*Article* **The Effect of Brewing Process Parameters on Antioxidant Activity and Caffeine Content in Infusions of Roasted and Unroasted Arabica Coffee Beans Originated from Different Countries**

**Anna Muzykiewicz-Szyma ´nska \* , Anna Nowak, Daria Wira and Adam Klimowicz**

Chair and Department of Cosmetic and Pharmaceutical Chemistry, Pomeranian Medical University in Szczecin, Powsta ´nców Wielkopolskich Ave. 72, 70-111 Szczecin, Poland; anna.nowak@pum.edu.pl (A.N.); daria.wira@pum.edu.pl (D.W.); adam.klimowicz@pum.edu.pl (A.K.)

**\*** Correspondence: anna.muzykiewicz@pum.edu.pl

**Abstract:** Coffee is one of the most often consumed beverages almost all over the world. The multiplicity of beans, as well as the methods and parameters used to brew, encourages the optimization of the brewing process. The study aimed to analyze the effect of roasting beans, the brewing technique, and its parameters (time and water temperature) on antioxidant activity (determined using several in vitro methods), total polyphenols, flavonoids, and caffeine content. The infusions of unroasted and roasted Arabica beans from Brazil, Colombia, India, Peru, and Rwanda were analyzed. In general, infusions prepared from roasted beans had higher antioxidant activity and the content of above-mentioned compounds. The hot brew method was used to obtain infusions with a higher antioxidant activity, while the cold brew with higher caffeine content. The phenolic compound content in infusions prepared using both techniques depended on the roasting process. Moreover, the bean's origin, roasting process, and brewing technique had a significant effect on the tested properties, in contrary to brewing time and water temperature (below and above 90 ◦C), which had less impact. The results confirm the importance of coffee brewing optimization.

**Keywords:** coffee Arabica; roasting process; brewing methods; antioxidant activity; polyphenols; flavonoids; caffeine; pH of infusions; tannins

#### **1. Introduction**

Coffee trees belong to the constantly green trees and shrubs of the Rubiaceae family [1]. The optimal conditions for the growth of different coffee varieties occur in so-called the Coffee Belt located in the intertropical areas [2]. Ninety-five percent of global production of Coffea Arabica and Coffea Robusta are used for consumption, especially Coffea Arabica due to its delicate and less bitter taste [3]. An annual production of green coffee beans is about nine million tons. Even though Ethiopia is the cradle of Coffea Arabica [4], Brazil is the greatest world coffee exporter [1]. Other major suppliers of coffee beans are South American countries (Colombia and Peru) as well as African and Asian, such as Rwanda, above-mentioned Ethiopia, and Vietnam, Indonesia, and India [4].

After harvesting, the coffee beans are selected, shelled, and roasted at 200–250 ◦C [5]. The desired aroma and flavor of coffee are related to the volatile and non-volatile compounds. Their content is affected by the selection of appropriate roasting parameters and the subsequent Maillard reaction, which additionally leads to a change of beans color as a result of the production of melanoidins [6]. There are many techniques of coffee brewing. Generally, they can be divided into high- and low-temperature processes. The most popular methods of hot coffee brewing are the Turkish technique, French press, Aeropress, espresso, and simple infusion method [7,8]. Turkish coffee preparation involves boiling a mixture of coffee with water. The brew cannot be filtered [8]. A simple infusion based on pouring the

**Citation:** Muzykiewicz-Szyma ´nska, A.; Nowak, A.; Wira, D.; Klimowicz, A. The Effect of Brewing Process Parameters on Antioxidant Activity and Caffeine Content in Infusions of Roasted and Unroasted Arabica Coffee Beans Originated from Different Countries. *Molecules* **2021**, *26*, 3681. https://doi.org/10.3390/ molecules26123681

Academic Editors: Jan Oszmianski, Sabina Lachowicz and Francesco Cacciola

Received: 10 May 2021 Accepted: 14 June 2021 Published: 16 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

ground grains with hot water (~85–95 ◦C) and macerated for about 5 min [7,9]. During the preparation of espresso, the roast ground coffee is briefly pressurizing with hot water using a percolator to obtain a small concentrated cup of coffee. The pressure is one of the most important parameters of the coffee brewing process [10]. Aeropress created in 2005 by Adler is a device that uses pressure, overflow, and French press techniques. Hot water and the short brewing time (~2 min) are used to prepare infusions [7,9]. Methods of cold coffee extraction include direct and indirect immersion methods, dripping, and the French press [10]. To prepare cold brew coffee using the direct method, water at room temperature is applied, as in the cold drip case. Then, the ground coffee beans are immersed in a proper amount of water for at least 6–8 up to 24 h followed by filtration [11,12]. In the cold drip method, water at a temperature of 20–25 ◦C is slowly dripped onto coffee powder placed in the filter to obtain a relatively intensive coffee extract in the beaker. For this purpose, one drop of water is added usually every 5 or 10 s. Recently, a French press method, the so-called plunger pot, has gained interest. This technique uses the pot with the plunger, to which hot or cold water is poured inside [7,10]. It allows to get fresh unfiltered infusion immediately. French press coffee maker enables to depress the plunger to separate coarsely ground coffee beans from the liquid [13].

Many researchers confirmed the health-promoting properties of consuming coffee brew. The most important benefits are summarized in Table 1. In addition to the taste, coffee beans are a valuable source of bioactive compounds, whose profile depends on coffee species, thermal processing, cultivating conditions, and harvesting time [11,14]. Due to the content of antioxidants, as well as caffeine, the brews of coffee belong to the functional food sector [14]. The caffeine content in seeds ranges from 0.3% to 2.5% and is twice as high in Coffee Robusta as in Coffee Arabica [3,5]. Ground coffee brews are also rich in phenolic compounds, such as chlorogenic acid, hydroxycinnamic acids, and their derivatives such as caffeic and ferulic acids and alkaloids (mainly caffeine and small amounts of theobromine and theophylline), diterpenoid alcohols (cafestol and kahweol), carbohydrates, lipids, and volatile and heterocyclic compounds [5,15].


**Table 1.** Health-promoting effect of coffee infusions on the human body.


There are many reports on the coffee brewing process, but most of them are focused on the assessment of selected parameters, such as the impact of the intensity of the roasting process, the thickness of the grinding, or the method of brewing. The present study aimed to evaluate several parameters of the coffee brewing process, such as the method of making infusions (cold and hot brew), water temperature (cold, below or above 90 ◦C), and brewing time (9 and 24 h as well as 4 and 10 min). The details of the various brewing methods are presented in Table 2. For this purpose, unroasted and roasted Arabica coffee beans, imported from various countries (Brazil, Colombia, India, Peru, and Rwanda), belonging to the world's largest suppliers of this raw material, were analyzed. The influence of these factors on the in vitro antioxidant activity of infusions as well as on the content of polyphenols, flavonoids, and caffeine was analyzed. The results show that the bean's origin, roasting process, and brewing technique had a significant effect on the tested properties. The factors such as the difference in brewing time and water temperature (below and above 90 ◦C) had less impact. The obtained results could extend our knowledge on the brewing process of the most frequently consumed beverage in the world to verify parameters affected the content of health-promoting biologically active compounds, especially antioxidants.

**Table 2.** Description of the brewing methods.


**Table 1.** *Cont.*

summarized in Table 2.

#### **2. Results and Discussion**

#### *2.1. In Vitro Antioxidant Activity of Coffee Infusion*

Figure 1 presents free radical scavenging activities (RSA [%]) of infusions prepared using various brewing methods from unroasted (green) and roasted (brown) coffee beans from different countries, assessed by the DPPH and ABTS techniques. The brew's abilities to reduce ferric ions evaluated by the FRAP and PFRAP methods as well as to reduce cupric ions determined with CUPRAC method are presented in Figure 2.

**Figure 1.** Radical scavenging activity (RSA[%]) of coffee infusions evaluated using DPPH and ABTS **Figure 1.** Radical scavenging activity (RSA [%]) of coffee infusions evaluated using DPPH and ABTS methods. Vertical lines represent standard deviation (SD). Details regarding brewing methods are summarized in Table 2.

methods. Vertical lines represent standard deviation (SD). Details regarding brewing methods are

summarized in Table 2.

**Figure 2.** The ability of coffee infusions to reduce cupric and ferric ions determined by CUPRAC, FRAP, and PFRAP methods. Vertical lines represent standard deviation (SD). Details regarding brewing methods are **Figure 2.** The ability of coffee infusions to reduce cupric and ferric ions determined by CUPRAC, FRAP, and PFRAP methods. Vertical lines represent standard deviation (SD). Details regarding brewing methods are summarized in Table 2.

The infusions prepared from roasted beans were characterized by a higher antioxidant potential expressed as RSA [%] evaluated by DPPH technique (Figure 1), as compared to infusions from unroasted beans. The highest activity evaluated using DPPH method was found for the Colombian roasted coffee beans infusion (86 ◦C, 4 min), whereas the lowest for the Brazilian green coffee beans infusion brewed using the cold brew method for 9 h. The highest RSA [%] in the roasted coffee beans infusions group prepared using different brewing methods was most often observed for extracts from Colombian beans, whereas the lowest was for Indian bean brews. In the case of unroasted beans, the highest activity of infusions was found for Indian coffee, whereas the lowest was for infusions prepared from Brazil and Rwanda coffee beans. Prolongation of roasted beans brewing time generally increased the activity of cold brew infusions but reduced the activity of infusions brewed at a temperature above 90 ◦C. In the case of unroasted beans, the increase of activity due to the extension of the brewing time was observed only for coffee brewed at a temperature above 90 ◦C. All the results obtained by DPPH method could suggest that the hot brew method led to obtain infusions with higher antioxidant activity.

The infusions analyzed using ABTS method (Figure 1) were characterized by very high activities, ranging from 95.76% RSA (green beans from Colombia, brewed at 95 ◦C for 4 min) to 99.43% RSA (infusion of roasted coffee from Brazil—86 ◦C, 4 min). As a rule, the roasted coffee bean infusions were characterized by a higher antioxidant potential than analogously prepared extracts from unroasted beans. The highest activities of the infusions prepared from roasted coffee beans were observed most often for Brazilian beans, while in the group of extracts from green coffee beans, for the brews from Indian and Colombian beans. In the group of unroasted beans extracts, it was also found that infusions of Peru beans are frequently characterized by slightly lower activity assessed by ABTS method. However, prolongation of brewing time led to enhance of the antioxidant activity, especially for extracts prepared from unroasted beans. This tendency was also found for infusions prepared with the cold brew method and at a temperature above 90 ◦C. Therefore, infusions prepared using the cold and hot brew techniques showed a very high free radical scavenging potential evaluated using the ABTS technique; however, it is rather difficult to unequivocally evaluate the effect of brewing temperature on antioxidant activity of brews.

The reduction potential of the infusions was evaluated using the FRAP, PFRAP, and the CUPRAC methods (Figure 2). The highest activity determined by the FRAP method was found for roasted beans from Peru (95 ◦C, 4 min), while the lowest for the unroasted beans infusion from Rwanda beans (CB, 24 h). Moreover, it was found that activity of extracts from unroasted beans was often higher as compared to infusions prepared from roasted beans. This tendency was especially observed for infusions prepared using the cold brew method. In the case of roasted beans, the highest activity was often observed for coffee beans from India and Peru, and the lowest for beans from Colombia and Rwanda. Moreover, the highest activity evaluated by FRAP method was found for unroasted Brazil and Colombia beans, whereas the lowest for Rwanda coffee. Evaluation of the brewing time effect of roasted and unroasted beans showed that in the case of the cold method and during application water of above 90 ◦C, a shorter brewing time seemed to be more effective. In the case of the hot brew method using water at a temperature of ~85 ◦C, brewing for 10 min seems to be more effective. The impact of brewing method on the infusion's potential showed that the hot brewing technique led to obtain infusions with a higher reduction activity.

The highest reduction activity of iron ions assessed by the PFRAP method (Figure 2) was observed for unroasted Colombian coffee beans infusion (95 ◦C, 10 min), whereas the lowest for unroasted beans from Peru (CB, 24 h). Based on the analysis of the results obtained for roasted and unroasted coffee beans, it is rather difficult to clearly define the group showed higher activity. However, the highest potential for roasted beans was found for infusions brewed at 86 ◦C, whereas for unroasted beans at 95 ◦C. Moreover, in the group of unroasted and roasted beans infusions, the highest activities were obtained most often for extracts from Rwanda beans. Additionally, also roasted Indian coffee beans

brews were highly active. The lowest potential showed the infusions prepared from Peru coffee, both roasted and unroasted. In contrary to roasted, unroasted Indian coffee bean infusion had generally quite low antioxidant potential. The analysis of brewing time has established no beneficial effect of its prolongation on the tested potential of the extracts. The results suggested that in the case of hot brew (brown beans), 10 min process is generally more effective, while in the case of unroasted beans (85 ◦C)—4 min. Similar to the FRAP method, the analysis of the impact of the brewing method showed that hot coffee brewing techniques led to infusions with a greater reduction in the activity of ferric ions evaluated using PFRAP technique.

Application of CUPRAC method (Figure 2) to determine cupric ion reduction capacity suggest that Colombian roasted coffee extract (CB, 24 h) was characterized by the highest activity, whereas the lowest was found for the infusion prepared using the same method from Colombian unroasted beans. Comparison of the activities of roasted and unroasted coffee bean infusions leads to the conclusion that after application either the cold brewing method or 86 ◦C (4 min) roasted coffee bean infusions are more active, in contrary to the other brewing methods, where the green bean extracts were characterized by higher potential. Moreover, the highest ability to reduce cupric ions was found most often for Colombian coffee extracts (both roasted and unroasted), while the lowest for Rwanda roasted and Brazil unroasted coffee beans. Shorter brewing time was more effective to prepare infusions from brown and green beans using water below 90 ◦C and the cold brew method for unroasted beans. In the other brewing methods, prolongation of brewing time up to 24 h in cold method and up to 10 min in hot method seems to be more efficient.

The analysis of the effect of roasting coffee beans on the antioxidant activity of infusions showed that as a rule antioxidant potential of roasted coffee beans was higher, as compared to unroasted beans. Priftis et al. [21] evaluated the antioxidant activity of 13 varieties of coffee (roasted and unroasted) and confirmed that both types of coffee showed the high ability to scavenge free radicals. The impact of roasting on the activity of the obtained infusions depended on the variety of beans. This phenomenon may be related to the different chemical composition of beans. Moreover, the burning time also influenced the antioxidant activity of the obtained extracts. Dybkowska et al. [22] analyzed the antioxidant properties of coffee brews prepared from beans cultivated in Brazil, Ethiopia, Colombia, and India. The beans with various roasting degrees were analyzed—light, medium, and dark. They noticed that the process of roasting coffee beans increased the antioxidant potential of the obtained infusions. This phenomenon is directly related to the formation of melanoidins as a result of the Maillard reaction in coffee beans upon roasting. Melanoidins influence the antioxidant activity of coffee and its sensory properties. Their content in coffee depends, among others, on the intensity of the roasting process—the higher the roasting temperature, the higher content of these compounds, but lower their molecular weight. Ribeiro et al. [23] confirmed a higher content of melanoidins in roasted coffee bean infusions than in unroasted. In their study, higher antioxidant activity evaluated using FRAP and ABTS methods was found for roasted bean infusions as compared with infusions prepared from green coffee beans. Heˇcimovi´c et al. [24] also confirmed our observations, that both factors—coffee variety and the roasting degree—affect the antioxidant activity. In their study, similarly to ours, infusions prepared from unroasted beans showed a lower potential than extracts from roasted beans.

The country of the coffee bean's cultivation could also affect the antioxidant activity of infusions. High antioxidant potential was found quite often for coffee from Colombia and India, while infusions from beans from Peru and Rwanda were characterized by lower activity. Oszmia´nski et al. [25,26], based on the apple analysis, concluded that the factors such as year of harvest, growth period, storage conditions, geographic location and genetic variation, the effect of the region, agricultural practices, and cultivation method could have an impact on the plant biochemical profile including antioxidant potential. It is assumed that the roasting degree of beans could have a significant impact, apart from the individual composition of particular varieties. The beans from Colombia and India were burned medium-light to medium. Coffee from Peru was the most roasted of all the tested varieties, in contrary to the least roasted from Rwanda. The observations are confirmed by the suggestion of Priftis et al. [21] that the roasting degree of coffee beans should be optimized for high antioxidant activity. Dybkowska et al. [22] also suggest that for nutritional reasons, consumption of coffee with a short or medium degree of burnout is the most beneficial for human health.

The analysis of the brewing method suggests that the hot technique could have a positive effect to obtain infusions with a higher antioxidant potential than the cold method. Rao and Fuller [27] showed that antioxidant activity of hot-brewed infusions was higher as compared to cold-brewed infusions. They suggested that this phenomenon could be related to the hot water extraction of additional bioactive compounds, including those with antioxidant potential. It may be related to the higher content of caffeoylquinic and chlorogenic acid isomers in hot-brewed coffee than in cold-brewed infusions. Rao et al. [28] also suggested that hot water can more easily wet the oily surface of coffee beans, to enhance the efficiency of extraction of active compounds, including those with antioxidant potential. Coffee brewing using hot or cold water could affect the different solubilities of compounds with higher molecular weight, mainly melanoidins.

Temperature of the water to be used to prepare infusions (hot brew method) is another parameter to be taken into consideration. The antioxidant activity of infusions brewing with water at a temperature of above 90 ◦C was higher. There are many reports on the effect of the roasting degree or the method of brewing coffee on the antioxidant activity of infusions. However, there is less information on the influence of small differences in water temperature on the activity of the obtained infusions. Castiglioni et al. [29] investigated the effect of water temperature (70 ◦C, 90 ◦C, and cold) on the antioxidant activity of white and green tea from China and Malawi. They found a maximum efficiency using water at temperature of 90 ◦C and cold brewing method. Pramudya and Seo [30] studied the effect of the temperature of serving coffee (5 ◦C, 25 ◦C, 65 ◦C) on the sensor attributes and emotional responses of the coffee brews. The respondents most often described the taste of coffee served at 65 ◦C as "roasted flavor", while the taste of the brew served at 5 ◦C as "pungent aroma", "metallic flavor", and "skunky flavor". They also perceived similar observations when serving green tea at different temperatures. The authors found that the respondents valued more positive sensory features of infusions served at higher than at lower temperature.

However, it is difficult to clearly evaluate how prolongation of the brewing time could affect the antioxidant activity of the infusions. Górecki and Hallmann [31] compared the antioxidant activity of infusions brewing for 3 and 6 min. They obtained slightly higher results for infusions prepared during a 6 min process. However, these differences were not significant. Moreover, the authors also suggest that other factors, such as the roasting degree of beans, could affect the antioxidant potential of the infusions.

#### *2.2. Total Polyphenols and Flavonoids Content*

The total polyphenols and flavonoids content in the studied infusions are presented in Figure 3. The highest polyphenols content was found in the Rwanda roasted coffee beans infusion, brewed using the cold brew method for 9 h, whereas the lowest in unroasted Colombian coffee extract prepared during a 4-min brewing (86 ◦C). Generally, a higher content of polyphenols was found in roasted coffee bean infusions, particularly in those prepared using hot water. The cold brewing method was effective to obtain infusions with high content of polyphenols from unroasted coffee beans. In brews from roasted beans, the highest polyphenol content was found in the Rwanda coffee extracts, while the lowest in Colombian and Indian infusions. In unroasted bean infusions, the highest concentration of polyphenols was found most often in Indian coffee bean brews, while the lowest in Colombian. In the case of roasted beans, the extension of brewing time to 10 min in the hot brewing method usually increased the polyphenols content. However, no similar relationships were observed in the case of green coffee. The obtained results suggest that

### cold brewing increases the concentration of total polyphenols in infusions, as compared to traditional hot brewing.

**Figure 3.** The total polyphenols and flavonoids content in coffee infusions. Vertical lines represent standard deviation (SD). Details regarding brewing methods are summarized in Table 2.

The highest content of flavonoids (Figure 3) was found in the unroasted Indian coffee infusion (85 ◦C, 10 min), while the lowest in unroasted Brazilian coffee (85 ◦C, 4 min). In the group of infusions prepared with hot water, a higher content of flavonoids was found in extracts from roasted beans, while in the cold brew method, from unroasted beans. In unroasted bean infusions, the brews prepared from Indian coffee showed the highest flavonoid content, while Brazil beans the lowest concentration of these group of compounds. Among infusions obtained from roasted beans, the highest content of flavonoids was found for samples obtained from Rwanda beans, whereas the lowest for Indian coffee infusions. It is difficult to clearly assess the effect of time extension on the content of flavonoids in the tested brews. In cold brewing, the infusions prepared during the 9-h process are generally characterized by a higher content of the flavonoids. A shorter brewing time also seems to be rather optimal for brewing roasted coffee with water below 90 ◦C. Extending the brewing time to 10 min seemed to be more effective in the case of roasted beans brewing with water at 95 ◦C and green coffee with water below 90 ◦C. The obtained results suggest that in most cases the cold brewing contributes to obtain infusions

with a higher content of flavonoids than hot brewing. Similar results were observed for total polyphenols content.

Similar to the antioxidant activity, it was found that roasted bean infusions contained more polyphenols, including flavonoids, than unroasted. The highest content of these compounds was found in Rwanda coffee characterized by the lowest roasting degree. Dybkowska et al. [22] showed a decrease in the content of polyphenols in 100% Arabica beans infusions and blend coffee of Arabica and Robusta beans because of the roasting process. The thermolabile nature of these compounds contributes to their degradation after prolonged exposure to high temperature. The authors also emphasize that the loss of polyphenols is unfavorable due to the health-promoting effect of these compounds on the human body. Król et al. [32] also indicate a decrease of polyphenols content in the brews because of the extension of the roasting beans process.

The brewing method could also affect the polyphenols and flavonoids content. Cold brewing infusions of unroasted beans lead to higher content of these compounds, while the hot brewing technique seems to be more effective for roasted beans. Fibrianto et al. [33] compared the effect of the hot and cold brewing method on the polyphenols content in infusions prepared from roasted Arabica beans. Similar to our results, they also found a higher content of these compounds in extracts obtained using hot water.

The evaluation of the effect of water temperature in the hot brewing method showed that a higher content of polyphenols and flavonoids was observed in slightly more infusions prepared with water at 95 ◦C rather than at 86 ◦C. Merecz et al. [34] investigated the effect of the brewing method (hot water, percolator, and coffee machine) on the content of polyphenols and flavonoids in brews. In their study, the roasted and unroasted Arabica and Robusta beans were evaluated. However, their results cannot clearly confirm the influence of the brewing method on the content of these compounds in infusions. The concentration of polyphenols and flavonoids depended on the type of coffee (species, roasted/unroasted, and country of the bean's origins), as well as on the number of the brewings (1 to 3). In the case of roasted beans, the most effective method seemed to be application of a percolator (flooding the ground beans with cold water and then placed over a heat source). The content of polyphenols and flavonoids in unroasted beans was lower than in roasted. In the case of green coffee, the content of these compounds was higher in infusions prepared in a percolator and using hot water as compared to the brews from a coffee machine. Results of our study also lead to the conclusion that the type of coffee used (degree of roasting, country of origin, brewing method) had an impact on the content of biologically active compounds, including polyphenols and flavonoids in contrary to the slight differences in temperature of the water used to prepare the infusions. Merecz et al. [34] also emphasized that factors such as storage method and the degree of grinding beans could affect the profile of biologically active compounds in coffee infusions.

Furthermore, infusion time could affect the polyphenols content. In our study, it was found that extending of the brewing time, both in the cold (up to 24 h) and hot methods (up to 10 min), generally had a positive effect on the content of these compounds. In contrary, this tendency was not observed for the content of flavonoids, because in some cases, shorter brewing time seemed to be more optimal. Slightly different observations were made by Górecki and Hallmann [31]. They found that extending brewing time from 3 to 6 min contributed to a slight increase of flavonoids content and to a slight decrease of total phenols in coffee brews. However, these authors analyzed only the extracts prepared using the hot brewing technique. On the other hand, Cordoba et al. [11] confirmed our observation: extending the brewing time from 14 to 22 h (cold brewing method) increased the polyphenol content in coffee infusions.

#### *2.3. Caffeine Content in Coffee Infusion*

The caffeine content in studied infusions is presented in Table 3, whereas the chromatogram presenting the analysis of the selected infusion (unroasted beans from Peru— 86 ◦C, 4 min) in Figure 4. The highest caffeine content determined by HPLC method

was found in the Indian roasted bean infusion (CB, 24 h), while the lowest in the Peru unroasted beans brews (95 ◦C, 4 min). The higher caffeine concentration was found rather in brown bean infusions than in green coffee beans. The cold method was more effective to prepare infusions with a higher caffeine content than the traditional hot brewing technique. However, infusions brewed with water at 95 ◦C for 4 min were characterized by the lowest caffeine content. The obtained results suggest that the relatively high content of caffeine (over 60 mg/100 mL in 5% (*w*/*w*) infusion) was found in brews prepared from Indian coffee. The lowest caffeine content was generally found in coffee from Peru and Rwanda, especially in brews obtained from unroasted beans.

**Figure 4.** Chromatogram of the determination of caffeine in the infusion of unroasted beans from Peru (temperature of water—84–86 ◦C, brewing time—4 min).

Similar to the antioxidant activity as well as to the content of total polyphenols and flavonoids, higher caffeine concentration was usually found in roasted coffee infusions than in brews of green coffee beans. The highest caffeine content was observed in Indian coffee infusions, while the lowest concentrations were observed in coffee beans grown in Peru and Rwanda. Similar results, i.e., higher caffeine content in roasted than unroasted beans, were also obtained by Mubarak [14] and Motor and Beyen [35]. The influence of the roasting degree on the caffeine content is emphasized by Górecki and Hallmann [31]. They observed a significant decrease in caffeine content after long-term roasting of the beans. Moreover, they compared the content of this alkaloid in beans from conventional and organic crops. Conventional crops samples were characterized by a higher caffeine content, probably due to the use of nitrogen fertilizers that lead to increase the percentage of caffeine in coffee beans. Moreover, according to Gebeyehu and Bikila [36], the growing conditions of coffee trees could also modify the caffeine content in beans.


**Table 3.** The mean (±SD) caffeine content in studied infusions. Details regarding brewing methods are summarized in Table 2.

The evaluation of the effect of the brewing method showed that cold brew infusions usually are characterized by a high caffeine content. Moreover, it can be assumed that longer brewing time favor the preparation of infusions with a higher content of this alkaloid. Similar results were obtained by Fuller and Rao [37]. They found that cold-brew infusions

were characterized by a higher caffeine content as compared to extracts prepared using the hot method. The authors suggest that the higher caffeine content in cold infusions may be caused, among others, by extending the brewing time. In their study, the time was extended from 6 min (hot brew) to even 24 h (cold brew). Such a procedure could increase the intragranular diffusion and decrease the concentration of extractable coffee compounds in the hot brew, as compared to the cold brew. Moreover, the extraction from the surface and near-surface matrix occurs more rapidly than the diffusion of compounds through the intragranular pore network to the grain surface. In another study, Rao et al. [28] compared, among others, the content of bioactive compounds in infusions prepared using cold and hot methods, from coffee beans with varying roasting degrees. They found that the roasting process led to several chemical and physical changes in the bean matrix. Depending on the water temperature using to prepare the infusion, the above-mentioned changes affected most likely the ability, speed, and permeation efficiency of the various compounds. In their study, the water temperature affected the caffeine content in the brews. As in the case of our research, cold brew infusions seemed to be more effective to obtain higher caffeine content than hot water infusions. The high caffeine content in cold drip infusions was also demonstrated by Córdoba et al. [38].

Moreover, the water temperature could affect the caffeine content. Infusions prepared using the hot brew method with water temperature not exceeding 90 ◦C were a little more often characterized by a higher content of this alkaloid. Caprioli et al. [39] analyzed the Arabica and Robusta beans infusions prepared in two espresso machines used different pressure and temperature to brew coffee—at a temperature of 88–92 ◦C at a pressure of 9 bar and a temperature of 92–98 ◦C at a pressure of 7 bar. The caffeine content in Arabica bean infusions was higher in brews prepared at a higher temperature and lower pressure, while for Robusta coffee at lower temperature and higher pressure. In another study, Caprioli et al. [40] confirmed our observations: it is rather difficult to find consistent comparative data in the literature on the influence of coffee brewing parameters on the concentration of biologically active compounds in brews. The reason for this may be as a rule the application of a non-standard brewing method, characterized by different parameters such as the coffee-water ratio, the degree of beans roasting as well as differences in the units of the presented results. These factors can make it difficult to compare the data from different studies.

The analysis of the impact of brewing time on the caffeine content suggests that for the tested coffee varieties in our study, a longer brewing time seems to be more optimal. Similar results were obtained by Fuller and Rao [37]. In their study, the extending of brewing time from 400 to 1440 min increased the caffeine content in coffee infusions. The influence of the brewing method on the content of biologically active compounds, including caffeine, is also mentioned by Zaguła et al. [41]. The authors analyzed the influence of the application of a variable magnetic field on the caffeine content in black and green tea infusions. They suggested that a magnetic field assisted extraction could enhance the effectiveness of extracting the active substances from tea leaves to the infusion. Moreover, the authors emphasized that properly selected techniques designed to facilitate water-based extraction, using ultrasounds, magnetic fields, or microwaves may lead to technological advancements in the extraction of bioactive compounds from plant material.

#### *2.4. Tannins Content and pH of Coffee Infusions*

In our study, all the analyzed infusions contained tannins. The tannins in roasted coffee beans infusions were also found by Choi and Koh [42]. Patay et al. [43] also confirmed the content of these compounds in ripe and unripe seeds and pericarp of coffee beans.

The pH of all tested infusions was slightly acidic (Table 4). The pH of the unroasted bean infusions ranged from 5.16 (beans from Rwanda, cold-brewed for 9 h) to 6.58 (beans from Peru, brewed for 4 min with water at 95 ◦C). The pH of roasted bean infusions was more acidic, from 4.99 (Rwanda coffee, cold-brewed for 9 h) to 5.71 (Indian beans, cold-brewed for 24 h). Fibrianto et al. [33] analyzed the pH of infusions depending on the

degree of roasting beans (light, medium, and dark). As the degree of roasting increased, the pH of the infusions was more alkaline. Slightly roasted bean infusions were more acidic.


**Table 4.** pH of coffee infusions. Details regarding brewing methods are summarized in Table 2.

The analysis of the influence of the brewing method on the pH of the obtained infusions showed that in the case of unroasted beans, cold brewing leads to extracts with a lower pH, while the hot method, with a higher pH. The opposite relationship was noticed in the group of roasted bean infusions, as application of the hot-brew method generally led to more acidic infusions than with the cold-brew method. The exception was coffee from Rwanda (CB, 9 h) with the lowest pH. Rao and Fuller [27] compared the pH of infusions of roasted beans from different countries, prepared using the hotand cold-brew methods. In most cases, cold infusions were characterized by a more alkaline reaction. These observations were confirmed in our study. Rao and Fuller [27] also noted that coffee vendors often suggested that infusions prepared using cold- and hot-brew methods were characterized by different taste profiles due to different acidity levels. Therefore, it is believed that consumption of cold-brewed coffee, due to its lower acidity, could cause fewer gastrointestinal symptoms, sometimes observed after consuming coffee infusions. Rao and Fuller [27] clearly distinguish the pH assessment of infusions and their total titratable acidity. pH refers to the concentration of aqueous hydrogen ions, providing a metric for the quantity of deprotonated acid molecules in a tested sample, whereas the total titratable acidity is a measure of all acidic protons in a sample, including non-dissociated protons. Based on the obtained results, authors concluded that coffee infusions prepared using cold and hot brewing technique are similar, taking into account the total concentration of deprotonated acid compounds; however, they differ in the concentration and possibly the complexity of protonated acids at the pH of extraction. No correlation between perceived acidity in the flavor of coffee brews and pH was observed by Gloess et al. [44] and Andueza et al. [45]. Furthermore, Gloess et al. [44] found no correlation between the pH and the titratable acidity of the coffee brews. The authors explain that many of the acids presented in the coffee infusion may not be completely deprotonated at pH measurement of this infusions and as a consequence does not affect its pH, but could be measured during titration with alkali.

The evaluation of the impact of water temperature in the hot brew method on the pH showed that infusions prepared with water at 95 ◦C were often characterized by a more alkaline reaction. Salamanca et al. [46] evaluated the effect of the type of coffee (natural and washed Arabica as well as natural Robusta) and the extraction temperature profile (88–93 ◦C, 90 ◦C, 93–88 ◦C) on the pH of the infusion. The pH varied depending on the type of coffee, as well as the temperature of brewing. In the case of Arabica beans, the infusions prepared at 93–88 ◦C were more acidic, while those obtained in a constant temperature of 90 ◦C were more alkaline. In the case of the Robusta variety, the most acidic infusions were those brewed at 88–93 ◦C, while the alkaline were prepared at 93–88 ◦C. Regardless of the water temperature, as in our study, all prepared infusions were acidic, and pH ranging from 5.01 ± 0.82 to 5.74 ± 0.04.

The analysis of the influence of the brewing time on the pH of the obtained infusions showed that in the case of the cold brew method, the extracts obtained during a longer brewing time (24 h) had a higher pH, whereas, in the hot brewing technique, the infusions obtained during a shorter brewing time (4 min) had a more alkaline pH. Fuller and Rao [37] analyzed the pH of infusions prepared using the cold brew method for 400 and 1440 min. They observed more alkaline reactions for extracts prepared during a shorter brewing time.

The comparison of the pH of the infusions depending on the country of origin of the beans has shown that in the case of unroasted beans, the most alkaline are usually the infusions from Rwanda coffee, while the most acidic—the beans grown in Brazil. In the case of roasted beans, the most alkaline pH, regardless of the water temperature and infusion time, was found for infusions of beans imported from India, while acidic for extracts obtained from beans grown in Colombia and Peru. Rao and Fuller [27] also compared the pH of infusions made from roasted beans grown in different countries (Brazil, Ethiopia, Myanmar, Colombia, and Mexico). In the case of the hot-brew method, the infusion made from Brazilian beans was the most alkaline, whereas the most acidic was made from the coffee grown in Ethiopia. In the case of cold brewing, the most alkaline was the infusion of beans grown in Myanmar, and the most acidic, also the extracts of Ethiopian beans. In another study of Fuller and Rao [37], it was noted that pH of infusions depended not only on the hot or cold method, but also on the degree of roasting and grinding the beans. The most acidic reaction was found for medium-coarse coffee, while the most alkaline for dark roast and medium coarse ground coffee (dark-medium).

#### *2.5. Statistical Analysis*

The statistically significant Pearson correlation coefficients were obtained between methods: DPPH vs. ABTS (r = 0.489; *p* < 0.0001), FRAP vs. F-C (r = 0.314; *p* < 0.02), ABTS vs. flavonoids content (r = 0.270; *p* < 0.04) as well as F-C vs. flavonoids content (r = 0.688; *p* < 0.0001). Correlations between the caffeine content and the results of antioxidant activity as well as the total polyphenols and flavonoids content, were also assessed. Statistically significant correlation coefficients were obtained between the caffeine content and antioxidant activity evaluated with DPPH (r = 0.390; *p* < 0.003) and ABTS (r = 0.453; *p* < 0.001) methods. Moreover, a significant correlation was found for caffeine vs. total polyphenols (r = 0.355; *p* < 0.006) and for caffeine vs. flavonoids content (r = 0.445; *p* < 0.001). The statistical significance of differences between the results obtained for infusions prepared using various brewing parameters was also assessed. The differences between the antioxidant activity of infusions prepared from roasted and unroasted beans were statistically significant (z = 4.871; *p* < 0.0001). Caffeine content in infusions of green and brown beans also differed significantly (z = 4.206; *p* < 0.0001). The above-mentioned differences, between the concentration of polyphenols and flavonoids, were statistically insignificant. The assessment of differences between the coffee brewing temperature (CB vs. 84–86 ◦C and CB vs. 93–95 ◦C) showed that in the case of antioxidant activity, the differences were statistically significant between the cold-brew method and 93–95 ◦C (z = 2.274; *p* < 0.03). The content of polyphenols and flavonoids differed significantly between the cold-brew method and 84–86 ◦C as well as cold brew method and 93–95 ◦C (z = 4.436 and z = 3.212, respectively;

*p* < 0.01). Furthermore, the caffeine content differed depending on the brewing method (CB vs. 84–86 ◦C—z = 3.808; *p* < 0.001 as well as CB vs. 93–95 ◦C—z = 2.688; *p* < 0.01). The differences between the brewing temperatures in the hot brew method (84–86 ◦C vs. 93–95 ◦C) were not statistically significant taking into account antioxidant activity, polyphenols and flavonoids content, as well as caffeine concentration in the analyzed infusions. The evaluation of differences between shorter and longer brewing time in individual brewing techniques (9 h vs. 24 h and 4 min vs. 10 min) showed that only the differences between caffeine content in cold brew infusions (9 h vs. 24 h), were statistically significant (z = 1.988; *p* < 0.05). The statistically significance of differences between the antioxidant activity, the content of caffeine, polyphenols, and flavonoids in infusions from beans (both roasted and unroasted) cultivated in different countries was also assessed. In the case of antioxidant activity, the differences were statistically significant between coffees imported from Brazil and Colombia (z = 3.806; *p* < 0.001), Colombia and Peru (z = 3.622; *p* < 0.001), as well as Colombia and Rwanda (z = 4.167; *p* < 0.001). The content of polyphenols and flavonoids differed significantly between infusions obtained from beans cultivated in Brazil vs. Colombia (z = 2.057; *p* < 0.04), Brazil vs. Rwanda (z = 4.171; *p* < 0.001), Colombia vs. Rwanda (z = 2.743; *p* < 0.01) as well as Peru vs. Rwanda (z = 2.600; *p* < 0.01). Caffeine content differed significantly between infusions brewed with coffee beans from Brazil and India, Colombia and India, India and Peru as well as India and Rwanda (z = 3.059; *p* < 0.01).

#### **3. Materials and Methods**

#### *3.1. Chemicals*

Acetic acid (99.5%), aluminum chloride hexahydrate, copper(II) chloride dihydrate, disodium hydrogen phosphate dihydrate, 96% ethanol, 36% hydrochloric acid, iron(III) chloride hexahydrate, methanol, phosphoric acid, potassium persulfate, potassium dihydrogen phosphate, potassium hexacyanoferrate(III), sodium acetate anhydrous, sodium carbonate anhydrous, sodium hydroxide, sodium nitrite, and trichloroacetic acid were purchased from Chempur, Poland. Neocuproine was delivered by J&K Scientific, Germany. Folin–Ciocalteu reagent, acetonitrile, iron(II) sulfate heptahydrate, gallic acid were supplied by Merck, Germany, whereas rutin trihydrate by Roth, Germany. ABTS (2,20 azino-bis(3-ethylbenzothiazoline-6-sulfonic acid), DPPH (2,2-diphenyl-1-picrylhydrazyl), TPTZ (2,4,6-tris(2-pyridyl)-s-triazine), Trolox (6-hydroxy-2,5,7,8-tetramethylchromane-2 carboxylic acid), caffeine were purchased from Sigma-Aldrich, USA. All the chemicals were of analytical grade.

#### *3.2. Preparation of Coffee Brews*

Unroasted and roasted Arabica coffee beans from Brazil, Colombia, India, Peru, and Rwanda were used to prepare the infusions. The region of coffee from individual countries and the degree of roasting are summarized in Table 5. Coffee beans were ground using electric grinder (CTC Clatronic KSW 3306, Germany), immediately before sample preparation. The roasted beans were ground for 15 s, while unroasted, due to the greater hardness, for 70 s, until the coffee was finely ground. The ground beans were poured over with boiled tap water at different temperatures (hot-brew and cold-brew method) and subjected to different brewing times. The details of the various brewing methods are presented in Table 2. The completed brews were filtered through filter papers. Prepared 5% (*w*/*w*) infusions were filtered through Whatman's filter papers no. 4. All the extracts were stored at +4 ◦C until the analysis.


**Table 5.** The origin of the analyzed coffee beans.

#### *3.3. Evaluation of Antioxidant Activity*

The antioxidant activity of infusions was evaluated by several in vitro methods. The ability to scavenge free radicals (RSA [%]) was assessed by the DPPH and ABTS methods. Moreover, the ability of samples to reduce ferric and cupric ions was evaluated using FRAP (ferric reducing antioxidant power), PFRAP (potassium ferricyanide reducing power), and CUPRAC (cupric ion reducing antioxidant capacity) methods. The evaluation of antioxidant activity by DPPH, ABTS, and FRAP methods was performed as described by Muzykiewicz et al. [47]. To evaluate ferric reducing capacity of infusions, the FRAP method, as described by Apak et al. [48], and the PFRAP technique (with slight modifications), according to Jayaprakasha et al. [49], were used. The incubation time was reduced to 10 min and absorbance was measured at 734 nm. The spectrophotometric measurements were performed in 1 cm cuvettes using Hitachi U-5100 spectrophotometer (Japan). In DPPH and ABTS methods the activity was expressed as RSA [%], whereas in CUPRAC technique as Trolox equivalents (TEAC)—mg Trolox/g RM (raw material). The reducing power evaluated using FRAP and PFRAP method was presented as FeSO<sup>4</sup> equivalents mg FeSO4/g RM. Three samples were prepared from each extract and the results are presented as an arithmetic mean ± standard deviation (SD).

#### *3.4. Evaluation of Total Polyphenols and Flavonoids Content*

The total polyphenols content (Folin–Ciocalteu method) was evaluated as described by Muzykiewicz et al. [47], whereas the flavonoids according to Saeed et al. [50]. The spectrophotometric measurements were performed in 1 cm cuvettes using Hitachi U-5100 spectrophotometer (Japan). The total polyphenols content was expressed as gallic acid (GA) equivalents (GAE)—mg GA/g RM, whereas flavonoids content as rutin equivalents—mg rutin/g RM. Three samples were prepared from each extract and the results are presented as an arithmetic mean ± standard deviation (SD).

#### *3.5. Evaluation of Tannins Content and pH of Infusions*

Tannins content in infusions was analyzed according to Saeed et al. [50]. The few drops of 0.1% FeCl<sup>3</sup> were added to the coffee infusion. The appearance of a blue color indicated the presence of tannins in the coffee infusion. The pH of brews was measured using Thermo Electron Orion Benchtop 410A pH meter (USA).

#### *3.6. HPLC Analysis*

The concentration of caffeine in all coffee infusions was determined by high-performance liquid chromatography (HPLC-UV, Knauer, Germany). The tested compound was separated on a 125 × 4 mm column containing Hyperisil ODS (C18), particle size 5 µm. The mobile phase consisted of 0.5 M H3PO<sup>4</sup> (pH 2.5), acetonitrile and MeOH in the ratio 180:20:10 (*v*/*v*/*v*), flow rate was 1 mL/min 20 µL of the analyzed sample was injected on the column. The determinations were carried out at 272 nm. The correlation coefficient of the calibration curve was r = 0.999 (y = 370683x + 32.205, retention time—2.05 min). Each sample was analyzed in triplicate, and the results are presented as arithmetic mean ± standard deviation (SD).

#### *3.7. Statistical Analysis*

The Pearson's linear correlation between the antioxidant activity (DPPH, ABTS, FRAP, PFRAP, CUPRAC methods), polyphenols, flavonoids as well as caffeine content was determined. The significance of differences between the results of antioxidant activity, the content of polyphenols, flavonoids, and caffeine, obtained for infusions prepared using various parameters (method, time and temperature of brewing), considering the roasting and origin of beans was determined with Wilcoxon signed rank-test (parameter z). *p* < 0.05 was considered to be statistically significant. All the calculations were done with Statistica 13.3PL Software (StatSoft, Poland).

#### **4. Conclusions**

The results of the study showed that the selection of beans and brewing methods could have a significant effect on antioxidant activity, polyphenols, flavonoids, and caffeine content, as well as the pH of the infusions prepared from Arabica coffee beans. In general, a higher antioxidant activity and content of the above-mentioned biologically active compounds were obtained in the infusions prepared from roasted beans, as compared to the unroasted coffee beans. The origin of the beans and the brewing technique (hot or cold brew) also influenced the tested properties. Cold-brew infusions were generally characterized by a higher caffeine and total polyphenols (including flavonoids) content in the case of unroasted beans. The hot brewing method led to obtain extracts with higher antioxidant activity and the content of phenolic compounds in the case of roasted beans. In this study, the coffee beans were imported from different countries and were characterized by different degree of roasting, which also had a significant impact on the characteristics of infusions. It seems that factors such as brewing time (9 h vs. 24 h as well as 4 min vs. 10 min) and water temperature (below and above 90 ◦C) had a less significant impact on the tested properties. All infusions were slightly acidic and contained tannins. The results suggest that origin of coffee beans and brewing parameters seem to be responsible for the tested properties of infusions, therefore of their preparation should be optimized to obtain infusions with the most favorable content of biologically active compounds.

**Author Contributions:** Conceptualization, A.M.-S., A.N. and D.W.; methodology, A.M.-S., A.N., D.W. and A.K.; validation, A.M.-S., A.N. and A.K.; formal analysis, A.M.-S. and A.N.; investigation, A.M.-S., A.N. and D.W.; resources, A.M.-S. and D.W.; writing—original draft preparation, A.M.-S., D.W. and A.N.; writing—review and editing, A.M.-S. and A.K.; visualization, A.M.-S.; supervision, A.K.; project administration, A.M.-S.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding. The APC was funded by Pomeranian Medical University in Szczecin.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available in this article.

**Acknowledgments:** We express our gratitude to Adam Sołtysiak from the coffee roastery "To Kawa" in Szczecin (http://tokawa.pl/, accessed on 15 June 2021), for providing the coffee samples used in the study.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **References**


## *Article* **Development of a High-Fibre Multigrain Bar Technology with the Addition of Curly Kale**

**Hanna Kowalska 1,\* , Jolanta Kowalska <sup>1</sup> , Anna Ignaczak 1,\*, Ewelina Masiarz 1,\*, Ewa Domian <sup>1</sup> , Sabina Galus <sup>1</sup> , Agnieszka Ciurzy ´nska <sup>1</sup> , Agnieszka Salamon <sup>2</sup> , Agnieszka Zaj ˛ac <sup>1</sup> and Agata Marzec <sup>1</sup>**


**Abstract:** The aim of this study was to find the effect of kale and dietary fibre (DF) on the physicochemical properties, nutritional value and sensory quality of multigrain bars. A recipe of multigrain bars was prepared with the addition of fresh kale (20% and 30%) and DF preparations (apple, blackcurrant, chokeberry and hibiscus). The bars were baked at 180 ◦C for 20 min. These snack bars, based on pumpkin seeds, sunflower seeds, flaxseed and wholegrain oatmeal, are a high-calorie product (302–367 kcal/100 g). However, the composition of the bars encourages consumption. In addition to the ability to quickly satisfy hunger, such bars are rich in many natural ingredients that are considered pro-health (high fibre content (9.1–11.6 g/100 g), protein (11.2–14.3 g/100 g), fat (17.0–21.1 g /100 g, including unsaturated fatty acids), carbohydrates (20.5–24.0 g/100 g), as well as vitamins, minerals and a large number of substances from the antioxidant group. The addition of kale caused a significant increase of water content, but reduction in the value of all texture parameters (TPA profiles) as well as calorific values. The content of polyphenols was strongly and positively correlated with the antioxidant activity (r = 0.92). In the bars with 30% addition of kale (422 mg GA/100 g d.m.), the content of polyphenols was significantly higher than based ones (334 mg GA/100 g d.m.). Bars with the addition of the DF were characterized by a higher antioxidant activity, and the content of carotenoids, chlorophyll A and B and polyphenols. High sensory quality was demonstrated for all (from 4.8 to 7.1 on a 10-point scale). The addition of fibre preparations was also related to technological aspects and allows to create attractive bars without additional chemicals.

**Keywords:** snack; baking; carotenoids content; chlorophyll content; total polyphenols content; calorific value; sensory properties

#### **1. Introduction**

Consumers are increasingly turning to snack foods. This trend may be due to the fast pace of people's lives, which results in a lack of time to prepare and eat traditional meals. In addition, they are paying attention to what they eat because they are more aware of the effects of food on their health. Cereal products, fruits and vegetables occupy an important place in the daily diet because they contain many valuable ingredients such complex carbohydrates, including dietary fibre (DF), vitamins, antioxidant compounds and minerals that can be considered pro-health. However, their consumption is still too low. Producers, to meet the requirements of consumers, introduce new, innovative products to the market such as cereal-based products with the addition of fruit or vegetables. Replacing traditional high-calorie snacks with products with a high content of bio-ingredients may have a

**Citation:** Kowalska, H.; Kowalska, J.; Ignaczak, A.; Masiarz, E.; Domian, E.; Galus, S.; Ciurzy ´nska, A.; Salamon, A.; Zaj ˛ac, A.; Marzec, A. Development of a High-Fibre Multigrain Bar Technology with the Addition of Curly Kale. *Molecules* **2021**, *26*, 3939. https://doi.org/ 10.3390/molecules26133939

Academic Editors: Jan Oszmianski, Sabina Lachowicz and Francesco Cacciola

Received: 20 May 2021 Accepted: 24 June 2021 Published: 28 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

beneficial effect on health. It can prevent many diseases, such as diabetes, arteriosclerosis, or high blood pressure.

Fresh and processed vegetables are a source of valuable nutrients. A greater degree of consumer attention should be focused on cruciferous vegetables, which are little appreciated. Kale is not a popular vegetable in the diet of consumers because of its taste, flavour and characteristic texture. Cruciferous vegetables are good sources of fibre, polyphenols, and glucosinolates. It is rich in biologically active substances with antioxidant, bactericidal and fungicidal properties [1]. This vegetable contains polyphenols, including flavonoids, which by inhibiting the activity of phosphodiesterase and cyclooxygenases can reduce platelet aggregation; therefore, it is recommended in atherosclerotic diseases. Glucosinolates may reduce the risk of cancer development [2]. Sulforaphane has the strongest anticancer properties. It shows an inhibitory effect on angiogenesis and the formation of metastases [3]. Kale is characterized by a high content of vitamins, such as in edible parts: C (120 mg/100 g), A (0.9 mg/100 g), B1 (0.1 mg/100 g), B2 (0.2 mg/100 g), B6 (1.6 mg/ 100 g), and E (1.7 mg/100 g), and also contains folic acid and niacin. In addition, it contains a large number of essential micro and macro elements, including, in edible parts: calcium (157 mg/100 g), potassium (530 mg/100 g) and iron (1.7 mg/100 g) [4,5]. Carotenoids, but also quercetin and camferol, are responsible for its strong antioxidant capacity [6]. The multitude of active compounds in kale translates into the health benefits of its consumption. Many studies have shown that frequent kale consumption reduces the incidence of cancer in various parts of the digestive tract, lung cancer and others. Son et al. [7], in view of the nutritional needs of patients with impaired renal function, attempted the production of kale with reduced potassium content without compromising the yield and quality. The potassium deficiency in kale was eaten up by an increase in total glucosinolate content, which is an indicator of the anticancer activity of cruciferous vegetables. Chlorophyll, contained in the raw material, has an antiseptic and immunizing effect [8,9]. The fresh leaves are suitable for direct consumption, and they can be added to salads or boiled as an ingredient of soups or fried. Michalak et al. [10] presented the possibility of using kale fermentation by autochthonous lactic acid bacteria in the creation of bioactive derivatives of phenolic compounds that may have anticancer properties. In recent years, it has also become a fashionable addition to juices or smoothies [11]. In the USA, powdered kale is used to make dietary supplements. On the local market, there are capsules with kale powder as an ingredient.

Dietary fibre is a heterogeneous mixture of carbohydrate polymers found in plant raw materials. It refers to a large number of substances that exhibit a wide variety of physicochemical properties, with a general division into water-soluble and insoluble compounds. However, the way of processing, e.g., cereals, reduces its content in products. There is a need to make DF preparations that can be added to enrich various products. In addition, DF has important health and technological functions in food production. The raw materials for the production of DF preparations are industrial fruit and vegetable waste (apple pomace, blackcurrant pomace, waste from the processing of carrots, tomatoes) as well as bran, corn cobs, chaff and straw, and legumes (mainly soybeans and peas) [12]. The soluble fraction includes pectin, ß-glucans, gums, mucilage, and a wide range of indigestible oligosaccharides (including inulin). The insoluble fraction includes: lignin, cellulose, and hemicellulose. Each of the two types of fractions has different physiological effects. Soluble DF is less common in food than insoluble DF, but it has a significant impact on digestive and absorbent processes [12–14]. Fruit and vegetable DF has a much higher proportion of soluble DF, while cereal DF contains more insoluble cellulose and hemicellulose [15].

The addition of DF to bakery products, such as cereal snacks or multigrain snack bars, is justified both in terms of health and technology. Consuming products with DF helps to prevent many civilization diseases, such as: obesity, type 2 diabetes, ischemic heart disease, gallstone disease, constipation, and flatulence. It also reduces the risk of developing certain cancers. Its use in food production results from its ability to bind water, gel formation,

emulating and stabilizing properties, and fat mimetic properties, therefore it is primarily a structure-forming component and filler and affects the sensory quality of food [12,16]. The European Food Safety Authority (EFSA) allows the nutrition claims "source of fibre" and "high fibre" on food packaging with a content of at least 3% (1.5 g/100 kcal) or at least 6% fibre, respectively (3 g/100 kcal) [17–19]. Food has an increased level of fibre when it is at least 25% higher than similar foods. Products that contain health claims for dietary fibre must also meet requirements for adequately low fat, including saturated fat and low cholesterol [20]. Products containing a large amount of DF can be classified as functional and/or health-promoting food. In the technology of producing granular bars, DF allows for a better, more compact combination of ingredients and obtaining the appropriate structure of the product, mainly due to water absorption and mechanical durability. When deciding to use a DF blend in the product, it should be taken into account that their effect on texture may vary depending on the product formulation.

The aim of the research was to evaluate the use of kale and fibre preparations as an added value to multigrain bars. The scope of the study was the effect of kale and dietary fibre (DF) on the physicochemical proper-ties, nutritional value and sensory quality of multigrain bars.

#### **2. Results**

#### *2.1. Development of the Recipe Composition and Production Technology of Multigrain Bars*

At the stage of preliminary research, the recipe composition of the base bars was developed using baking. To improve the attractiveness of such a snack in terms of sensory and health, the recipe has been enriched with kale and DF preparations. Table 1 presents the chemical characteristics of fresh kale and multigrain bars with the addition of fresh and dried (microwave-blanched) kale.



The raw material composition allowed to obtain a product with a DF content of 11.5%. Unrefined cereal products are characterized by a high content of DF, especially the insoluble fraction, similar to vegetables. Whole-grain oat flakes and flaxseed could play a significant role in the resulting DF content. The addition of kale did not have a significant effect on the differentiation of the bar composition in terms of DF content, but it caused a partial reduction in the caloric content of the bars (Table 1). The snacks obtained in this way can

be a valuable source of both DF fractions, as well as fat (pumpkin and sunflower seeds) and protein. Due to their high calorific value, they can be a tasty and valuable snack that can replace one of the main meals.

Korus [21] analysed the composition of fresh vegetables. She showed a slightly higher content of proteins (about 4.3 g/100 g), emphasizing the beneficial composition of amino acids, including exogenous ones. Korus [5] showed that the pre-treatment of kale leaves reduced the content of minerals and vitamins by 26–52% (blanching) and 29–75% (cooking). The highest content of minerals, B vitamins and tocopherols was recorded in the frozen kale leaves after blanching. After 12 months of storage of frozen leaves, they contained 24–74% of macronutrients, 40–71% of micronutrients, 45–71% of vitamin B1, 27–47% of vitamin B<sup>2</sup> and 69–85% of total tocopherols. In the study by Olsen et al. [22], green and red kale extracts have undergone a treatment including blanching, freezing and heat treatment by boiling in a bag. In both kale varieties, processing significantly decreased total phenolics, antioxidant capacity, and the content and distribution of flavonols, anthocyanins, hydroxycinnamic acids, glucosinolates, and vitamin C. Both extracts continued to inhibit colon cancer cell proliferation, but fresh kale extract had a much stronger effect. According to Korus [21], vegetables are low in fat, high in carbohydrates and DF and minerals, as well as vitamins and other important ingredients like antioxidants. Cruciferous vegetables deserve special attention. The author [21] showed that among all vegetables, this vegetable contains the most easily digestible calcium and protein and an exceptionally high amount of iron. Kale is also a significant source of DF, vitamins C and E (more than spinach and lettuce), provitamin A and antioxidants (more active than garlic, spinach, Brussels sprouts and broccoli).

Based on literature [5,21,22] data showing that kale processing reduces the content of many components, and because the content of individual ingredients was quite similar in both types of bars with fresh and dried kale (Table 1), the bars with 20 and 30% addition of fresh kale and DF preparations were used for a more detailed assessment of the physicochemical properties. This was to make bars with lower calories and to facilitate the preparation of ingredients. Depending on the method of drying the kale, it could take up to several hours.

#### *2.2. Water Activity and Its Content in Multigrain Bars*

The water activity (Aw) in the bars was in the range of 0.857–0.953 (Table 2). This level of Aw classifies bars in the group of moist foods (in the range of 0.90–1.00) and with average Aw (in the range of 0.55–0.90), and thus in the food in which some microorganisms can develop (no microbiological stability). With the increase in the proportion of kale in the composition, the water activity increased. The base bars, i.e., bars without kale, were characterized by the lowest water activity (0.857). The type of added DF preparation had a significant effect on the water activity of the obtained bars (Table 2). Among the bars with 20% addition of kale, these with apple and chokeberry DF (0.943–0.944) were characterized by higher Aw, while bars with hibiscus DF (0.914) has significantly lower Aw.

The analysis of water content in the tested bars showed a significant effect of the type of added DF preparation and the amount of added kale on this parameter (Table 2). This varied widely, from approximately 17.3 to 41.1%. The water content in the control bars (without the addition of kale) differed significantly from the water content in the samples with the addition of kale. Increasing the proportion of kale from 20 to 30% resulted in a significant increase in the water content in the bars.


**Table 2.** The effect of kale and fibre preparation on water content, water activity (Aw), and compression force (Fmax) and work (W) in multigrain bars.

\*—means significant difference at a confidence level of 0.05; a, b, c and A, B, C—homogeneous groups, the same letters mean no statistically significant differences between the analysed values of indicators; the codes are described in Table 4.

### *2.3. The Effect of the Addition of Kale and Fibre Preparation on the Texture of the Multigrain Bars* 2.3.1. Compression Test

The texture of the bars was tested on the basis of the compression test and the work required was calculated. The control bars showed the highest compression work (604.3 mJ), about 2 times higher than the other bars (Table 2). A significant effect of increasing the amount of kale on the value of this indicator was demonstrated; for bars with 20 and 30% addition of kale, the work/deformation energy of the samples was about 383.5 and 271.4 mJ, respectively. Higher humidity decreased the hardness and the work needed for the deformation in the compression test was lower. The type of DF had a significant effect on the hardness of the bars. The bars with the addition of DF-Chokeberry were distinguished by significantly greater hardness (Fmax = 438.3 mJ), and P-Hibiscus by significant softness (268.6 mJ). This type of DF turned out to be less useful in the production of bars already at the stage of preparation before baking.

#### 2.3.2. Texture Profile Analysis (TPA) Test

The addition of kale had a statistically significant effect on all parameters of the texture profile (Table 3), but no significant differences were observed in the amount of kale addition. The DF preparations had a significant effect on the parameters of the texture profile, only in the case of elasticity, no such effect was observed (Table 3).

The values of the hardness parameter for the tested bars ranged from 89 to 299 N. The control bars (298.2 N) had the highest hardness value, significantly lower values were achieved in bars with 20 (121.5 N) and 30% (94.6 N) kale addition. Therefore, it can be concluded that the kale bars were more brittle or soft than the base bars. However, the amount of kale added did not significantly affect the hardness of the bars. The results of this parameter were significantly influenced by the type of DF preparation used. The bars with the addition of apple and hibiscus DF had significantly lower hardness values than the bars with the addition of blackcurrant and chokeberry DF.

Elasticity discriminants of the tested bars changed, as did the hardness of bars with and without kale. The addition of kale decreased the value of this parameter. The base bars (without the addition of kale) had an elasticity of 0.54, the bars with 20% addition of kale 0.40, and the bar with 30% addition of 0.383. For bars with the addition of various DF preparations, the values of elasticity were at a similar level (0.40–0.48 N).


**Table 3.** The effect of kale and fibre preparation on the texture profile of multigrain bars. The bar codes are explained in Table 4.

\*—means significant difference at a confidence level of 0.05; a, b, c and A, B, C—homogeneous groups, the same letters mean no statistically significant differences between the analysed values of indicators; the codes are described in Table 4.

> The addition of kale significantly influenced the cohesiveness. The bars without the addition (base) were characterized by higher cohesiveness than the bars with the addition of kale. The type of DF preparation also had a significant effect. The bars with the addition of apple DF and 20% kale had the lowest cohesiveness values (0.26), the bars with the addition of hibiscus powder (0.37), blackcurrant DF (0.43) and chokeberry DF (0.48) were characterized by higher values.

> Guminess significantly depended on the addition of kale. The base bars (without addition) had higher values (145.0 N) than the bars with the addition of kale. The greater addition of kale did not cause any significant changes in the guminess of the bars. The effect of the various DF preparations was also significant. The bars with apple and hibiscus DF were characterized by low values of guminess (about 32 N), while the value of this parameter for bars with the addition of blackcurrant DF (75.7 N) and samples with the addition of chokeberry DF (88.8 N) was significantly higher.

> The chewiness of the bars decreased significantly (to about 6 times) in the bars with the addition of kale compared to the control bars, from 77.0 to 12.9 N for bars with 20% additive and 14.4 N with 30% additive. Moreover, bars with apple DF (12.9 N) and hibiscus (13.8 N) were characterized by low chewiness values, significantly higher values were obtained for bars with blackcurrant and chokeberry DF addition.

#### *2.4. The Effect of the Addition of Kale and Fibre Preparation on the Color of the Multigrain Bars*

The addition of kale in the amount of 20 and 30% did not significantly affect the brightness of the colour of the bars (Figure 1a). On the other hand, the type of added DF preparation significantly influenced the brightness of the colour of bars. The addition of blackcurrant, chokeberry and hibiscus DF caused a significant darkening of the colour bars. The bars with the addition of apple DF showed up to about 16% higher values of the L\* parameter. This dependence may be due to the colour of the DF preparations, apple DF is a light beige powder, while the remaining DF are dark red powders. Dark red discoloration of the preparations may be due to the presence of anthocyanin pigments.

**Figure 1.** The influence of kale and fibre preparation addition on the colour parameters of multigrain bars: (**a**) Colour brightness L\* and absolute colour difference ΔE, (**b**) colour saturation C\* and colour tone h. Designation: a, b, c, d—homogeneous groups, the influence of: fibre preparations and A, B—kale at α = 0.05. The bar codes are explained in Table 4. To evaluate the effect of the addition of kale and DF preparation on the colour **Figure 1.** The influence of kale and fibre preparation addition on the colour parameters of multigrain bars: (**a**) Colour brightness L\* and absolute colour difference ∆E, (**b**) colour saturation C\* and colour tone h. Designation: a, b, c, d homogeneous groups, the influence of: fibre preparations and A, B—kale at α = 0.05. The bar codes are explained in Table 4.


changes of the bars, the absolute colour difference ΔE was calculated, and the colour of the control bars was set as a standard. The colour of the bars was clearly differentiated, because the absolute colour difference ranged from 6.3–13.7 (Figures 1a and 2), from the **Table 4.** The recipe of multigrain bars based on the weight of all ingredients, with the addition of kale and fibre preparation or powder of dried hibiscus flower.

The addition of kale in the amount of 20 and 30% did not significantly affect the brightness of the colour of the bars (Figure 1a). On the other hand, the type of added DF preparation significantly influenced the brightness of the colour of bars. The addition of blackcurrant, chokeberry and hibiscus DF caused a significant darkening of the colour bars. The bars with the addition of apple DF showed up to about 16% higher values of the L\* parameter. This dependence may be due to the colour of the DF preparations, apple DF is a light beige powder, while the remaining DF are dark red powders. Dark red discoloration of the preparations may be due to the presence of anthocyanin pigments.

**Figure 2.** Pictures of baked bars, respectively: 1—base (Control), 2—with apple fibre and 20% kale, 3—with apple fibre and 30% kale, 4—with hibiscus fibre and 20% kale, 5—with fibre from blackcurrant and 20% kale, 6—with chokeberry fibre and 20% kale. To evaluate the effect of the addition of kale and DF preparation on the colour changes of the bars, the absolute colour difference ∆E was calculated, and the colour of the control bars was set as a standard. The colour of the bars was clearly differentiated, because the absolute colour difference ranged from 6.3–13.7 (Figures 1a and 2), from the colour of the bars with apple DF to those with blackcurrant and chokeberry DF.

Color brightness L\*/

Absolute color difference,

ΔE

A, c A,c A,bc

A,a A,a

Type of bars

colour of the bars with apple DF to those with blackcurrant and chokeberry DF.

To evaluate the effect of the addition of kale and DF preparation on the colour changes of the bars, the absolute colour difference ΔE was calculated, and the colour of the control bars was set as a standard. The colour of the bars was clearly differentiated, because the absolute colour difference ranged from 6.3–13.7 (Figures 1a and 2), from the

Bd Ac Abc

Bd Bd

<sup>90</sup> C\* <sup>h</sup>

Type of bars

Ac

Color saturation C \* / Color tone h

<sup>a</sup> <sup>a</sup> <sup>b</sup>

b

bc

a

(**a**) (**b**) **Figure 1.** The influence of kale and fibre preparation addition on the colour parameters of multigrain bars: (**a**) Colour brightness L\* and absolute colour difference ΔE, (**b**) colour saturation C\* and colour tone h. Designation: a, b, c, d—homogeneous groups, the influence of: fibre preparations and A, B—kale at α = 0.05. The bar codes are explained in Table 4.

<sup>a</sup> <sup>a</sup> ab

L\* DE

<sup>c</sup> <sup>c</sup> <sup>b</sup>

The addition of kale in the amount of 20 and 30% did not significantly affect the brightness of the colour of the bars (Figure 1a). On the other hand, the type of added DF preparation significantly influenced the brightness of the colour of bars. The addition of blackcurrant, chokeberry and hibiscus DF caused a significant darkening of the colour bars. The bars with the addition of apple DF showed up to about 16% higher values of the L\* parameter. This dependence may be due to the colour of the DF preparations, apple DF is a light beige powder, while the remaining DF are dark red powders. Dark red discoloration of the preparations may be due to the presence of anthocyanin pigments.

**Figure 2.** Pictures of baked bars, respectively: 1—base (Control), 2—with apple fibre and 20% kale, 3—with apple fibre and 30% kale, 4—with hibiscus fibre and 20% kale, 5—with fibre from blackcurrant and 20% kale, 6—with chokeberry fibre and 20% kale. **Figure 2.** Pictures of baked bars, respectively: 1—base (Control), 2—with apple fibre and 20% kale, 3 with apple fibre and 30% kale, 4—with hibiscus fibre and 20% kale, 5—with fibre from blackcurrant and 20% kale, 6—with chokeberry fibre and 20% kale.

The higher colour saturation is perceived by consumers as the more "alive", while the lower the colour saturation, the more muffled it is, the closer to grey [23]. The values of the C\* parameter in the bars tested ranged from 5 to 18% (Figure 1b). The base bars (approx. 17.5%) had higher values of the C\* parameter, while the lower values of the bars with the addition of blackcurrant DF and chokeberry (approx. 5.5%). The addition of kale to the recipe significantly influenced the saturation of the colour of the bars, bars without the addition of kale had significantly higher values of the C\* parameter than bars with the addition of 20 or 30% kale. The type of DF preparation used also significantly influenced the colour saturation. Bars with the addition of chokeberry and currant DF were characterized by the lowest values, significantly higher values were found for bars with the addition of hibiscus, while the most "vivid" colour was characterized by the bars with the addition of apple DF.

The colour hue of h bars was also analysed, which informs how much a given colour differs from white [23]. All tested bars had the h parameter in the range of 40–75◦ , which corresponds to the range of colours from red to yellow. Low h values were characteristic for bars with the addition of chokeberry DF (approx. 40◦ ), higher bars with the addition of blackcurrant DF (approx. 50◦ ) and bars with the addition of hibiscus (approx. 57◦ ). The bars with the addition of apple DF had the highest colour shade values. The type of DF preparation used had a significant effect on the colour shade. The amount of kale addition used in the bar recipe was not significant. On the other hand, the addition of kale (20 and 30%) resulted in significantly higher h values than in bars without this additive.

#### *2.5. The Effect of the Addition of Kale and Fibre on the Antioxidant Content of Multigrain Bars*

The addition of kale significantly (Figure 3a) increased the antioxidant activity from 23.6 mM Trol/g d.m. for base bars without kale addition to 29.3 mM Trol/g d.m. for bars with 30% addition of kale. Bars with a 20% addition of kale and chokeberry DF (43.2 mM Trol/g d.m.) were characterized by high antioxidant activity, the value in these bars was nearly two times higher than in the based bars. It can be assumed that the DF preparation had a significant effect on this index.

**Figure 3.** The influence of the kale (factor I) and fibre preparations (factor II) addition in multigrain bars on the: (**a**)—DPPH antioxidant activity [mM Trol/ g d.m.] and (**b**)—polyphenol content [mg GA/100 g d.m.]. Designations: A, B—homogeneous groups (factor I), and a, b, c—(factor II) at α = 0.05. The bar codes are explained in Table 4. **Figure 3.** The influence of the kale (factor I) and fibre preparations (factor II) addition in multigrain bars on the: (**a**)— DPPH antioxidant activity [mM Trol/ g d.m.] and (**b**)—polyphenol content [mg GA/100 g d.m.]. Designations: A, B—homogeneous groups (factor I), and a, b, c—(factor II) at α = 0.05. The bar codes are explained in Table 4.

Korus [21] showed a high content of polyphenols in kale, at the level of 256–531 mg/100 g of fresh weight. Such a large range is influenced by the variety, growing conditions and maturity. This vegetable is therefore a very valuable source of these compounds, which are largely preserved in the bars. Green vegetables are rich in chlorophyll and often contain carotenoids, the colour of which is not always discernible due to the predominant chlorophyll. In fresh kale tested by Korus [21], the total content of chlorophyll was 81–165 mg/100 g and carotenoids 16.8– 34.2 mg/100 g. The control bars were characterized by a low content of chlorophyll (approx. 5.0 mg/100 g d.m.) and carotenoids (0.14 mg/100 g d.m.), and the addition of kale caused a 3–5 fold increase in chlorophyll content and 23–34 fold in carotenoids content (Figure 4a,b). This content increased with the increase in the percentage share of kale in In the study by Korus [21], the antioxidant activity of kale was at the level of 14.7–23.7 µM Trol/g, so lower than in the bars tested. In terms of dry matter content, the results would be more similar. Moreover, the values of this indicator depend on many factors related to both the preparation of the product and the method of determination [24]. Due to the large number of different compounds influencing the antioxidant activity (soluble in water or organic solvents), the obtained results may depend on the method of preparation of the extract. The bars enriched with chokeberry DF had the highest value of polyphenol content (Figure 3b), i.e., about 552.5 mg GA/100 g d.m., significantly lower bars with the addition of hibiscus powder, about 457.5 mg GA/100 g d.m. For bars with the addition of blackcurrant DF and bars with the addition of apple DF, this value was in the range 377–396 mg GA/100 g d.m.

the recipe. The DF preparations did not cause significant changes in the chlorophyll content in the bars. On the other hand, the content of carotenoids was significantly different, depending on the type of DF in the bar recipe (Figure 4b). The bars with the addition of blackcurrant DF contained significantly less (approx. 2.4 mg/100 g d.m.) carotenoids than Korus [21] showed a high content of polyphenols in kale, at the level of 256–531 mg /100 g of fresh weight. Such a large range is influenced by the variety, growing conditionsand maturity. This vegetable is therefore a very valuable source of these compounds, which are largely preserved in the bars.

the bars with the addition of chokeberry DF (approx. 4.1 mg/100 g d.m.). Green vegetables are rich in chlorophyll and often contain carotenoids, the colour of which is not always discernible due to the predominant chlorophyll. In fresh kale tested by Korus [21], the total content of chlorophyll was 81–165 mg/100 g and carotenoids 16.8–34.2 mg/100 g. The control bars were characterized by a low content of chlorophyll (approx. 5.0 mg/100 g d.m.) and carotenoids (0.14 mg/100 g d.m.), and the addition of kale caused a 3–5 fold increase in chlorophyll content and 23–34 fold in carotenoids content (Figure 4a,b). This content increased with the increase in the percentage share of kale in the recipe. The DF preparations did not cause significant changes in the chlorophyll content in the bars. On the other hand, the content of carotenoids was significantly different, depending on the type of DF in the bar recipe (Figure 4b). The bars with the addition of blackcurrant DF contained significantly less (approx. 2.4 mg/100 g d.m.) carotenoids than the bars with the addition of chokeberry DF (approx. 4.1 mg/100 g d.m.).

**Figure 4.** The influence of the kale (factor I) and fibre preparations (factor II) addition in multigrain bars on the: (**a**) chlorophyll content A + B [mg/100 g d.m.] and (**b**)—carotenoids content [mg/100 g d.m.]. Designations: A, B, C—homogeneous groups (factor I), and a, b, c—(factor II) at α = 0.05. The bar codes are explained in Table 4. **Figure 4.** The influence of the kale (factor I) and fibre preparations (factor II) addition in multigrain bars on the: (**a**) chlorophyll content A + B [mg/100 g d.m.] and (**b**)—carotenoids content [mg/100 g d.m.]. Designations: A, B, C homogeneous groups (factor I), and a, b, c—(factor II) at α = 0.05. The bar codes are explained in Table 4. (**a**) (**b**) **Figure 4.** The influence of the kale (factor I) and fibre preparations (factor II) addition in multigrain bars on the: (**a**) chlorophyll content A + B [mg/100 g d.m.] and (**b**)—carotenoids content [mg/100 g d.m.]. Designations: A, B, C—homogeneous groups (factor I), and a, b, c—(factor II) at α = 0.05. The bar codes are explained in Table 4.

#### *2.6. The Effect of the Addition of Kale and Fibre on the Sensory Quality of the Multigrain Bars 2.6. The Effect of the Addition of Kale and Fibre on the Sensory Quality of the Multigrain Bars 2.6. The Effect of the Addition of Kale and Fibre on the Sensory Quality of the Multigrain Bars*

The bars were positively assessed by potential consumers (Figure 5). Within the five sensory discriminants on a 10-point scale, apart from the overall quality of the chokeberry DF bars, all bars scored higher than 5.0, but not higher than 7.3 points. Most of the lower scores were given to bars with the addition of chokeberry DF (from 4.8 to 6.4 points). Most of the higher ratings were given to bars with P-Hibiscus, especially for texture and overall quality. The bars were positively assessed by potential consumers (Figure 5). Within the five sensory discriminants on a 10-point scale, apart from the overall quality of the chokeberry DF bars, all bars scored higher than 5.0, but not higher than 7.3 points. Most of the lower scores were given to bars with the addition of chokeberry DF (from 4.8 to 6.4 points). Most of the higher ratings were given to bars with P-Hibiscus, especially for texture and overall quality. The bars were positively assessed by potential consumers (Figure 5). Within the five sensory discriminants on a 10-point scale, apart from the overall quality of the chokeberry DF bars, all bars scored higher than 5.0, but not higher than 7.3 points. Most of the lower scores were given to bars with the addition of chokeberry DF (from 4.8 to 6.4 points). Most of the higher ratings were given to bars with P-Hibiscus, especially for texture and overall quality.

multigrain bars. Designations: a, b, c—homogeneous groups (the influence of type of bars) at α = 0.05. The bar codes are explained in Table 4. **Figure 5.** The effect of kale and fibre preparation or powder of dried hibiscus flower addition on sensory properties of multigrain bars. Designations: a, b, c—homogeneous groups (the influence of type of bars) at α = 0.05. The bar codes are explained in Table 4. **Figure 5.** The effect of kale and fibre preparation or powder of dried hibiscus flower addition on sensory properties of multigrain bars. Designations: a, b, c, d, e—homogeneous groups (the influence of type of bars) at α = 0.05. The bar codes are explained in Table 4.

#### **3. Discussion**

Currently, the daily intake of DF by most consumers is too low. This is due to the high degree of processing of many products. DF is supplied mainly from cereals and cereal products, seeds of legumes and fruits and vegetables. These products differ not only in their DF content, but also in the type of DF compounds. They are found in vegetables and cereals are grouped into water-soluble (pectins, gums) and water-insoluble (cellulose, lignin, some of the hemicellulose) [25]. All bars contained fibre at the level above 9% (Table 1), which allows them to be classified as products with a high fibre content. Bars tested by Márquez-Villacorta and Vásquez [26] with a composition of 4.12% oat bran; 10.04% of pineapple peel powder and 17.18% of quinoa flakes contained more DF (13.28%) and protein (11.37%), and the overall acceptance score was slightly higher (7.47 points). Epidemiological studies suggest that regular consumption of fruits and vegetables containing both DF and natural antioxidant compounds may reduce the risk of many chronic diseases [27,28]. The current diet also focuses on the caloric content of food. Vatankhah et al. [29] investigated the suitability of stevioside, a natural low calorie sweetener, as a replacement for sucrose in Iranian sweet bread. They showed that the replacement of sucrose in the amount of 50%, the physical, chemical and sensory properties of the bread were similar to the base product, but the calorific value was reduced by 11%. Ibrahim et al. [30] assessed the possibilities of using date fruit in the bar recipe and replacing honey with date paste. With regard to the use of date paste up to 70%, bars with a share of 50% were characterized by the highest overall acceptability.

Due to their sensory value, wide availability and convenience, snacks are popular and frequently consumed products [31]. Consumers like snacks very much, but also pay increasingly attention to what they eat and are aware of the issue of healthy eating. The current market trends force the food industry to introduce such products that can be part of a healthy and balanced diet, but also tasty and encouraging consumption [32].

Depending on the composition, various methods are used for the production of snacks, which can generally be divided into the so-called "cold" and "hot", respectively, without and with the use of increased temperature. The use of baking has benefits in terms of quality and product safety without the use of chemicals. During baking, starch gelatinization, browning reactions, changes in structure, surface properties and other mechanical behaviour of the bakery products occur [29], and the formation of their characteristic sensory properties. Based on the composition of the bars, they may be classified as snack bakery products. An additional advantage of choosing this method of producing bars was obtaining relatively soft products, such as bread. In order to increase the content of natural ingredients in multi-grain (wholegrain oatmeal, sunflower seeds, pumpkin seeds, flaxseed) bars, kale and DF were added to increase the high health-promoting potential of the bars, without chemicals. The taste of fresh kale does not encourage consumption, so an attempt was made to mask it. As a result, in the sensory evaluation, the addition of kale was less significant than the type of fibre. The addition of chokeberry to the fibre preparation was the least acceptable for most indications (4.8–5.2 points), but its flavour was distinguished (6.4 points).

The water activity (Aw) in the bars was high (0.857–0.953), but when analysing bars with the addition of 20% kale, all Aw were below 0.95. No pathogenic microorganisms develop in such a product. Many bakery products are characterized by higher water activity [33]. The shelf life of the bars is short. To retain all the value of the snacks and extend their freshness, for example, modified atmosphere packaging should be used. Water is an essential ingredient in many foods. Affects a number of processes and reactions that can reflect the quality and stability of food during storage. Whether certain reactions will occur is primarily determined by the state of the water, which is characterized by its activity [34]. From the point of view of water activity, food can be divided into [35] wet with water activity in the range of 0.90–1.00, medium water content—water activity in the range of 0.55–0.90 and low water content—water activity in the range 0.00–0.55.

In general, a stable food is considered the one with a water activity in the range of 0.07–0.35. However, the development of microorganisms is almost completely limited already at the water activity below 0.60 [36].

With the increase in the share of kale in the recipe, the water content in the bars increased. This was due to the increase in the water content in the bar recipe, which was caused by the addition of kale containing about 85% water [6]. This form of kale addition was justified by the possibility of enriching the bars. However, from the technological point of view, the addition of kale in the amount of 20% was sufficient. The type of DF preparation significantly influenced the water content in the tested bars (Table 2). The water content in bars with the addition of currant and chokeberry DF was the lowest (about 27%), while the bars with the addition of hibiscus preparation (about 35%) had a significantly higher water content. This may indicate the different sorption properties of the DF preparations used. According to Miastowski et al. [36], water binding is one of the most important features of DF preparation. However, their large diversity in terms of the presence of DF compounds, depending on the source of origin, has a large impact on the degree of water binding in bars. This property also depends on the degree of micronization and the particle size composition. Therefore, the use of various preparations resulted in different values of water activity. This is advantageous in the manufacture of bars that should have the desired texture.

The water content has influenced the mechanical properties of bars. The correlation coefficient for the water content and compression force was about −0.92 (*p* = 0.0098), and for the compression work −0.94 (*p* = 0.0053). This proves a strong negative correlation between these properties and the water content.

The TPA test is used to test the texture of food based on indicators that reflect the consumer's perception of the chewing experience [37]. For multigrain bars about 2 cm thick, softness is required and those up to about 1 cm thick can be crunchy. In the case of thin ones, especially those with increased carbohydrate content, one should aim to obtain a glassy (amorphous) state. The structure of the bars is influenced by the method of their production, especially the temperature value. In the research by Nikmaram et al. [38], the optimal conditions for the production of extrudates depended on the amount of sesame seeds added and the temperature of the process. The addition of kale and DF preparations had a statistically significant effect on the parameters of the texture profile. The higher content of sesame seed, incorporated into corn expanded extrudates, increased the hardness of the extrudates, possibly due to the content of fat, protein and fibre [38]. Similarly relationships Kowalczewski and Ivanišová [39] showed that the addition of fruit to the muffin recipe had a significant impact on the parameters of the texture profile. According to Wójtowicz and Baltyn [40], the hardness of snacks should be as low as possible, as it proves the fragility of these products. In the study by Kubiak and Dolik [41], the result for apples was 62.02 N, Wójtowicz and Balatyn [40], potato pancakes were characterized by a hardness in the range of 102–106 N, while in Heo et al. [42], muffins enriched with DF were characterized by a hardness of 412–491 N. In the research by Kubiak and Dolik [41], the bread was characterized by elasticity at the level of 0.94. The tested bars were characterized by almost two times lower elasticity. This may indicate their compact structure.

In a properly functioning organism, it is necessary to ensure a balance in so-called redox processes. If reactive oxygen species are not effectively quenched, it may lead to oxidative stress [39]. To prevent the formation and protect the body against reactive oxygen species, one should eat food rich in antioxidant compounds [43]. Polyphenols are compounds synthesized by plants. Several thousand compounds belong to the group of phenol compounds, but they all have one thing in common, which is the antioxidant properties. The content of polyphenols was strongly and positively correlated with the antioxidant activity (r = 0.92). The use of kale addition caused changes in the content of polyphenolic compounds (Figure 3b). In the case of 30% addition of kale (422 mg GA /100 g d.m.), the content of polyphenols was significantly higher than for bars without addition (334.6 mg GA/100 g d.m.). The amount of polyphenolic compounds largely

depended on the DF preparation used. The bars enriched with chokeberry DF had the highest value, i.e., about 552 mg GA/100 g d.m. In the study by Nawirska et al. [44], chokeberry pomace was also characterized by the highest polyphenol content among the tested fruits, and significantly lower values were obtained for blackcurrant. Biega´nska-Marecik et al. [45] showed that kale has one of the highest values of antioxidant activity. The addition of frozen and freeze-dried kale on beverages based on apple juice resulted in a two- and three-fold increase in antioxidant activity, respectively. Murugesan et al. [1] showed that the antioxidant capacity of kale leaf ethanol extract was 62.9% (DPPH\*), and GC-MS chromatographic analysis included profiles of more than 17 major phytochemicals in the extract. Additionally, Satheesh et al. [46] reported that there has been a growing trend in recent times to include more green leafy vegetables in the human diet, and kale has great potential for use in a variety of food and nutritional applications. Kale has been shown to have the nutritional and anti-nutritional components of kale, with research showing its multiple health benefits.

The anthocyanin pigments present in the bars had a positive effect on the antioxidant activity, but they could cause colour changes of the product. Colour is a parameter that has a large impact on the perception of food by consumers, as it can reflect the quality of food products. According to Kowalczewski and Ivanišová [39], these changes may not be accepted by consumers for some products. Moreover, in the case of bars, it can be noticed that the addition of chokeberry DF preparation, which had a positive effect on the content of polyphenols, translated into high oxidative activity and caused colour changes (Figure 1a,b). In addition, the bars were darker because the L\* values were lower (Figure 1a). However, no significant differences were observed in consumer assessments regarding the colour of the bars (Figure 5).

Carotenoids and chlorophylls are plant pigments that give colour to vegetables and fruits, they are located in chloroplasts. Carotenoids are responsible for the red, orange and yellow colours. They are considered one of the strongest antioxidants and are credited with the ability to extinguish free radicals. They are also precursors of vitamin A. Chlorophylls are credited with bacteriostatic and anti-inflammatory properties, supporting the removal of carcinogenic toxins and with antioxidant properties. They give plants and products a characteristic green colour [47].

According to Karwowska et al. [47], fresh kale is characterized by the content of chlorophyll A and B at the level of 904.5 mg/100 g d.m. Kale added to the recipe increased the content of chlorophyll in the bars. This content increased with the increase in the percentage share of kale in the recipe. The addition of kale in the amount of 20 and 30% resulted in a 3.6 and 5.3-fold increase in their chlorophyll content, respectively, in comparison with the control samples (approx. 5.0 mg/100 g d.m.). The DF preparations of blackcurrant, chokeberry and hibiscus did not cause significant changes in the chlorophyll content in the bars. However, in the bars with the addition of chokeberry and hibiscus DF, the content of chlorophyll was lower by 10–14% than for those with chokeberry DF.

The addition of kale enriched bars with carotenoids. As in the case of chlorophyll, the addition of kale caused a greater increase in the carotenoid content. The bars with 20% additive (3.2 mg/100 g d.m.) contained about 23 times more dye than the base bars (0.136 mg/100 g d.m.). Bars with more kale contained the most carotenoids (4.7 mg/ 100 g d.m.). According to Karwowska et al. [47], it contains carotenoids in the amount of about 175 mg/100 g d.m. The effect of the DF preparation used on the content of dyes was also observed. The bars with the addition of blackcurrent DF contained significantly less carotenoids than the bars with the addition of chokeberry DF. Aronia contains 140–230 mg of carotenoids per 100 g of d.m., while blackcurrant only 20–40 mg/100 g of d.m. [48]. This translated into the final content of these dyes in the baked bars.

#### **4. Materials and Methods**

#### *4.1. Material*

The materials for the research were multi-grain bars with the addition of fresh kale prepared according to the established recipe (Table 4). The type of DF added and the percentage of fresh kale (*Brassica oleracea* L. var. *acephala*) added was variable in the recipe. The raw materials were purchased in a large-area store, while the DF preparations were obtained directly from the producer (GreenField, Poland).

#### *4.2. Experimental Procedure*

#### 4.2.1. Preparation of BARS

The dry ingredients were ground in a grinder (Bosch MKM6000) for 20 s. Ground flaxseed was poured over with hot water to gel. The kale was ground in a Thermomix TM 31 device (Vorwerk Ltd., Wroclaw, Poland) for 10 s, speed of rotation—level 7. Then, all ingredients were combined and mixed for about 2 min. After receiving the mass, it was placed in rectangular form with dimensions of 100 × 40 × 20 mm. The formed bars were baked or dried in three different variants.

#### 4.2.2. Baking

Baking was carried out in an electric Piccolo oven (Winkler Wachtel Ltd., Wroclaw, Poland) for 25 min. The temperature of the lower and upper chamber of the furnace was 180 ◦C.

#### *4.3. Analytical Methods*

#### 4.3.1. Determination of Dry Matter Content

The dry matter content of the bars was determined by drying in a laboratory dryer (WAMED SUP-65 WG, Warsaw, Poland) at 130 ◦C for 1 h. The vessels with/without samples were weighed on an analytical balance (ME54E/M, Metler, Warsaw, Poland) with an accuracy of 0.001 g. The measurement was performed in duplicate.

#### 4.3.2. Determination of Water Activity

Water activity was determined with an AQUALAB CX-2 device (Decagon Devices Inc. Pullman, WA, USA). Measurements were carried out at the temperature of 23 ± 1 ◦C. The measurement was performed in duplicate, the final result was the mean of the measurements.

#### 4.3.3. Colour Parameters

The colour of the bars was measured with the Konica Minolta CR-300 colorimeter (standard observer CIE 2◦ , illuminat D65, measuring gap 8 mm) in the CIE Lab system. The measurement was performed in 5 replications. The mean of the measurements was taken as the result.

#### 4.3.4. Examination of Bars Structure

#### Compression Test

The mechanical properties were tested in a TA-HD plus texturometer (Stable Micro Systems, Godalming, UK). The compression test was performed with a 75 mm diameter head. Bars with dimensions of 25 × 40 × 20 mm were used for the measurement. The head speed was 1 mm/s. The samples were compressed to 50% of their height. The measurement was performed in 10 replications. The compression test was performed for the bars 4 h after the end of drying or baking. On the basis of the test, the compression work calculated as the product of the half of the area under the deformation curve and the head travel speed were determined.

Texture Profile Analysis (TPA) Test

The texture profile test was performed with a TA-HD plus texturometer (Stable Micro Systems, Godalming, UK). The measurement was performed 4 h after the end of drying or baking. Bars with dimensions of 25 × 40 × 20 mm were used for the measurement. The tested samples were compressed twice to about 50% of the original height. The head speed was 1 mm/s. The measurement was performed in 10 replications. On the basis of the test, the mechanical determinants of texture, such as hardness, elasticity, cohesiveness, gumminess, and chewiness, were determined as follows:


#### 4.3.5. Chemical Determinations

Chemical determinations were carried out in an accredited laboratory at the Institute of Agriculture and Food Biotechnology—State Research Institute in Warsaw, Poland. All determinations were performed at least in duplicate.

#### Nutritional Value

Determination of the nutritional value, i.e., protein, fat, ash, DF and carbohydrates, was carried out in the accredited laboratory of the Institute of Agricultural and Food Biotechnology—National Research Institute in Warsaw Poland. All determinations were performed at least twice.

#### Determination of Protein Content

Total nitrogen content was determined by the reference titration method (Kjeldahl) and converted into total protein content, taking into account the nitrogen to protein conversion factor 6.25 according to PN-EN ISO 20483: 2014 standard.

The principle of the method consists in converting organic nitrogen compounds contained in a dry sample of ammonium sulphate with concentrated sulfuric acid in the presence of a catalyst, basifying the solution, distilling and titrating ammonia bound in boric acid with the addition of indicators with sulfuric acid.

#### Determination of Fat Content

The fat content was determined in accordance with the PN-A-79011-4: 1998 standard. The principle of the method is based on the extraction of fat from a dry sample under predefined conditions using petroleum ether by means of a Soxhlet apparatus, and then weighing the residue of the sample after complete evaporation of the solvent.

#### Determination of Ash Content

The ash content was determined by the gravimetric method after the samples were incinerated according to the PN-EN ISO 2171: 2010 standard. The principle of the method is based on incineration of the dry sample (pre-dried) at the temperature of 900 ◦C and determination of the inorganic residues after ashing by weight.

#### Determination of Total Dietary Fibre Content

The total dietary DF content, including the soluble and insoluble fractions was determined by the gravimetric method after prior enzymatic hydrolysis of the samples using the Megazyme Total Dietary Fibre Kit (Bray, Bray Business Park, Co. Wicklow, A98 YV29, Ireland).

Calculation of Carbohydrates Content, Including Sugars

The carbohydrate content (CC) in g/100 g d.m. was calculated from the formula:

$$\text{CC} = 100 - \text{(H} + \text{A} + \text{F} + \text{P} + \text{DF)} \tag{1}$$

where:

H—humidity of the sample, [g/100 g d.m]; A—ash content of the sample, [g/100 g d.m]; F—fat content of the sample, [g/100 g d.m]; P—protein content of the sample, [g/100 g d.m]; DF—dietary fibre content of the sample, [g/100 g d.m].

Determination of the content of individual sugars, including: fructose, glucose, disaccharides (sum of sucrose and maltose) was performed using high performance liquid chromatography (HPLC) with refractometric detection of sugars contained in the aqueous solution obtained from the sample. The result of the sugar content was given as the sum of individual sugars [g/100 g d.m.].

#### Calculation of Energy Value

The energy value (EV) of the product (bars) was calculated on the basis of the energy content of protein (1 g = 4 kcal or 17 kJ), carbohydrates (1 g = 4 kcal or 17 kJ), fat (1 g = 9 kcal or 37 kJ) and dietary fibre (1 g = 2 kcal or 8 kJ) contained in them. The energy value in kcal/100 g and kJ/100 g of product was calculated [49]:

$$\text{EV [kcal / 100 g]} = (\text{P} + \text{CC}) \cdot 4 + \text{F} \cdot 9 + \text{DF} \cdot 2 \tag{2}$$

$$\text{EV [k]} / 100 \text{ g} \text{]} = (\text{P} + \text{CC}) \cdot 7 + \text{F} \cdot 37 + \text{DF} \cdot 8 \tag{3}$$

where:

P—protein content of the sample [g/100 g]; CC—carbohydrates content of the sample [g/100 g]; F—fat content of the sample [g/100 g]; DF—dietary fibre content of the sample [g/100 g].

#### Determination of DPPH Radical Scavenging Activity

The antioxidant activity (AA) was determined using the spectrophotometric method with the DPPH radical based on the method of Urba ´nska at el. [50] and Wong at el. [51]. For the preparation of samples, 2.4 mL of DPPH methanolic radical solution (60 µM) was used and 100 µL of acetone extract of the samples (the extract was prepared in the same way as for the determination of carotenoids/chlorophylls) was added. The samples were mixed and incubated at room temperature for 30 min in the dark. After this time, the absorbance was measured at the wavelength λ = 515 nm against the blank. The acetone solution and DPPH solution were collected for the control sample. The blank was a sample containing of methanol and of 80% acetone.

The antioxidant activity (quenching/scavenging capacity) of the DPPH radical (% inhibition) was calculated:

$$\% \text{inhibition} = \frac{A\_0 - A\_1}{A\_0} \cdot 100 \tag{4}$$

where:

*A*1—absorbance of the DPPH radical with acetone extract from the sample;

*A*0—absorbance of the DPPH radical with acetone (control sample).

When the calculated inhibition was greater than the 95% value, the sample was diluted with 80% (*v*/*v*) acetone solution so that the absorbance value was linear over the range of the analysed concentrations.

The antioxidant activity (AA) based on the DPPH free radical scavenging ability of the extract was expressed as mM Trolox per 1 g of dry matter (d.m.) of the sample.

#### Determination of Total Polyphenol Content by the Folin–Ciocaletau Method

The total polyphenol content was determined by spectrophotometric method with the use of the Folin–Ciocaleteu reagent, which consisted of a coloured reaction of polyphenolic compounds with this reagent [50]. To the test tube was added 15% sodium carbonate (0.5 mL), distilled water (8.9 mL), acetone extract of the sample (0.5 mL; the extract was obtained in the same way as for the determination of carotenoids/chlorophylls, chapter 3.3.6), and 100 µL of Folin–Ciocalteu reagent. The sample was then mixed and incubated for 45 min in the dark (room temperature). After this time, the absorbance was measured at the wavelength λ = 765 nm against the blank. When the measured absorbance of the sample was greater than 0.650 value, the sample was diluted with 80% (*v*/*v*) acetone solution. The determination was performed in duplicate. The content of total polyphenols was expressed as mg of gallic acid (GA) per 100 g dry matter (d.m.) of the sample.

#### Determination of Carotenoids and Chlorophyll A and B Content

Determination of carotenoids and chlorophyll content in the bar samples was performed using the BECKMAN DU-530 spectrophotometer (Beckman, UK). The samples were milled with a Sencor grinder to obtain the extract. An 80% (*v*/*v*) acetone solution (25 mL) was added to the weighed sample (about 1.0 g). The samples were homogenized for 30 s at a speed of 13,500 rpm in an ULTRA-TURRAX T25 basic homogenizer (IKA-WERKE, Germany). Then, the obtained homogenate was centrifuged in a laboratory centrifuge MPW 375 (MPW-Med-Instruments, Poland) for 3 min at a speed of 10,000 rpm. The measurements were made for chlorophyll A at wavelengths λ = 663 nm, for chlorophyll B λ = 647 nm, and at λ = 470 nm for carotenoids with the blank, which was an 80% (*v*/*v*) acetone solution. When the measured absorbance of the sample was greater than 0.900 in value, the sample was diluted with an 80% (*v*/*v*) acetone solution. The determination was performed in duplicate. The content of carotenoid pigments, chlorophyll A and B in the acetone extract was calculated from the equations [52]:

$$\mathcal{C}\_{\mathbb{C}} = \frac{1000 \cdot A\_{470} - 1.82 \cdot \mathcal{C}\_A - 85.02 \cdot \mathcal{C}\_B}{198} \tag{5}$$

$$\mathbf{C}\_{A} = 12.25 \cdot A\_{663} - 2.79 \cdot A\_{647} \tag{6}$$

$$\mathbf{C}\_{B} = 21.50 \cdot A\_{647} - 5.10 \cdot A\_{663} \tag{7}$$

$$\mathbf{C}\_{A+B} = 7.15 \cdot A\_{663} + 18.71 \cdot A\_{467} \tag{8}$$

where:

*CC*—content of carotenoids in acetone extract [µg/mL]; *CA*—content of chlorophyll A in acetone extract [µg/mL]; *CB*—content of chlorophyll B in the acetone extract [µg/mL]; *CA+B*—content of chlorophyll (total A + B) in the acetone extract [µg/mL]; *A*663—absorbance of acetone extract measured at wavelength λ = 663 nm; *A*647—absorbance of acetone extract measured at a wavelength of λ = 647 nm; *A*470—absorbance of acetone extract measured at wavelength λ = 470 nm.

The content of chlorophyll or carotenoid dyes in the sample was calculated in mg per 100 g dry matter (d.m.). All determinations were performed at least in duplicate.

#### Sensory Evaluation

The sensory evaluation was performed by a team of 30 unqualified people, aged 18 to 45, using a 10-point scale. The evaluators were instructed on how to evaluate the selected discriminants such as taste, colour, smell, texture, and overall desirability (Table 5).


**Table 5.** Quality features to be assessed and their characteristics.

#### *4.4. Statistical Analysis*

The statistical analysis of the obtained results was performed with the use of Microsoft Excel and STATISTICA 13 PL programs. To determine the effect of the amount of curly kale and the addition of DF on the selected indicators, a one- or two-factor analysis of variance and Tukey's HSD test were performed to determine homogeneous groups (post hoc test). Pearson's correlation was also performed to investigate the relationship between the selected indicators.

#### **5. Conclusions**

The addition of kale and DF preparations had a beneficial effect on the physicochemical, sensory and pro-healthy properties of snacks. In the production of bars, DF also played a technological role, enabling the appropriate consistency of the mix before baking and the texture of the final products. Multigrain raw materials are characterized by a high content of DF. Multigrain bars with the addition of kale and DF preparations can be a valuable source of both DF fractions, antioxidant compounds, as well as fat, protein, vitamins and minerals. They can be an offer of snacks for people struggling with health problems, as well as for healthy people who are looking for tasty and valuable products.

**Author Contributions:** Conceptualization, H.K., A.Z. and A.I.; methodology, H.K., A.S. and J.K.; software, H.K. and A.I.; validation, H.K., J.K., E.D., S.G., A.C. and A.S.; formal analysis, H.K., J.K., S.G. and A.M.; investigation, A.Z. and E.M.; resources, A.Z.; data curation, H.K. and E.M.; writing original draft preparation, H.K., A.C. and A.I.; writing—review and editing, J.K., A.M., E.D., A.I. and E.M.; visualization, E.M.; supervision, H.K.; funding acquisition, H.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was financed under the Warsaw University of Life Sciences (WULS) Support System (decision No. SMPB 7/2020).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

**Sample Availability:** Samples are available from the corresponding author.

#### **References**


*Article*
