3.1. Yields and composition of the different SWE fractions
Figure 1 shows the flow chart of the brewer spent grain (BSG) or beer bagasse fractionation throughout the SWE, giving rise to soluble extracts (E) and insoluble residues (R) at each temperature. The latter were submitted to a bleaching step to purify cellulose while the extracts were freeze-dried to obtain extract powders. Images of the different obtained products, together with the obtained mass yield of each process step are also shown in Fig. 1. Previously to the SWE, a defatting step of the BSG yielded around 8% oil from the dried bagasse, being this value in agreement with that reported in the literature (Faulds et al., 2008; Del Rio et al., 2013). The main beer bagasse lipid compounds has been reported, these being triglycerides (55–67%), free fatty acids (18–30%), such as palmitic, oleic and linoleic acids and free steroids (5%), such as sitosterol and campesterol. These lipid compounds have a wide range of nutraceutical, pharmaceutical and cosmetic applications (Del Rio et al., 2013).
The values of mass yields of solid extracts (E-110, E-130, E-150 and E-170) and dried residues (R-110, R-130, R-150 and R-170) of the SWE process performed at different temperatures, shown in Fig. 1, indicate that the extraction yield increased (from 7–41%) when the extraction temperature rose from 110°C to 170°C. This is mainly explained by the changes in the water solvent properties when the temperature increased, which reduces the strength of hydrogen bonds and leads to an important reduction in dielectric constant, this becoming closer to the dielectric constant value of some organic solvents, such as methanol (ɛ = 33) or ethanol (ɛ = 25) (Carr et al., 2011; He et al., 2012). The sum of both yields (extract and residue) at a given temperature closed the mass balance, thus indicating a low mineralisation degree of the organic matter present at the processing conditions used.
The TGA curves of the defatted bagasse (DB), extracts and residues obtained after SWE at the different temperatures is shown in Fig. 2a, together with de derivative curves (Fig. 2b). The DB presented three main degradation steps: the first mainly corresponding to the loss of bound water; a second step associated with the degradation of polysaccharides with different thermostability such as hemicelluloses (150–350 oC ), celluloses (275–350 oC ) and a part of lignin (160–900 oC ), and the third one, related to the degradation of residual lignin and secondary metabolites from the previously thermo-degraded compounds, as previously described by other authors for lignocellulosic biomass (Freitas et al., 2023; Carichino et al., 2023). The major weight losses took place between 225 and 625 oC (80%), in line with the lignocellulosic nature of this residue, in agreement with the results obtained by other authors (Ortiz et al., 2020; Carichino et al. 2023). Very similar TGA patterns were obtained for every lignocellulosic residue. Nevertheless, it is remarkable that the highest extraction temperatures (150 and 170 oC) gave rise to the samples with the highest peak temperatures (temperature of the maximum degradation rate), which indicates that these were the most enriched in cellulose that shows peak temperature between 330–350 oC (Zhang et al., 2012). In contrast, the TGA and DTGA curves of the extracts revealed a more complex compositional profile, exhibiting several thermodegradation steps. The extract obtained at the highest temperature showed a higher proportion of compounds that degrade at higher temperature according to a greater extraction of polymeric components, such as hemicellulose or lignin. A higher final mass residue was also observed in comparison with the untreated sample (defatted bagasse), which can be due to the extract enrichment in minerals or formation of degraded organic matter from the soluble compounds.
Some compositional differences in the extracts and residues obtained at each temperature can be observed in Tables 1 and 2.
Table 1
Chemical composition of defatted beer bagasse (DB) and insoluble fractions after SWE at different temperatures.
Sample | Extractive (%) | Protein (%) | Ash (%) | Lignin* (% ) | Cellulose (%) | Hemicellulose (%) |
DB | 13,1 ± 0,5 ᵃ | 22 ± 2a | 3,71 ± 0,01 ᶜ | 9,5 ± 1,6 ᵃᵇ | 17 ± 2 ᵃ | 17,9 ± 0,6 d |
R-110 | 20,9 ± 1,5 ᵇ | 28 ± 2 ᵇ | 3,12 ± 0,11 ᵇ | 9,1 ± 0,4 ᵃ | 16 ± 2 ᵃ | 15 ± 2 ᶜ |
R-130 | 21,79 ± 0,05 ᵇ | 26,2 ± 0,2 ᵃᵇ | 2,79 ± 0,04 ᵇ | 11,56 ± 0,08 ᵇ | 20 ± 2 ᵃ | 14,9 ± 1,2 ᶜ |
R-150 | 34,5 ± 0,9 ᵈ | 26 ± 2 ᵇ | 2,27 ± 0,11 ᵃ | 14,6 ± 0,5 ᶜ | 21 ± 2 ᵃ | 7,8 ± 1,2 ᵇ |
R-170 | 28 ± 3 ᶜ | 35,7 ± 0,2 ᵃᵇ | 3,1 ± 0,2 ᵇ | 29,9 ± 0,3 ᵈ | 30 ± 3 ᵇ | 2,01 ± 0,08 ᵃ |
a,b,c..Different superscript letters (a,b,c) in the same column indicate significant differences (p < 0.05) |
* acid insoluble lignin |
In Table 1, the total (water and ethanol) extractive content, protein, ashes, cellulose, hemicellulose and acid insoluble lignin contents of the DB can be observed, together the values obtained for the different SWE solid residues. The obtained values for raw brewer´s spent grain were within the range previously reported (around 16–22% for cellulose, 24–28% hemicellulose and 9–27% total lignin) (Verni et al., 2020; Alonso-Riaño et al., 2023; Qazanfarzadeh et al. 2023). In the extraction residues, the hemicellulose content was very low at temperatures greater than 150 oC, in accordance with the selective dissolution of hemicellulose under the subcritical water conditions (Cocero et al., 2018; Ruthes et al., 2017). Thus, the hemicellulose started to be removed from the beer bagasse matrix when using temperatures greater than 130 oC, reaching very low values at 170 oC (2%). At these temperatures, the lignin content significantly increased, which confirmed that this fraction of the biomass was not released under SWE. as it has been previously observed by other authors working with BSG (Alonso-Riaño et al, 2023). On the other hand, the increment in the cellulose significantly increased (p < 0.05) in the residues treated at the highest temperature (R-170) in comparison with the untreated DB.
The insoluble-acid lignin in the solid residues accounted for the 87, 101, 99 and 160% of the total lignin in the raw material for 110, 130, 150 and 170ºC, respectively. Thus, the obtained lignin values are surely overestimated as the outcome of such gravimetric analysis is highly disturbed by the presence of non-lignin acid-insoluble material, e.g. proteins (Erven et al., 2017). The corrected lignin (calculated by substracting the protein content) was not given because in most cases, negatives values were obtained. According to Alonso-Riaño et al. (2023), changes in the lignin structure took place during the SWE such as condensation reactions and structural alterations.
In both extract and residue fractions, the greater mass loss in TGA curves was observed for the temperature range of 200–700 oC, where the lignin is mainly degraded, in line with the formation of secondary metabolites from the previously thermo-degraded compounds. This thermal degradation behaviour agreed with that found in the literature for other lignocellulosic residues (Freitas et al., 2023).
So, the application of SWE led to a selective fractionation of DB, giving rise to aqueous extracts richer in different compounds of lower molecular (sugars, phenolic compounds and minerals) and polysaccharide and lignin-rich insoluble residues.
Table 2
Total phenolic content (TPC), antioxidant activity (EC50), protein content and ashes of the aqueous extracts (E) obtained from SWE process at different temperatures. (mean values ± standard deviation).
| E-110 | E-130 | E-150 | E-170 |
% Protein (db) | 15,1 ± 0,1 a | 16,6 ± 0,2 a | 22,4 ± 1,2 b | 28,7 ± 0,6 c |
%Ashes (db) | 1,54 ± 0,06 c | 1,5 ± 0,1 c | 1,20 ± 0,04 b | 0,46 ± 0,01 a |
TPC1 (mg GAE/g extract) | 16,8 ± 0,1a | 22 ± 2a | 17,91 ± 0,07b | 59,1 ± 0,2c |
TPC2 (mg GAE/g DB) | 1,27 ± 0,08a | 3,2 ± 0,3ab | 6,34 ± 0,02ab | 24,18 ± 0,08b |
EC501 (mg extract/mg DPPH) | 15 ± 3a | 19 ± 2a | 48 ± 4b | 71,1 ± 0,4c |
MIC (mg/ml) against L.innocua | 264 | 198 | 168 | 80 |
MIC (mg/ml) against E. Coli | 234 | 204 | 162 | 140 |
a,b,c…different superscripts in the same row indicates significant differences among extracts (p < 0.05)
The ash content of the DB, extracts and residues (shown in Tables 1 and 2) showed that minerals were mainly present in the insoluble residues, whereas small amounts were released to the extracts. The value obtained per DB was in the range of the ash content reported by other authors for beer spent grain (2–5 g/100 g dry DB), being the most abundant constituents phosphorous, magnesium, calcium and potassium (Ortiz et al., 2020).
In Table 2, the protein content of DB is also shown (around 22%), being this value in the range of previously reported values for beer bagasse (Rodriguez et al., 2023; Alonso-Riaño et al.,2021) considering a fat-free basis. The partition of the protein content during the SWE gave rise to greater content in the insoluble residues, thus suggesting a low solubility of the bagasse proteins under the used water subcritical conditions, especially at the lowest SWE temperatures. At 110 oC, the 95% of the total protein remained in the insoluble residue, this percentage decreasing to around 53% at 170 oC. These proteins are extracted and/or hydrolysed during the thermal treatment, leading to peptide chains of different sizes or free amino acids or even amino acid decomposition, especially at high temperatures, producing different carboxylic acids and other nitrogen containing compounds such as ethanolamine (Trigueros et al., 2023; Rogalinski et al, 2005). Therefore, SWE treatment of beer bagasse can be considered as an efficient extraction method to recover the protein fraction of the BSG generated in the beer industry, the maximum recovery of solubilised protein in the SWE extracts being of 47% at 170 oC.
3.2. Functional properties of the SWE extracts: Antioxidant and Antibacterial properties.
In Table 2, the Total phenolic content (TPC) and antioxidant activity of the aqueous extracts obtained from SWE process at different temperatures is displayed. The antioxidant activity of the different extracts was determined through the total phenolic content by the Folin–Ciocalteu method and the EC50 parameter with DPPH radical, which quantifies the amount of extract needed to reduce the initial concentration of the radical up to 50%. The TPC determined in the solid extracts (TPC1) was also referred per mass unit of defatted bagasse (TPC2). The TPC values increased from 16 to 59 mg GAE/g dried extract as the extraction temperature rose. Similarly, Rahman et al (2021) observed that BSG aqueous extract obtained at 160 oC showed highest TPC values than that obtained at 100 and 140 oC, the main phenolics compounds being flavan-3-oles, hydroxycinnamic acids (such as chlorogenic, gallic, protocatechuic, ferulic and p-coumaric acids) and flavonols. The found TPC values in defatted bagasse (7,57 mg GAE/g DB) are in the range of those reported by other authors (0.89-15 mg GAE/g sample), depending on the solvent and extraction method used (Santi et al., 2018). Nevertheless, when expressed per mass unit of DB, TPC2 values in the extracts were lower than the TPC value of DB, except for the extract obtained at the highest temperature (E-170). During the SWE extraction, an increment in the TPC content occurred as temperature increases due to the promotion of hydrolysis of lignin/phenolics-carbohydrate complexes, fostering the decomposition of these structures and releasing phenolic acids. Likewise, the neoformation of antioxidant compounds under severe SWE conditions has also been described (Plaza et al. 2010a,b). This neo-formed antioxidant compounds could be also quantified as phenols by the unspecific Folin-Ciocalteu reagent. These compounds are formed through Maillard and/or caramelization reactions, producing 5-HMF and sugar condensation compounds, and exhibit different bioactivities, including antioxidant activity (Trigueros et al., 2023). On the other hand, the thermal degradation of the phenolic compounds at high temperatures could also occur. Specifically, flavonoids, one of the main phenolic compounds in the beer bagasse, are highly thermo-sensitive. Therefore, the extent of the different phenomenon occurred during SWE, depending on the composition of each natural matrix, will determine the final content and nature of phenolics in the extracts. Thus, the marked increment in the TPC observed at 170 oC could be attributed to the high progress of the hydrolytic phenolic release, compared to the potential degradation ratio, as well as to the neo-formation of higher amounts antioxidant species.
The antioxidant capacity, measured throughout the EC50 values are also shown in Table 2. This value increased when the temperature rose, thus indicating a decrease in the radical scavenging capacity of the extracts. This decrease in the antioxidant capacity when the temperature rose, despite the promotion of higher phenolic content, can be attributed to the different phenolic profile in each extract with different radical scavenging capacity.
The antimicrobial potential of the DB extracts was also studied against the Gram-negative E.Coli bacteria and the Gram positive L. Innocua, which are well-known pathogenic microorganisms responsible for food poisoning. The minimal inhibitory concentration (MIC values) of the extracts with both bacteria were determined and shown in Table 2. The antibacterial effectiveness increased when the extraction temperature rose, being the gram positive bacteria (Listeria) more sensitive to the extracts. The E-170 MIC value for E.Coli was similar to that found for SWE extract of almond peel (90 mg/ml) obtained at 160 oC (Freitas et al., 2023). Barbosa-Pereira et al. (2014) also reported the antimicrobial efficiency of the polyphenols from the brewery waste stream against S. Aureus, L. monocytogenes, Salmonella spp. and E.Coli bacteria, while ferulic and caffeic acids and flavonoids being the main responsible for the observed antimicrobial activity. The obtained results indicate that the SWE extracts from beer bagasse are excellent candidates to be used as antioxidants or antilisteria compounds in food preservation or in the pharmaceutical sector.
3.3. Bleaching of the Extraction Residues
The extraction residues (R-110, R-130, R-150 and R-170) were bleached to recover the cellulose fraction, as they can be used for different applications in the material developing and pharmaceutical sectors. The bleaching treatment was carried out using a greener bleaching agent than the usual chlorine bleaches, to minimize the environmental impact of the process. Thus, the insoluble fractions were submitted to four successive 1 h cycles with 4% H2O2 solution at pH 12. To evaluate the efficiency of the process, the white index (WI) and the yield of the process was determined in cycle for the different samples (Fig. 3). As expected, the application of four successive bleaching cycles significantly decreased yield and increased the WI values, in accordance with the progressive purification of cellulose in each cycle.
Table 3
Chemical composition (%wt) of the insoluble fractions subjected to the four bleaching cycles with 4% hydrogen peroxide.
Sample | Ashes (%) | Lignin (%) | Protein (%) | Cellulose (%) | Hemicellulose (%) |
BR-110-1C | 6,96 ± 0,09a,1 | 14,69 ± 0,03a,1 | 11,3 ± 0,5a,3 | 52 ± 2a,1 | 41 ± 5a,1 |
BR-110-2C | 5,9 ± 0,3b,1 | 15,9 ± 0,8a,2 | 8,2 ± 0,3b,2 | 44 ± 2a,1 | 33 ± 4b,1 |
BR-110-3C | 6,20 ± 0,08b,1 | 13,88 ± 1,09a,2 | 5,1 ± 0,2c,2 | 57 ± 5a,1 | 30 ± 3c,1 |
BR-110-4C | 6,2 ± 0,3b,1 | 11,5 ± 0,9b,1 | 1,9 ± 0,3d,2 | 53 ± 4a,1 | 26 ± 3d,1 |
BR-130-1C | 6,5 ± 0,3b,1 | 16,2 ± 0,5ab,2 | 21 ± 0,6a,1 | 62 ± 3a,1 | 28 ± 3a,2 |
BR-130-2C | 7,07 ± 0,07b,2 | 17,3 ± 0,4bc,1 | 11,3 ± 0,4b,1 | 65 ± 4a,2 | 31 ± 6a,1 |
BR-130-3C | 6,5 ± 0,3b,1 | 17,97 ± 0,04c,3 | 7,7 ± 0,2c,1 | 67 ± 3a,1 | 22 ± 3ab,2 |
BR-130-4C | 5,8 ± 0,2a,1 | 15,6 ± 0,8a,2 | 2,5 ± 0,1d,1 | 62 ± 5a,12 | 15 ± 2b,2 |
BR-150-1C | 5,07 ± 0,41a,2 | 20,88 ± 0,06a,3 | 16,2 ± 0,9a,2 | 58 ± 6a,1 | 15 ± 2a,3 |
BR-150-2C | 7,4 ± 0,5a,2 | 15,35 ± 0,11b,23 | 11,6 ± 0,4b,1 | 64 ± 5a,2 | 14 ± 2a,2 |
BR-150-3C | 5,5 ± 1,4a,1 | 15 ± 2b,2 | 7,3 ± 0,2c,1 | 62 ± 5a,1 | - |
BR-150-4C | 4,9 ± 0,7a,1 | 14,77 ± 1,02b,2 | 2,9 ± 0,7d,1 | 71 ± 6a,2 | - |
BR-170-1C | 5,5 ± 0,4a,2 | 21,9 ± 0,3a,4 | 6,7 ± 0,2a,4 | 60 ± 3a,1 | - |
BR-170-2C | 5,7 ± 0,3a,1 | 14,4 ± 0,2b,3 | 4,8 ± 0,3b,3 | 42 ± 2c,1 | - |
BR-170-3C | 5,7 ± 0,4a,1 | 11,5 ± 0,2c,1 | 2,9 ± 0,3c,3 | 44 ± 2c,2 | - |
BR-170-4C | 6,2 ± 0,2a,1 | 10,2 ± 0,2d,1 | 1,2 ± 0,2d,3 | 53 ± 4b,1 | - |
a,b,c, different letters indicate significant differences(p < 0.05) between samples at the same extraction temperature
1,2,3..: different numbers indicate significant differences (p < 0.05) between samples at the same bleaching cycle
The cellulose purification progress was monitored through the analysis of lignin and sugars by means of the NREL method (Sluiter 2005 and 2008). After removal of the water (which includes soluble sugars) and ethanol extractives in the samples (between 13–34%), the acid insoluble lining and hydrolyzed sugars were quantified in the different samples. Glucose was the major component, followed by xylose and arabinose. As established in the NREL method, hemicellulose content was considered as total xylose and arabinose and total glucose was attributed to the cellulose content. The obtained values are given in Table 3.
The hemicellulose content was selectively removed when successive bleaching cycles were applied (p < 0.05) in every sample. This hemicellulosic fraction significantly decreased when using 3–4 cycles in BR-110 and BR-130 samples, and completely disappeared in BR-150 and BR-170 samples, after two and one bleaching cycles, respectively. Nevertheless, no significant increase in the cellulose content of the samples occurred during the successive cycles, except for R-170, in which the cellulose content significantly decreased with successive cycles. This suggests that cellulose is progressively degraded through the bleaching cycles with hydrogen peroxide. In fact when referring the cellulose content per mass unit of initial DB, a progressive decrease was observed, ranging from 16 g cellulose/ 100 g DB in the non-bleached residues to 7–13 g cellulose/g DB in the fourth bleaching cycle of the different samples. Degradation of cellulose by the oxidative action of hydrogen peroxide has been reported by other authors (Vismara et al., 2009) through free radical mechanisms forming alpha hydroxyalkyl radicals and subsequent chain scission. This process is largely affected by the substrate composition and the presence of catalysers or inhibitors of the reaction. Therefore, the use of hydrogen peroxide as bleaching agent of BD cellulosic fractions did not yield proper results, since an important part of cellulose is degraded during the delignification process.
In general, the acid insoluble lignin content decreased when successive cycles were applied, especially after the fourth bleaching cycle. Nevertheless, as commented on above, it has to be taken into account that these values were affected by the protein content of the samples. As can be observed, the protein is progressively removed by successive bleaching cycles, especially in sample BR-170 where higher protein solubilization occurred in the SWE step.
The TGA and DTGA curves of the insoluble and bleached residues are shown in Fig. 4. All samples exhibited a first weight loss step 25 and 125 oC corresponding to the loss of bonded water, and the typical degradation steps of lignocellulosic residues, previously commented. The TGA curves of the bleached fractions showed the expected differences in the thermal behavior with respect the non-bleached samples, related with the compositional changes that occurred in the bleaching step. The partial removal of hemicellulose during the bleaching cycles are reflected on the TGA curves where the double peak in DTGA curves of polysaccharides became a single peak, mainly attributed to cellulose degradation, and the temperature of the maximum degradation rate increased from 280 to 300 oC in BR-110 and BR-130 samples. Nevertheless, no relevant changes in the cellulose purification degree can be deduced from the scarce increase in the weight loss step attributed to this polymer, remaining other compounds whose degradation overlapped with the cellulose degradation, as also observed in the analyzed composition. In sample BR-170, very few changes in the TGA curve were observed after the first bleaching cycle, coherently con the small composition changes reflected in Table 3. Therefore, the successive cycles reduced the bleaching mass yield but did not significantly promote cellulose purification, but its degradation. In the other cases the bleaching cycles promoted the removal of hemicellulose, but also did not result in higher cellulose purity due to its partial degradation. The cellulose degradation products probably contributed to the increase in the final residual mass obtained for most of bleached samples. So, the oxidative process applied with hydrogen peroxide in an alkaline medium seems to partially degrade cellulose generating other compounds and reducing the process yield.