive function of eggs in a pound cake system and the effects of commercially available egg replacers on batter and cake quality, as well as on nutritional value.
A recent survey on the current market situation stated that manufacturers face challenges developing plant-based eggs, making most of the available products purely starch-based. In conclusion, most of the problems lie within replicating the functionalities and taste of conventional eggs, the high ingredients costs, and the limited capacity of start-ups to scale the product [
37]. Starch-based egg replacers fail to replicate the high nutritional value of eggs. In contrast, applying plant-protein-based substitutes (soy, whey, lentil) results in a low consumer acceptance, considering the low volume of the cakes, inappropriate texture, and sensory scores [
22,
23,
24,
25,
26]. Promising results were found upon applying lentil protein as an egg replacement in muffins and angel cake, with minor concerns about the texture of those bakery products [
29]. Other studies claimed chickpea aquafabe as a fitting egg replacement. Treatment with citric acid and table salt resulted in a remarkable similarity to baked products containing egg whites [
38]. Another study added lentil protein to a mixture of aquafabe and citric acid and achieved acceptable cakes [
39]. A potential approach for the substitution of eggs in baked goods involves the usage of plant milk. In a recent study, 50% of eggs were successfully replaced with soy or lupin milk [
40]. Europe’s trending vegan food market [
2] contributes to multiple European countries selling and developing vegan egg replacements such as the UK, Spain, Austria, and Poland. Product claims on egg replacer packaging serve as advertisements for the advantages of using plant-based egg alternatives such as their lower environmental impact, considering that many consumers still perceive conventional eggs as a good option [
37]. Additionally, egg replacers are considered healthy by recent surveys, providing essential minerals, vitamins, and a high amount of fibre [
41]. However, most of the analysed egg replacers display a poor nutritional value not comparable to eggs, with only 10% of the available products reaching a high in protein claim (according to EU, Regulation (EC) No 1924/2006; and Canadian regulations, B.01.513 Food and Drug Regulations (FDR)). Moreover, the protein quality of egg replacers is often inferior to that of eggs as these replacers contain pulses that contain low amounts of sulphur amino acids. To overcome this problems, adding cereal proteins containing natural sources of sulphur amino acids and hydrocolloids is suggested [
12].
The quality of a cake is directly related to the properties of the cake batter. In
Figure 5, the different attributes of the cake samples are plotted in a principal component analysis (PCA). In the PCA, a direct correlation occurs between the analysed parameters of cell elongation and the number of holes. Correlation analysis reveals a direct correlation between the viscosity of the cake batter and the number of holes (
p-value: 0.000843; r-value: 0.80). Hence, the batter viscosity influenced the expansion and stability of air bubbles in the batter throughout the baking process [
28]. In less viscous cake batter, the carbon dioxide can evolve, releasing the water vapour from the system. Comparatively, highly viscous cake batters retain more air bubbles, leading to more holes [
28]. On the other hand, the PCA displays no positive or negative correlation between the viscosity and the texture properties, resulting in a limited influence of the cake batter’s viscosity on the overall textural quality of the product. It has been reported that hydrocolloids prevent structural collapse by increasing the viscosity of the batter, particularly xanthan gum, a compound whose high molecular weight correlates with the radius of gyration and water hydration capacity, leading to a high batter viscosity [
28,
42,
43,
44]. A clear trend involving the usage of hydrocolloids and the occurrence of higher viscosity is not visible in the results. However, it is possible that the application of R4 resulted in the most viscous batter due to the included xanthan gum. Contradictory to this, the other egg replacers containing xanthan gum (R2, R9) are characterised by a low viscous batter, even with the presence of salt-based ingredients (cream of tartar, rock salt, black salt) that are known to promote the association and gelation properties of xanthan gum [
45]. A direct correlation was observed between the water content and the viscosity (
p-value: 0.00809; r-value: 0.77), resulting in the assumption that the water level has a stronger influence on the viscosity than the included hydrocolloids. For better comparability and the estimation of the effects of the ingredients, the same amount of water should be added instead of following the supplier’s instructions.
Additionally, the structural changes during the baking process of the cake batter were analysed. The lowest structure setting temperatures occurred in the control cakes, especially with fresh eggs (C3). This can be explained through the cake structure formation. During baking, the structure sets due to the gelatinisation of starches and egg protein coagulation [
15]. Ovalbumin, the most abundant protein in egg white, completes the heat setting of the cake batter to a solid cake foam at 85 °C [
15]. Comparable results for the egg replacers were only achieved by incorporating R1, R8, and R10, implying higher structure-setting temperatures for the remaining egg replacers due to the lack of egg proteins. This hypothesis is supported by the fact that plant proteins, such as pea protein isolate (88.9–94.5 °C), used in R4, R5, and R9, have a higher general denaturation temperature than egg proteins [
46]. Hence, improving the quality of egg-replaced cakes could be achieved by lengthening the baking time or increasing the baking temperature. Another factor that influences the ability of structure setting during the heating process is the presence of hydrocolloids. The highest determined temperatures for the transition to a solid cake structure can be found in cakes including gelling agents like xanthan gum (R4, R6, R9). For the gelatinisation of starch, the water in the pound cake system is required, resulting in the swelling of starch granules [
47]. The degree of gelatinisation is determined through the availability of water [
14]. Since xanthan gum hydrates in cold solutions, water availability could be limited in egg replacers including this hydrocolloid [
44]. In contrast, egg replacements, including the hydrocolloid methylcellulose (R1) and locust bean gum (R8), without additional xanthan gum, displayed lower structure-setting temperatures in this study. One explanation could be that locust bean gum requires heat to be soluble in water, and the gelation mechanism of methylcellulose is caused by the hydrophobic-induced association of methyl groups in methylcellulose at high temperatures. Therefore, they cannot absorb water during the mixing process, displaying no direct competition for the starch granules during cold temperatures, leaving enough available water for the baking process [
48].
Nevertheless, the parameter specific volume displays the most noticeable difference when comparing the control pound cakes to the egg replacer cakes. The interaction of egg proteins and the gluten network form a strong thermal gel that strengthens the cake structure during baking [
15,
49,
50]. Between the different control cakes C1, C2, and C3, significant differences in the specific volume are visible. This can be explained by the influence of the spray drying process used to manufacture whole egg powder (C1). High temperatures and pressure conditions during the drying process result in the denaturisation of the albumin fraction of the egg and changes in the protein conformation, positively affecting the emulsion stability and gel-forming capacity after heating, hence providing a higher specific volume in the cake [
51]. The egg replacers achieving the highest specific volume are R4 and R3, containing no or minimal protein contents. The high specific volume of R4 is due to mono- and diglycerides, which delay starch gelatinisation and structure settings in cakes and allow for longer rising times [
43]. Moreover, the formation of amylose–lipid complexes between certain emulsifiers and the amylose in wheat flour starch influences the starch gelatinisation during baking and results in a soft crumb, which correlates negatively with the specific volume (
p-value: 0.0014; r-value: −0.68) [
52]. R3 resulted in a high specific volume, most likely due to containing baking soda and psyllium husk fibre. Psyllium husk fibre is a highly branched acidic arabinoxylan with a high molecular weight (~1500 kDa), used as a stabiliser and gel former in bakery products [
53]. However, arabinoxylan can beneficially influence the cake’s quality by increasing the batter’s viscosity, improving the water binding capacity and stabilising the expansion of gas cells during the baking process, resulting in a higher specific volume. Contradictory, it has been found that the increase in the batter viscosity above a critical level restricts the expansion of gas cells, directly relating to small volumes in cakes [
54]. This could also relate to the collapsed structure that most of the egg replacement cakes display. Another explanation for the collapsed structure of the egg replacer cakes could be the restricted starch gelatinisation due to the competition for water by high-fibre ingredients, additional starches, and hydrocolloids [
55]. The lowest specific volume was found in egg replacers, containing the hydrocolloids guar gum, locust bean gum, and xanthan gum (R2, R5, R8, R9), potentially due to the high molecular weights of the hydrocolloids creating overly strong interactions and competing with the present water in the system, thus preventing the formation of a stable cake network [
44,
56]. Horstmann et al. (2018) could determine similar results by applying those hydrocolloids in a bread system. It was hypothesised that the high molecular weights of the hydrocolloids create overly strong interactions, subsequently competing over the present water in the system that prevents the formation of a stable cake network.
Similar to other published studies, a negative correlation between specific volume and hardness of the cake was observed (
p-value: 0.0014; r-value: −0.68). Cakes with a softer crumb structure mainly contained starch-based egg replacers, with some containing psyllium, relating to the previously stated theory that adding arabinoxylans improves cake quality. In contrast, no clear trend was discernible for the cakes with the hardest crumb structure (R6, R2, R5), which were equally and mainly starch-based. Hence, it can be assumed that the composition of the applied starch-based egg replacers provides optimal cake quality. The egg replacers displaying a comparable hardness value to that of C1 contain xanthan gum (R4, R9). This effect can be linked to the negatively charged surface of xanthan gum, creating repelling forces that restrict the starch granules from swelling further and retard the leaching of amylose, resulting in reduced retrograded amylose in the cake [
44]. Accordingly, choosing the right hydrocolloid and protein source to simulate the textural properties of egg-based cakes is one of the general concerns for all upcoming egg replacer developments. Chewiness is a measure of the energy required for chewing semi-solid foods and can often be compared to the hardness parameter in egg-replaced cakes [
25,
30]. Comparable results to the egg powder cake (C1) were achieved by R5, R6, and R10. Similarities in the composition of the egg replacements lie within the addition of hydrocolloids in R5 and R6. On the other hand, R10 contains Chia as a natural hydrocolloid, and this enhances the cake’s structure by inhibiting the kinetics of amylopectin retrogradation and the mucilage, influencing the water holding capacity and the protein network formation [
57]. Furthermore, Chia influences the texture attributes cohesiveness, springiness, and resilience, which codetermine the quality characteristics of pound cake. Those properties are indicators of the internal resistance of food to compression and the development of a three-dimensional protein network [
27]. None of the egg replacer cakes correspond to the cohesiveness values of the control cakes, implying the possibility of product damage during the production process and the formation of an unstable network [
22]. A high springiness value is desired for cakes and indicates a fresh, aerated, and elastic end product [
58]. This attribute was achieved by R4 and R10, both containing legumes as protein sources. Legume proteins form gels through electrostatic interactions, hydrophobic interactions, and hydrogen bonds whereas disulfide bonds do not contribute to forming a gel-like structure [
59]. As a result, the electrostatic interactions significantly positively influence texture properties like hardness and, notably, springiness [
59].
Extremely high and shallow bake loss values are typical for exceptionally moist and dry cakes, respectively, while both extremes can have a significant impact on the quality of the product [
22]. The egg replacer R3, having the highest bake loss, contains psyllium husk fibre. Geera et al. (2011) could determine a fibre- and gum-based egg replacer as relatively incapable of binding water during the heating process, compared with other egg replacers and 100% egg formulations [
22]. For the egg replacement R4, it could be argued that the included starch and hydrocolloid (xanthan gum), combined with less water added to the recipe, resulted in the lowest bake loss due to the higher water holding capacity [
60]. Another egg replacement displaying a significantly low bake loss is R10, containing chia. Chia gel was found to have a high water holding capacity compared to hydrocolloids like guar gum and other protein isolates (lupine, cowpea, winged bean) due to the high amount of fibre and protein within the ingredient [
61].
The overall appearance of the final cake products plays a significant role in meeting the consumer’s needs and increasing product acceptance.
Figure 2 shows a major difference between the control and egg replacement cakes. Additionally, differences in the colour of the egg replacer products are visible and calculated using the Scofield equation (
Table 6). The crust colour is formed through non-enzymatic chemical reactions involving the Maillard reaction and caramelisation, reactions that are highly influenced by pH and water activity (a
w [
15,
49,
50]. Regarding the a
w results, there are no significant differences between the egg replacers and the control cakes, except for R2 displaying a low a
w. Those results were comparable with the crust colour of C1. It was also reported that a higher protein content promotes browning through the Maillard reaction [
26]. However, this observation seems to be only partially true, as the high-protein egg replacers differ more from the control cake than the no- and low-protein egg replacers. The crust colour primarily depended on the colour of the products, with the only significant differences pronounced in R3 and R10 (
Figure 2,
Table 6). Both egg replacers contain high-fibre ingredients like psyllium husk and chia, with reports claiming that adding chia to pound cakes results in a darker product colour due to the darker colour of the ingredient [
62]. This effect is particularly striking when looking at the results of the crumb colour analysis, with the most comparable replacer being R9. The added turmeric powder provides a yellow colour to the egg replacer powder, tinting the cake batter and the resulting product after baking. Accordingly, turmeric powder might be a suitable alternative to replicate the orange–yellowish colour of the egg yolk.
The cake crumb characterisation was accomplished through image analysis. Except for the slice area and the number of cells, which are directly related to the cake’s specific volume, no significant differences seem to appear. The more air is incorporated into the batter during mixing and held in the batter by functional ingredients, the higher is the number of final cells observed in the cell structure. During the baking process, water evaporates, and the vapour is entrapped in the already existing air cells. The more air cells are present in the batter, the higher the amount of water vapour being entrapped is, which increases the internal pressure, resulting in a leavening effect (supported by the reaction of baking powder). The cell diameter indicates, firstly, how resistant the cake batter is towards extension. A high resistance limits the expansion of the batter and hence causes a small diameter (R2, R4). Secondly, the cell diameter is influenced by the stretchability of the batter, which avoids the collapse of air cells leading to holes [
63]. It is worth noting that egg replacer R4 is the most comparable to the control cakes without considering cell elongation. The parameter cell elongation is directly related to the viscosity of the cake batter (
p-value:0.000472; r-value: 0.75). A cell elongation of 1 represents a circle structure while values higher or lower than 1 indicate an elongation of the cells. All cakes showed cells vertically elongated, a phenomenon caused by the internal pressure in the batter during leavening in the oven. An ideal round shape of the cells can only be achieved if the internal pressure is equally released in all dimensions. Due to the tin walls, all cakes rise vertically during baking, causing a cell elongation [
63]. This explains the higher cell elongation in R4 and R10, where the high viscosity of the cake batter restricted the even expansion of the cake cells, resulting in elongated cells that were less comparable to the control cakes C1 and C2.
5. Conclusions
Recent shifts in consumer behaviour are being caused by a growing concern for health, well-being, animal welfare standards, and the environment. As a response to these changing trends and consumer demands, manufacturers have brought 20 different egg replacers to the market, suitable for baking applications. Different approaches regarding the compositions of the egg replacers were used to face the challenges that arise with replacing eggs in a pound cake system. The ten analysed egg replacements were sorted into three different categories, displaying egg replacements with no protein, a low amount of protein (1–10 g/100 g), and a high amount of protein (>10 g/100 g) and were compared to three different control cakes including powdered whole egg, fresh egg, and liquid whole egg. Applying a low-protein egg replacer, consisting of a corn-starch-based alternative, with the addition of mono- and diglycerides of fatty acids as emulsifiers, xanthan gum as a hydrocolloid, and pea as the protein source led to the most comparable results, especially regarding specific volume, textural properties, and colour. The added emulsifier was suspected to positively influence the results due to the replication of the emulsifying characteristics of the egg yolk. Psyllium husk fibre was another ingredient that positively affected textural properties, with its incorporation resulting in a softer crumb texture in one of the no-protein egg alternatives and one of the high-protein egg alternatives. Hydrocolloids, like xanthan gum, positively influenced the cake batter’s viscosity and texture properties, especially hardness. By contrast, they restricted the structure setting of the cake and inhibited the formation of a high specific volume as a direct competitor for the water in the system. Hot-swelling hydrocolloids (methylcellulose, guar gum, locust bean gum) performed better regarding the structure setting properties of the cakes. However, the nutritional composition of the analysed egg replacers could not match the positive attributes of eggs, particularly the high protein content and the even distribution of essential amino acids throughout the egg protein. According to the performed market research on the available egg replacers for baking purposes, this is a recurring issue. Additionally, market research has shown that egg replacers cannot match the functionality of conventional eggs in terms of volume, taste, and texture, which reflects the findings of the current study. On the other hand, egg alternatives containing higher amounts of plant-based proteins performed poorly in simulating the egg proteins’ characteristics during the baking process. Therefore, a new approach for developing egg replacements must be taken into consideration, using ingredients that provide the nutritional and techno-functional properties of eggs in a pound cake system. Limitations that need to be considered for the results of this research are the small sample size and the application in a controlled baking environment. Future research should focus on using more samples under different baking conditions, replicating the cake production in the industry or for the average consumer. Additionally, using protein sources not used regularly in commercial egg replacers, e.g., microalgae, faba bean, and lentil, could provide progress for closing the gap in the egg replacer market.
Summarising all determined attributes, none of the egg replacers achieved comparable results to the control cakes, displaying a significant gap in the current development state of commercial egg replacers.