Effect of Mashing-in pH on the Biochemical Composition and Staling Properties of the Sweet Wort

Abstract

Mashing is a decisive brewing step, affecting wort quality. Critical mashing parameters include time, temperature, pH, mash density and oxygen uptake. While the pH-dependent biochemical changes during mashing have been researched, the oxidative processes remain unclear. This work explores the impact of pH conditions during mashing on the biochemical composition and oxidative properties of wort. In laboratory trials, mashing was conducted at different mashing-in pH (4.5, 5.0, 5.2, 5.4, 6.0). The sweet worts were analyzed for extract, fermentable sugars, limit dextrins, amino acids, soluble proteins, polyphenols, color, aldehydes, transition metals, reducing potential, and rate of radical formation. Differences were found in enzymatic activities (particularly proteolysis), color, aldehydes, transition metal ions, the rate of radical formation and the rate of oxygen consumption. Notably, when adjusting the pH at mashing-in, the obtained wort tends to gravitate toward the pH of the unadjusted wort, underscoring its buffering capacity. This was reflected by similar properties of the produced worts, indicating pH playing a lesser role during mashing compared to time/temperature. However, under extreme acidified conditions (pH 4.5 at mashing-in), the produced wort markedly deviates. An intriguing negative correlation between the reducing potential and the content of transition metal ions with the rate of radical formation is discussed.

Keywords:

• Buffering capacity
• mashing
• oxidative stability
• pH
• transition metal ions

Introduction

Mashing is a critical step in the brewing process, involving the mixing of milled barley malt (or other cereals) with hot water to facilitate the hydrolysis of starch and protein constituents. By controlling the temperature at specific temperature intervals, starch is hydrolyzed, leading to the production of fermentable sugars, primarily maltose. The enzymatic activity responsible for this conversion is primarily attributed to α-amylase and β-amylase. The time and temperature of the rests as well as the pH conditions are precisely selected based on the optimal enzymatic activity required.^[1] In traditional standard infusion mashing, a series of rests are employed, including a β-glucanase and protein rest at around 45 °C, a β-amylase rest at approximately 63 °C and an α-amylase rest at 72 °C.^[2] Recently, with the development of high quality well-modified malts it is possible to initiate mashing directly at temperatures where α- and β-amylase exhibit the highest activity.^[3]

In addition to temperature and time, pH is another critical control point during mashing, due to the pH optima for highest activity of given enzymes. As a consequence, at the brewhouse level, pH range will impact extract yield, fermentability, wort filtration, total and free amino nitrogen (FAN) levels and color intensity.^[4] Mash acidification, achieved through the addition of calcium salts (precipitating phosphates) or alternative methods, accelerates the degradation of starch, enhances the activities of carbohydrases and the proteolytic enzymes, leading to increased values of total soluble nitrogen (TSN) and FAN as well as a reduction in wort color.^[1,2,5] Besides addition of calcium salts, the mash can be also acidified by means of biological acidification or the addition of mineral and organic acids, which apart from influencing the pH can alter the matrix in different ways. For instance, organic acids may change the buffering capacity of the wort^[6] and calcium ions can bind to proteins, phosphates and oxalic acid.^[1,5] Furthermore, the pH adjustment results in modified solubility properties of certain proteins, increased buffering capacity of the wort and decreased hop utilization during later stages of brewing.^[1]

Next to the biochemical composition of the wort, the pH conditions during mashing are claimed to affect flavor stability of the beer. At first, much attention was drawn to trans-2-nonenal and its formation pathway via enzymatic oxidation of unsaturated fatty acids by lipoxygenases (LOX-1 and LOX-2). Lipoxygenases become completely inactivated at temperature above 65 °C. The enzymes show pH optima close to 6.5. LOX-1 exhibits an extended range of activity, but at mash pH 5.0 displays reduced activity by 50%. The pH optimum for LOX-2 activity is narrower and the enzyme is practically inactivated at pH 5.0.^[7] Typically, pH at mashing in standard conditions is 5.4–5.7. Drost et al.^[8] have reported decreased lipoxygenase activity and reduced nonenal potential when adjusting the pH to 5.0 at mashing-in. Reduced hydroperoxy fatty acids formation (lipoxygenase activity products) was observed when adjusting the pH to 5.0 at mashing-in^[9] or when using an elevated temperature at mashing-in.^[10] Thus, pH adjustment to 5.0–5.2 at mashing-in became a recommended brewing practice in relation to improved beer flavor stability.^[11–17] In addition to lipoxygenase activity, trihydroxy fatty acids can undergo autoxidation to hydroperoxy fatty acids, however, the pH effect during mashing in relation to autoxidation remains unexplored. Yet, over the years it became apparent that trans-2-nonenal is just a part of a bigger picture of beer staling. Increased aldehyde levels, the decrease in iso-α-acids responsible for the intensity and the quality of bitterness and the reduction in esters are the contemporary reported staling markers.^[18–21]

Bamforth^[4] has pointed out that brewing textbooks predominantly rely on an empirical approach, with limited or absent references, when discussing “optimum” pH values for parameters like extract and wort filtration, repeatedly referencing previous textbooks rather than scientific studies. An important challenge often arises from the practice of measuring wort or mash pH at room temperature and assuming that these values remain constant at higher temperatures, which is not the case, as studies have shown that at 65 °C the pH of a wort can be approximately 0.35 pH units lower than its room temperature measurement.^[22,23] Previous investigations have consistently highlighted the positive impact of mashing-in at elevated temperatures (63 °C) and pH adjustments to 5.2 on beer flavor stability, particularly in terms of reducing lipid oxidation products such as trans-2-nonenal and hexanal.^[24,25] Conversely, recent results of Mertens et al.^[23] and Van Mieghem et al.^[26] suggest that higher mash pH results in sweet worts with lower iron content, which may lead to lower formation of the reactive oxygen species via Fenton and Haber-Weiss reactions, potentially benefiting oxidative stability of the final beer. In addition, higher pH during mashing was reported to favor chelation and complex formation with transition metal ions, which can be removed during mash filtration.^[23,27]

This study aims to investigate the oxidative properties of wort in parallel with the biochemical conversion of malt components during mashing as a function of pH adjustments at mashing-in. Previous research has highlighted a positive correlation between the oxidative stability of wort and the quality as well as the prolonged shelf-life of the resulting beers.^[17,28–36] There is a noteworthy absence of detailed scientific literature pertaining to the impact of the pH conditions during mashing on the oxidative stability of wort. To monitor the extent of oxidation of the produced worts we employed two methods. The first was electron spin resonance (ESR) spectroscopy that combined with the spin trapping technique allows the evaluation of rate of formation of free radicals in wort at 60 °C. The ESR method has been previously employed in the analysis of wort^{.[23,34,36–38]} The second methodology determined the rate of oxygen consumption in the air-saturated wort samples. The investigated oxidative stability has been evaluated in parallel with oxidation-related parameters, for example, reducing potential and transition metal ion content as well as biochemical properties of wort, including extract yield, fermentable sugars, limit dextrins, soluble proteins, free amino acids, and polyphenols, among others. Finally, we have chosen to focus our investigation on a single malt, specifically a pilsner-type, as it serves as a representative sample for the majority of brewing conditions. This focused approach enabled us to conduct an extensive set of analyses, leading to a more comprehensive understanding of the observed phenomena.

Methodology

Malt quality parameters

Standard malt quality parameters were evaluated according to the methods of the European Brewery Convention.^[39] The following parameters were assessed: moisture content (4.2), homogeneity and partly unmodified grains (4.14), friability (4.15), total nitrogen content (4.17), diastatic power (4.12.1), α-amylase activity (4.13), extract yield (4.5.1), color (4.7.1), viscosity (4.8), total soluble nitrogen (4.9.2), free amino nitrogen (4.10), pH (8.17), β-glucan (4.16). Determination of the thiobarbituric acid index (TBI) in malt was carried out following the adapted method of Coghe et al.^[40] Trihydroxy fatty acids in malt were determined according to Jaskula-Goiris et al.^[19] Determination of S-methylmethionine and dimethyl sulfide and lipoxygenase activity in malt was carried out according to method described by Dugulin et al.^[41] Free aldehydes were determined according to Filipowska et al.^[42] The determination method for amino acids is covered in further methodology subsection of this manuscript. Thermo Scientific drying oven (Thermo Fisher Scientific, Waltham, MA, U.S.A.) was used for the evaluation of moisture content of the malt.

Sweet wort production

Wort samples were prepared on laboratory scale with continuous stirring, under atmospheric conditions (Lochner Labor und Technik, GmbH, Berching, Germany). Applied malt to water ratio was 1: 2.5; 150 g of Pilsner malt (supplied by Brewery Bavaria, Swinkels Family Brewery, Lieshout, the Netherlands) was mixed with 375 g of reverse osmosis water enriched with 80 ppm Ca²⁺ ions in form of calcium chloride dihydrate (Merck KGaA, Darmstadt, Germany). The following mashing protocol was applied: 63 °C (30 min), 72 °C (15 min), 78 °C (1 min); temperature rise 1 °C/min. Upon end of mashing, the mash was cooled down to room temperature and standardized by addition of distilled water up to a total weight of 525 g. The approved mash was centrifuged for 10 min at 9,000 rpm. The supernatant was separated and kept at −20 °C, prior being subjected for analysis. Headspace of the wort samples subjected for ESR oxidative stability analysis was flushed with N₂, prior to freezing. Each mashing sample was prepared in three biological replicates.

pH adjustment at mashing-in

The pH measurements at mashing-in and mashing-off were made at 63 °C by Consort C831 pH-meter (Consort NV, Turnhout, Belgium). The pH-meter was calibrated at 63 °C. The experimentally determined pH value at 63 °C of the non-adjusted mash was 5.43 and corresponded to a pH reading of 5.77 at ambient temperature. HCl and NaOH (Merck KGaA, Darmstadt, Germany) were used to adjust the pH at mashing-in. Based on pre-trials, amounts of NaOH and HCl were selected to obtain the target pH value at mashing-in. The pH 4.5, 5.0, 5.2, and 6.0 were achieved by the addition of 1.3 mL 5 M HCl, 2.0 mL 1 M HCl, 1.0 mL 1 M HCl, and 1.0 mL 1 M NaOH, respectively. Acid and base were added in one portion (not by titration) directly after mixing malt with water. Immediately after the acid/base addition, samples were subjected for mashing procedure. In all samples, the pH at mashing-in was measured 3 min after acid/base addition.

Wort analysis

Anton-Paar DMA 5000 Alcolyzer density meter and Anton-Paar Alcolyzer (Anton Paar, Graz, Austria) was used to determine basic wort parameters. European Brewery Convention^[39] methods were used for the evaluation of wort color (9.6), total polyphenol content (9.11), and flavonoid content (9.12). The content of soluble proteins was determined according to Bradford.^[43] Proanthocyanidin content was determined according to Bate-Smith.^[44] The spectrophotometric measurements were carried out using the Spectrometer Thermo Scientific Evolution 60 S UV-Vis (Thermo Fisher Scientific, Waltham, MA, U.S.A.).

Fermentable and nonfermentable sugars analysis

Fermentable sugars (glucose, maltose, maltotriose) and limit dextrins with a degree of polymerization of 4–8 (maltotetraose, maltopentaose, maltohexaose, maltoheptaose, and maltooctaose) were quantified by high performance liquid chromatography (HPLC) coupled with refractive index (RI) detector integrated with Waters 717 autosampler (Waters, Milford, MA, U.S.A.). Prior the analysis, the samples were ×10 diluted. Proteins were precipitated by addition of 20  µL Carrez I solution (LCH Chimie, Les Aires, France) and 20 µL of Carrez II solution (LCH Chimie, Les Aires, France) to 1 mL of wort sample and subsequent centrifugation. Separation of components was obtained by Nucleodur NH2 100-5 column (250 × 4 mm i.d.; Macherey-Nagel, Düren, Germany) at 30 °C. Water/acetonitrile (37/63% v/v) (acetonitrile: Sigma Aldrich Co., St. Louis, MO, U.S.A.) was used as the mobile phase at a flow rate 1.0 mL/min. Waters 2414 RI-detector (Waters, Milford, MA, U.S.A.) was used for detection. Maltose was used for external calibration. Relative response factors compared to maltose were used for the determination of remaining components.

Amino acid analysis

Free amino acids were quantified using Acquity Ultra Performance Liquid Chromatography (UPLC) (Waters, Milford, MA, U.S.A.) with Photo Diode Array detector. Components were separated by AccQ Tag Ultra column (100 × 2.1 mm i.d.; Waters, Milford, MA, U.S.A.). Amino Acid standard H (2.5 mM of Ala, Arg, Asp, Cys, Glu, Gly, His, Ile, Leu, Lys, Met, Phe, Pro, Ser, Thr, Tyr, and Val) (Waters, Milford, MA, U.S.A.) and 2.5 mM L-Tryptophan (>98%, Merck, Darmstadt, Germany), L-Glutamine (>99%, Merck, Darmstadt, Germany), and L-Asparagine (>99%, Merck, Darmstadt, Germany) were used as reference compounds for external calibration. AccQ Tag Ultra derivatization kit (Waters, Milford, MA, U.S.A.) was used for component derivatization. AccQ Tag Ultra eluents A and B were used as mobile phases. The flow rate and temperature were set at 0.7 mL/min and 60 °C, respectively.

Transition metal ion analysis

Inductively coupled plasma-optical emission spectroscopy (ICP-OES) in combination with Avio 200 spectrometer (PerkinElmer, Rodgau, Germany) was used for determination of levels of Fe, Cu, Mn, and Zn, according to Mertens et al.^[23] Emission lines used for quantification of Fe, Cu, Mn, and Zn were 238.204, 327.398, 257.610, and 206.200 nm.

Rate of radical formation

The rate of radical formation was evaluated according to Frederiksen et al.^[45] with minor modifications. ESR spectra were obtained by Miniscope MS 200 X-band spectrometer (Magnettech GmbH, Wertheim, Germany). Following instrumental conditions were applied: microwave power, 10 mW; magnetic field, 336.5 mT; sweep width 6.0 mT; sweep time, 60 s; modulation, 0.2 mT. Spectra were measured at room temperature. The amplitudes of the spectra were measured and were reported as the mean value of the height of the first and the second doublets. The intensities of the spectra were standardized relative to 2 µM 2,2,6,6-tetramethyl-1-pipe-ridinyloxyl (TEMPO) (Sigma Aldrich Co., St. Louis, MO, U.S.A.). The TEMPO standard was measured as the first and the last sample of the day. A 600 mM of spin trap α-(4-pyridyl-1-oxide)-N-tert-butylnitrone (POBN) (Sigma Aldrich Co., St. Louis, MO, U.S.A.) dissolved in 96% ethanol (Sigma Aldrich Co., St. Louis, MO, U.S.A.) was prepared. An aliquot of 1.9 mL of wort sample was mixed with 100 µL of POBN solution. The reaction mixture was incubated at 60 °C and aliquots were transferred to ESR flow cell in 10-min intervals. The rate of radical formation for each sample was expressed as the slope of the function of the ESR signal relative to TEMPO and reaction time (min).

Oxygen consumption

Oxygen consumption was determined according to the method used by Pagenstecher et al.^[38] Oxygen was measured with a HQ 30d Luminescent Dissolved Oxygen (LDO) sensor (Hach Lange, Düsseldorf, Germany) that was initially calibrated against Milli-Q water saturated with atmospheric air, at 25 °C. Wort sample (20 mL), which had been saturated with atmospheric air, was transferred to a 25 mL-glass tube. The sensor was inserted, and the oxygen concentration was measured in the air tight system at 25 °C every 5 min for 10 h. The reported rates of consumption were calculated based on the slope of the linear curves of the measured oxygen concentrations between 30 and 300 min. The values are expressed in mg of consumed oxygen per 1 mL of the sample in 1 h.

Reducing potential

Reducing potential of wort samples was determined by ferric reducing antioxidant power (FRAP) assay according to Benzie and Strain^[46] by using 96-well micro-plates and the SpectraMax i3x Multi-Mode Microplate Reader (Molecular Devices, LLC., San Jose, CA, U.S.A.). A FRAP reagent (10/1/1, v/v/v) was prepared with an acetate buffer solution (300 mM, pH 3.6), 2,4,6-tripyridyl-s-triazine (TPTZ) (Sigma Aldrich Co., St. Louis, MO, U.S.A.); (10 mM in 400 mM HCl); and 20 mM aqueous FeCl₃ × 6H₂O (Sigma Aldrich Co., St. Louis, MO, U.S.A.). Aqueous solutions of known Fe^II concentrations (in FeSO₄ × 7H₂O: Sigma Aldrich Co., St. Louis, MO, U.S.A.) ranging from 0.025 to 0.80 mM were used for 9-point calibration. Aliquots of 25 µL of centrifuged wort samples were transferred into the wells of the micro-plate followed by addition of 250 µL of FRAP working solution. The micro-plate was agitated and incubated for 8 min at 25 °C prior to measuring the absorbance at 593 nm.

Volatile aldehydes

Volatile aldehyde quantification was based on o-(2,3,4,5,­6-pentafluorobenzyl)hydroxylamine (PFBHA) (Sigma Aldrich Co., St. Louis, MO, U.S.A.) derivatization followed by head-space solid phase microextraction gas chromatography-mass spectroscopy (HS-SPME-GC-MS) according to a method developed by De Clippeleer^[47] and further optimized by Baert.^[48] Detailed experimental conditions as described by Ditrych et al.^[49] The following marker compounds were investigated: 2-methylpropanal (2MP), 2-methylbutanal (2MB), 3-methylbutanal (3MB), hexanal (HEX), furfural (FUR), methional (MET), phenylacetaldehyde (PHE), and trans-2-nonenal (T2N). Authentic reference compounds (Sigma Aldrich Co., St. Louis, MO, U.S.A.) in N₂-flushed Milli-Q water were used for external calibration with linear calibration (R² ≥ 0.9800). Under oxygen limited conditions in a 850-Series nitrogen dry box (Plas-Labs, Lansing, MI, U.S.A.), 0.4 mL of centrifuged wort sample, and 3.6 mL of N₂-flushed Milli-Q water were transferred to a 20 mL amber glass vial, closed with crimp seals with PTFE/silicone septa (Sigma Aldrich Co., St. Louis, MO, U.S.A.) and subjected for analysis. Data were processed in Xcalibur^TM (Thermo Electron Corporation, Waltham, MA, U.S.A.).

Statistical analysis

Wort samples were prepared in three biological replicates. Each biological replicate was analyzed in three technical replicates, unless mentioned differently. Means and standard deviations were calculated. Statistical significance was evaluated by one-way analysis of variance (ANOVA) with post-hoc Tukey HSD test in SPSS statistical software (IBM Corporation, Armok, NY, U.S.A.).

Results

Malt quality

A light pilsner-type of malt produced under low kilning regime (temperature up to 65 °C) was selected for laboratory scale mashing experiments. The same malt was utilized in our previous investigation, where its quality parameters were compared with the literature recommended values.^[18] The low color intensity, low free amino nitrogen content, and low Kolbach index values indicate a low degree of proteolytic modification during the malt production process (Table 1). Nonetheless, relatively high levels of individual amino acids (especially precursors for Strecker aldehydes: valine, leucine, isoleucine, methionine, and phenylalanine) and relatively high α-amylase activity suggest sufficient substrate levels for the potential formation of staling compounds i.a. Strecker aldehydes and Maillard reactions intermediates and products. In addition, low kilning intensity resulted in relatively high lipoxygenase activity displayed by the produced malt, when compared to corresponding values in the literature.^[41,50]

Table 1. Quality parameters of Pilsner malt selected for this study.

Standard malt characteristics		Amino acids [mg/kg dm]
Moisture [%]	6.0 ± 0.2	Histidine	178.16
Extract yield [% dm]	82.6 ± 0.4	Asparagine	407.54
Color [EBC]	3.2 ± 0.2	Serine	262.90
TBI [for 10 g dm]	9.87 ± 0.00	Glutamine	70.25
Friability [%]	90.6 ± 1.7	Arginine	–
Homogeneity [%]	98.4 ± 0.5	Glycine	168.84
Partially unmodified grain [%]	1.60 ± 0.52	Aspartic acid	53.37
Whole grain [%]	0.6 ± 0.4	Glutamic acid	470.26
β-Glucan [mg/L]	148 ± 20	Threonine	199.53
Viscosity [mPa·s]	1.56 ± 0.02	Alanine	288.85
Total protein [% dm]	10.9 ± 0.3	Proline	3136.46
Total soluble protein [% dm]	4.00 ± 0.16	Cysteine + Cysteic acid	63.53
Kolbach index [%]	36.8 ± 1.4	Lysine	173.97
Free amino nitrogen [mg/L]	132 ± 7	Tyrosine	337.80
Diastatic power [°WK]	320 ± 25	Methionine	84.64
α-Amylase activity [DU]	61.0 ± 5.5	Valine	395.28
pH [-]	6.07 ± 0.05	Isoleucine	231.81
Volatile aldehydes [µg/kg dm]		Leucine	394.38
2-Methylpropanal	1131 ± 91	Phenylalanine	459.18
2-Methylbutanal	1069 ± 50	Tryptophan	197.33
3-Methylbutanal	2592 ± 132	Beer staling related parameters
Methional	202 ± 50	Total polyphenols [mg/L]	46.54 ± 0.52
Phenylacetaldehyde	963 ± 33	S-methylmethionine [mg/kg dm]	6.00 ± 0.80
Hexanal	615 ± 70	Dimethyl sulfide [mg/kg dm]	2.46 ± 0.01
trans-2-Nonenal	454 ± 42	Trihydroxy fatty acids [mg/kg dm]	32.90 ± 0.40
Furfural	402 ± 64	LOX-activity [U/g dm]	76.8 ± 4.2

*dm − dry matter.

Data represent mean values ± standard deviation from three technical replicates (n = 3), except for amino acids (n = 1).

Mashing and pH

Current brewing protocols advocate for a high mashing-in temperature, typically aligned with the gelatinization point (60 to 65 °C depending on the malt) and pH adjustment to 5.2 at mashing-in. These measures are claimed to have a positive impact on beer flavor stability (by inhibiting lipoxygenase activity) and at the same time allowing for efficient biochemical conversion of malt components for yeast growth and fermentation. In this study, all pH measurements were done at 63 °C in order to characterize the actual pH changes at a temperature that are relevant during mashing. The pH of the unadjusted mash was 5.4 at mashing-in and it barely changed during mashing (Figure 1). Pre-trials were conducted, during which acid/base was added at the onset of mashing in the range of 1-5 mL for 1.0 M HCl, 5.0 M HCl, or 1.0 M NaOH. It was observed that the pH of the mash stabilized after 3 min and consequently, we selected this time point for measuring the pH at mashing-in. The pre-experiments also allowed the precise determination of the amounts of acid/base that were required to achieve the targeted pH values at mashing-in (see Methodology, pH adjustment at mashing-in), which resulted in low variations in mashing-in pH, evidenced by the relative standard deviation among the biological replicates being less than 1%. The highest overall changes in the pH during mashing were observed for the samples mashed-in at pH 4.5, 5.0, and 6.0, which was on average +0.17, +0.22, and −0.33, respectively (Figure 1). It appears that regardless of the pH adjustment at mashing-in, the final pH of the sweet worts gravitated toward the unadjusted wort pH (5.4).

Figure 1. pH values at mashing-in (black) and in the final mash (grey). Results represent mean value and standard deviation (n = 3). Measurements were carried out at 63 °C. One-way analysis of variance (ANOVA) with Tukey post-hoc test at p < 0.05 was used for statistical evaluation. Statistically distinguished groups for samples whose pH was measured at mashing-in were labelled with upper case letters, whereas groups for samples whose pH was measured in the sweet wort were labelled with lower case letters.

Composition of the produced worts

The changes in the composition of wort as a function of mashing-in pH were assessed based on analysis of extract, fermentable sugars, limit dextrins, soluble proteins, individual amino acids, polyphenols, and transition metals. In addition, wort color and FRAP reducing potential were evaluated. Overall, the collected data were consistent, pointing to high level of repeatability among the tested samples.

Regarding fermentable sugars and extract content (Figure 2), the yield was dependent on the mashing-in pH, generally with highest levels obtained in the pH range 5.0–5.4. The low content of limit dextrins seen in all samples (Table 2) indicates extensive starch degradation within the produced samples. The lowest level of fermentable sugars was found in the sample with mashing-in pH 4.5. Naturally, maltose was the most abundant carbohydrate with the maximum level of 119 ± 1 g/L determined in the sample, which was mashed-in at pH 5.2. The contributions of maltose, maltotriose, and glucose in relation to the sum of total quantified sugars analyzed (fermentable sugars and limit dextrins) were approx. 59–65%, 20%, and 12%, respectively. In addition, a strong positive correlation (R² = 0.96) was seen between the extract content and maltose levels determined in the sweet worts.

Figure 2. Extract and fermentable sugars content in sweet wort as a function of mashing-in pH measured at 63 °C. The following symbols represent means and standard deviations (n = 3) for maltose (red circles), glucose (black diamonds), maltotriose (blue squares), and extract content (green triangles).

Table 2. Limit dextrins in sweet worts produced at varying mash pH conditions.

pH at mashing-in [-]	4.42 ± 0.06 ^E	5.01 ± 0.01 ^D	5.23 ± 0.03 ^C	5.43 ± 0.05 ^B	5.99 ± 0.03 ^A
pH at mashing-off [-]	4.60 ± 0.05 ^E	5.26 ± 0.04 ^D	5.36 ± 0.04 ^C	5.44 ± 0.01 ^B	5.66 ± 0.02 ^A
Maltotetraose [g/L]	11.4 ± 1.8 ^A	8.4 ± 2.0 ^A,B	7.1 ± 1.0 ^B	6.1 ± 0.4 ^B	6.0 ± 1.2 ^B
Maltopentaose [g/L]	1.6 ± 0.6 ^A	1.6 ± 0.5 ^A	1.1 ± 0.2 ^A	0.8 ± 0.3 ^A	0.9 ± 0.7 ^A
Maltohexaose [g/L]	2.2 ± 0.2 ^A	0.5 ± 0.6 ^B	0.5 ± 0.1 ^B	0.9 ± 0.7 ^B	0.7 ± 0.1 ^B
Maltoheptaose [g/L]	1.3 ± 0.5 ^A	0.8 ± 0.0 ^A	0.8 ± 0.4 ^A	0.9 ± 0.2 ^A	1.5 ± 0.4 ^A
Maltooctaose [g/L]	1.1 ± 0.4 ^A	0.6 ± 0.2 ^A	0.7 ± 0.4 ^A	0.5 ± 0.5 ^A	0.9 ± 0.7 ^A
SUM [g/L]	17.6 ± 3.5 ^A	12.0 ± 3.1 ^A,B	10.2 ± 2.1 ^B	9.2 ± 2.1 ^B	10.0 ± 3.1 ^B

pH measurements were carried out at 63°C. Data represent mean and standard deviations (n = 3). Capital letters in the superscripts (A, B, C) indicate statistically distinguished groups by one-way analysis of variance (ANOVA) with a Tukey post hoc test at p < 0.05.

The lowest total polyphenol content was observed in the sweet worts produced at mashing-in pH 4.5 (Table 3). The other samples showed no significant differences in the total polyphenols. The levels of proanthocyanidins indicated only a minor effect of mashing-in pH. However, a clear effect was observed for flavonoids, where levels were approximately 20% lower in the worts produced at mashing-in pH 6.0 as compared to samples produced at pH 4.5. The employed methodology for flavonoid quantification is not sensitive to glycosides. Taking into consideration the broad spectrum of pH optima for β-glucosidases (from 4 to 9)^[51] and especially for α-glucosidase in barley malt (pH optimum at 4.6),^[52] the variation of the flavonoid content could potentially be linked to the pH dependence of the enzymatic hydrolysis of barley polyphenol-glycosides.

Table 4. Levels of free amino acids and soluble proteins at varying mash pH conditions.

pH at mashing-in [-]	4.42 ± 0.06 ^E	5.01 ± 0.01 ^D	5.23 ± 0.03 ^C	5.43 ± 0.05 ^B	5.99 ± 0.03 ^A
pH at mashing-off [-]	4.60 ± 0.05 ^E	5.26 ± 0.04 ^D	5.36 ± 0.04 ^C	5.44 ± 0.01 ^B	5.66 ± 0.02 ^A
Alanine [mg/L]	175.23 ± 11.82 ^A	174.33 ± 15.00 ^A	162.39 ± 4.81 ^A	150.72 ± 11.84 ^A, B	127.34 ± 7.05 ^B
Arginine [mg/L]	328.96 ± 11.16 ^A	312.42 ± 14.59 ^A	275.74 ± 4.62 ^B	244.83 ± 13.20 ^C	195.13 ± 4.61 ^D
Asparagine [mg/L]	160.62 ± 6.83 ^A	164.27 ± 8.47 ^A	158.14 ± 2.95 ^A	154.61 ± 11.59 ^A	149.22 ± 5.03 ^A
Aspartic Acid [mg/L]	143.57 ± 11.45 ^A	140.95 ± 12.16 ^A	137.88 ± 4.71 ^A	132.71 ± 9.50 ^A	124.31 ± 8.84 ^A
Cysteine and Cystine [mg/L]	41.60 ± 4.28 ^A	38.01 ± 3.49 ^A	34.36 ± 1.21 ^A, B	26.92 ± 2.49 ^B	15.65 ± 0.53 ^C
Glutamine [mg/L]	390.28 ± 16.85 ^A	375.30 ± 23.27 ^A, B	352.93 ± 7.59 ^A, B, C	339.54 ± 23.28 ^B, C	324.43 ± 12.99 ^C
Glutamic Acid [mg/L]	214.36 ± 17.97 ^A	172.11 ± 17.40 ^B	160.59 ± 8.68 ^B	152.66 ± 4.60 ^B	163.02 ± 9.85 ^B
Glycine [mg/L]	62.57 ± 3.37 ^A	61.03 ± 2.81 ^A	55.66 ± 1.38 ^A, B	53.09 ± 3.51 ^B	48.87 ± 1.97 ^B
Histidine [mg/L]	103.67 ± 4.06 ^A	105.77 ± 3.20 ^A	97.76 ± 1.81 ^A, B	92.93 ± 6.31 ^B, C	84.80 ± 2.33 ^C
Isoleucine [mg/L]	127.84 ± 6.77 ^A	124.82 ± 8.85 ^A, B	115.35 ± 2.50 ^A, B	110.07 ± 8.20 ^B, C	97.09 ± 4.17 ^C
Leucine [mg/L]	352.94 ± 18.59 ^A	327.27 ± 23.60 ^A, B	289.30 ± 6.82 ^B, C	255.01 ± 19.07 ^C	194.65 ± 7.35 ^D
Lysine [mg/L]	178.42 ± 14.20 ^A, B	189.59 ± 19.76 ^A	174.17 ± 5.04 ^A, B	151.89 ± 13.44 ^B	109.11 ± 8.06 ^C
Methionine [mg/L]	82.82 ± 3.59 ^A	75.27 ± 3.22 ^A	65.14 ± 2.33 ^B	57.14 ± 4.28 ^B	42.88 ± 1.13 ^C
Phenylalanine [mg/L]	289.88 ± 10.02 ^A	265.99 ± 6.28 ^B	238.31 ± 4.52 ^C	217.71 ± 11.58 ^C	187.96 ± 2.77 ^D
Proline [mg/L]	>230.20	>230.20	>230.20	>230.20	>230.20
Serine [mg/L]	154.25 ± 7.80 ^A, B	160.11 ± 7.04 ^A	151.06 ± 4.71 ^A, B	148.16 ± 7.90 ^A, B	138.45 ± 7.44 ^B
Threonine [mg/L]	119.66 ± 6.01 ^A	115.84 ± 7.32 ^A	105.59 ± 2.39 ^A, B	97.66 ± 7.52 ^B, C	84.64 ± 3.45 ^C
Tryptophan [mg/L]	97.30 ± 2.48 ^A	90.00 ± 1.08 ^B	82.61 ± 1.66 ^C	78.61 ± 2.75 ^C, D	73.85 ± 0.83 ^D
Tyrosine [mg/L]	224.46 ± 8.94 ^A	206.81 ± 4.26 ^B	184.40 ± 2.26 ^C	163.08 ± 8.87 ^D	139.29 ± 2.90 ^E
Valine [mg/L]	250.70 ± 10.90 ^A	241.25 ± 15.79 ^A	220.20 ± 8.03 ^A, B	205.09 ± 15.42 ^B, C	173.18 ± 7.57 ^C
SUM of free amino acids [mg/L]*	3506.91 ± 172.38 ^A	3349.58 ± 195.34 ^A, B	3070.26 ± 73.30 ^B, C	2841.57 ± 184.29 ^C, D	2482.40 ± 95.33 ^D
Soluble protein [mg/L]	992.48 ± 76.07 ^B	1212.30 ± 44.24 ^B, A	1231.48 ± 6.52 ^B, A	1263.70 ± 21.30 ^A	1345.97 ± 83.04 ^A

*Sum of the individual amino acids excluding proline.

pH was measured at 63 °C. Data represent mean and standard deviations (n = 3). Capital letters in the superscripts (A, B, C) indicate statistically distinguished groups by one-way analysis of variance (ANOVA) with a Tukey post hoc test at p < 0.05.

Soluble proteins (determined according to Bradford^[43]) and the free amino acids were clearly dependent on the mashing pH conditions. UPLC-MS was applied to precisely measure individual free amino acids in solution, enabling accurate quantification. Furthermore, the methodology by Bradford^[43] was used for quantifying soluble proteins, which relies on colorimetric measurement based on dye-protein binding. The content of soluble proteins increased proportionally with the increasing pH of sweet wort, ranging from 992 ± 76 mg/L (worts produced at 4.5 mashing-in pH) to 1,346 ± 83 mg/L (worts produced at 6.0 mashing-in pH) (Table 4). An opposite trend was observed for the amino acids, where the content decreased with increasing mashing pH (Table 4). The highest levels were seen for glutamine, leucine, and arginine and the lowest for glycine, methionine as well as cysteine and cystine. Proline was present above its detection limit (>230.2 mg/L) in all samples. The data spread within the data sets varied depending on the compound from 1 to 10% (5% on average).

Regarding transition metal ions, sweet wort produced at mashing-in pH 4.5 contained markedly higher levels of Fe, Mn, and Zn ions (Figure 3). When comparing sweet wort produced at pH 4.5 and 6.0, the average reduction in Fe, Mn, and Zn ions was 84%, 83%, and 77%, respectively. Conversely, an opposite behavior was observed for Cu ions, where the levels were proportional—with a linear trend—to the pH of the final mash. The absolute differences between the extreme samples were somewhat lower (by 40%).

Figure 3. Levels of transition metal ions as a function of pH in the sweet wort at mashing-off. pH in the final sweet wort was measured at 63 °C. The symbols represent mean values (n = 3) for Fe (black squares), Mn (blue triangles), Zn (green diamonds), and Cu (red circles). The corresponding color filling represent the error bands.

Oxidative stability

Notably, the rate of radical formation was negatively correlated in relation to the content of transition metal ions, which is presumably the most intriguing outcome of this investigation. The oxidative properties of the produced sweet worts were examined by ESR spectroscopy. A general linear increase in the steady radical formation as a result of forcing wort samples at 60 °C was observed until t = 40 min and seemed to level off afterwards. Sweet wort produced at mashing-in pH 4.5 displayed the lowest radical formation, whereas the sample produced at pH 6.0 the highest (Figure 4). Moreover, the rate of radical formation determined in the sweet worts showed practically a linear dependence with the final pH measured in the sweet worts. Analysis of the oxidative stability of the wort samples by measuring their rate of oxygen consumption at 25 °C showed a similar trend as observed by the spin trapping and ESR method (Figure 5). The wort sample produced at mashing-in pH of 6.0 displayed a threefold higher oxygen consumption rate compared to the sample produced at mashing-in pH of 4.5. Except for the sample from the mashing without adjustment of the pH, then the oxygen consumption rates showed a clear increasing trend with the final mash pH.

Figure 4. ESR-determined rate of radical formation as a function of pH in the sweet worts produced at varying pH conditions. The pH of the sweet wort was measured at 63 °C. Linear regression fitting is represented by the line. Capital letters (A, B, C) indicate statistically distinguished groups by one-way analysis of variance (ANOVA) with a Tukey post hoc test at p < 0.05.

Figure 5. Rates of oxygen consumption at 25 °C in sweet worts as a function of final mash pH. The pH of the sweet worts were measured at 63 °C. Results represent means and standard deviations (n = 3).

These results suggest that the pH of the sweet wort could be the main factor determining the formation of free radicals, which was further investigated by adjusting the pH of the sweet worts to the pH of the extreme samples (sweet worts produced at mashing-in pH of 4.5 and 6.0), as illustrated schematically in Figure 6. The pH in the sweet wort produced at mashing-in pH 4.5 was adjusted to the same pH value as in the sweet wort produced at 6.0 and analogously, the wort sample produced at mashing-in pH of 6.0 was adjusted to the pH of the sweet wort of the sample produced at 4.5. Also, the pH of the reference sweet wort—whose inherent mashing-in pH was 5.4—was adjusted to the pH of both extreme wort samples. The pH adjustment was carried out in triplicates using the same acid and base as used for adjusting pH during mashing.

Figure 6. Effect of pH adjustment in the final wort on the electron spin resonance (ESR)-determined oxidative parameters. Scheme on the left represent the pH adjustment of the final sweet wort used for the evaluation of the pH-effect on the determined ESR oxidative properties of worts. pH was measured at 63 °C. Each pH adjustment was carried out in triplicate. The bar graph on the right represents the rate of radical formation in sweet wort in reference samples and pH-adjusted sweet wort samples. Statistical significance evaluated at p < 0.05.

The sweet worts, where the pH was adjusted to 5.7 (which corresponds to the reference sweet wort sample produced at mashing-in pH 6.0) displayed similar rates of radical formation as compared to the reference sample (sweet wort produced at mashing-in pH 6.0) in the ESR experiments. However, the sweet worts, where pH was acidified to 4.6 (corresponding to the reference—pH of the sweet wort produced at mashing-in 4.5) showed a significantly higher (p < 0.05) rate of radical formation in comparison to the reference sample (sweet wort produced at mashing-in pH 4.5) (Figure 6).

The ferric reducing potentials of the produced sweet worts were determined by FRAP assay (Table 3). The sweet wort produced at mashing-in pH 4.5 showed significantly higher ferric reducing potential in comparison to other samples. Among the remaining four samples, no significant differences were identified.

Aldehyde composition

The levels of volatile aldehydes quantified in the produced sweet worts are covered in Figure 7. Regarding Maillard reactions, the low mashing-pH conditions resulted in higher levels of furfural, which correlated with the color of the sweet wort. The levels of 2-methylpropanal, 2-methylbutanal, and 3-methylbutanal determined in the sweet worts seemed to somewhat increase with the increasing mashing-in pH. A contrary behavior was observed for furfural, methional, phenylacetaldehyde, and trans-2-nonenal, where maximum levels were observed in the sweet wort produced at mashing-in pH 4.5. The varying values of mashing-in pH did not appear to affect the concentrations of hexanal.

Figure 7. Volatile aldehydes as a function of mashing-in pH. The pH was determined at mashing-in at 63 °C. The values for the determined aldehyde levels represent means and standard deviations (n = 3). Capital letters (A, B) represent statistically distinguished groups by one-way analysis of variance (ANOVA) with Tukey post-hoc test at p < 0.05.

Discussion

The obtained results emphasize the significance of buffering capacity reflected by wort’s natural ability to resist changes in the biochemical composition and staling properties. In order to minimize matrix interactions (e.g., chelation), pH at mashing-in was adjusted with HCl and NaOH solutions. The pH measurements at mashing-in and in the final sweet wort were purposely decided to be made at 63 °C, since the mashing process typically is carried out in this temperature range. Furthermore, wort cooling to reach ambient temperature was omitted to avoid the time delay. The corresponding pH measurements at ambient temperature are anticipated to be higher by approximately 0.35 units. Regarding the pH change during mashing, it undergoes a pronounced shift immediately after the addition of acid or base and practically stabilizing during the first three minutes of mashing. Subsequently, although the pH continues to change towards the unadjusted mash pH, the rate of change became notably slower.

The results of this study clearly demonstrate that regardless of the pH adjustment during mashing, the final pH of the sweet wort tends to gravitate toward the unadjusted pH of the wort (Figure 1). In general, worts of higher pH range display greater buffering capacity.^[6] Several components in wort, such as phosphates, free amino acids, peptides and polypeptides containing residues like aspartate and glutamate, contribute to its buffering capacity.^[4,53] Factors elevating the levels of free amino acids, peptides and polypeptides (including the nitrogen content of the malt, its degree of modification and the extent of proteolysis occurring during mashing) enhance the overall buffering capacity. Low buffering capacities of wort were reported when using adjuncts with low contents of proteins.^[54] MacKenzie and Kenny^[55] indicated that besides amino nitrogen, phosphate play a significant role in buffering pH changes during wort fermentation. Coote and Kirsop^[56] suggested only a limited buffering capacity of phosphate and amino acids in fermenting wort. Reduction of the levels of buffering material and the accumulation of organic acids did not account for the similar buffering capacity observed in wort and consecutively in the produced beers. Low contribution from phosphate and free amino acids (glutamic acid, aspartic acid and histidine) to the wort’s buffering capacity was also reported in a recent study by Li et al.^[6] Finally, the presence of some organic acids has been reported to improve the buffering capacity, which may prevent rapid acidification due to Saccharomyces cerevisiae metabolism.^[57] Supplementation of wort with organic acids (especially acetic acid) was found to substantially enhance the buffering capacity of wort.^[6] The main components of the wort that contribute the most to its buffering capacity are still not clearly identified, however this study demonstrates that during mashing the buffering compounds in the wort are generated that appears to partly counteract adjustments of pH that have been made at mashing-in.

In general, it seems that even when the mash environment is alkalized or acidified at the onset of mashing within the pH range of 5.0 to 6.0, the buffering capacity is strong enough to ensure the produced sweet worts to exhibit relatively comparable properties. However, when mashing-in at very low pH conditions of 4.5, the obtained sweet wort markedly stands out from the other tested wort samples. Several significant differences were identified for the sweet wort produced with this low pH adjustment, i.a. increased cloudiness, lower soluble protein content, lower extract yield along with higher content of limit dextrins, higher ferric reducing potential, and higher solubility of transition metal ions. These observations points to the unique behavior of the sweet wort produced under the extreme low acidified conditions in relation to extraction/solubilization and precipitation properties of wort components as well as enzymatic activities.

Assessment of fermentable sugars and extract content determined in the sweet worts clearly demonstrate the highest extract yields (24.6–24.8 °P) for the samples produced at mashing-in pH range 5.0 − 5.4, which is in line with literature sources^[1,4] (Figure 2). The relative contributions of maltose, maltotriose and glucose in relation to the sum of the quantified fermentable sugars and limit dextrins (59–65%, 20% and 12%) is in accordance with the data covered by Kunze^[2] (65%, 17.5%, 17.5%). However, relatively high extract content was also found in the sweet wort produced at mashing-in pH 6.0 (24.4 °P). The optimal pH for α- and β-amylase activity is approximately 5.4. However, at pH 6.0, α-amylase retains about 80% of its activity,^[58] while β-amylase activity remains relatively unaffected.^[59] This may imply that the effect of pH on amylolytic enzymes is of lesser importance after temperature and thickness of the mash.^[3,60] Additionally, the stabilizing effect of calcium ions (80 ppm in this study) can partially be involved in the extended amylolytic activity when mashing-in at pH 6.0.^[61] Furthermore, high levels of glucose determined in the sweet wort produced at mashing-in pH 4.5, may result from enhanced activity of α-glucosidase, which has a pH optimum 4.0.^[62] However, greater extract content does not necessarily imply increased fermentability. This is demonstrated by the sweet wort produced at mashing-in 4.5, where the extract content is relatively high, although the sample has a considerably high content of limit dextrins (see Figure 2 and Table 2).

This investigation demonstrates a significant effect of mash pH conditions on protein and amino acid content in wort solution (Table 4). Existing literature indicate that the highest levels of free amino nitrogen and total soluble nitrogen are found at mash pH of 4.6 and reduce when pH increases.^[1,5] The decrease of soluble protein content at decreasing mash pH (Table 4) can be associated with lowered solubility and/or precipitation. In general, proteins in the pH conditions below or above the protein isoelectric point (pI) display higher solubility, however, when the pH conditions are close to the isoelectric point, proteins display the lowest solubility and precipitate. A comprehensive proteome map for sweet wort produced from pale malt has been developed by Iimure et al.^[63] The observed pI values for proteins found in the sweet wort ranged from 4.5 to 8.7 with protein Z covering the lower spectrum of the observed pI values. Hordeins, the most abundant barley proteins, are neither found in sweet wort nor beer, most likely due to the hydrophobic character of the proteins. However, their smaller polypeptide fractions may be released because of proteolytic activity.^[63] Protein precipitation may also occur as a result of complex formation with i.a. polyphenols. The effect of pH on haze formation was studied by Siebert et al.^[64] in a model system with gliadin and tannic acid. The maximum value of haze reported occurred at pH 4.0 to 4.2 (close to gliadin pI value) and declined as the pH increased or decreased. In addition, the sweet wort produced in this study at mashing-in pH of 4.5 displayed markedly higher cloudiness when compared to the remaining sweet wort samples, which supports the argument for decreased protein solubility/extraction and greater precipitation at the pH of 4.5.

The diminished content of soluble proteins in the sweet wort produced at acidified conditions may not be solely attributable to the decreased protein solubility and/or precipitation but also due to elevated proteolytic activity. Irrespective of the pH conditions during mashing, higher levels of free amino acids than soluble proteins were determined in the produced worts. The ratio between the sum of free amino acids and soluble proteins tended to increase with the decreasing mash pH, pointing to better proteolytic activity at acidic conditions. Clearly, the increased content of amino acids in sweet worts mashed-in at low pH conditions is mainly related to the proteolysis by endopeptidase and carboxypeptidase (Table 4). The reported pH optima for proteolytic enzymes are: endopeptidase 3.9–5.5, carboxypeptidase 4.8 to 4.6, aminopeptidase 7.0 to 7.2, and dipeptidase 8.8.^[65] Endopeptidases and carboxypeptidases furthermore seem to be more heat stable than aminopeptidases and dipeptidases.^[65]

Up to this point one could anticipate that mashing-in at 4.5 would not produce significant challenges from the perspective of brewing efficiency and fermentation processes. Such strategy could benefit, for instance, brewing with adjuncts poor or lacking in nitrogenous compounds (e.g., glucose-fructose syrup). Higher proteolytic activity at pH 4.5 could potentially compensate the reduced pool of free amino nitrogen and at the same time enhance the buffering capacity of wort necessary for fermentation. However, higher level of free amino nitrogen will have a negative impact on flavor stability of beer, owing to higher availability of amino acid substrates for Maillard reactions and Strecker degradation (valine, leucine, isoleucine, methionine, and phenylalanine).^[24] Regarding polyphenols, our study did not show any significant effect of pH conditions during mashing on their levels. Statistical analysis revealed no notable alteration in total polyphenol content and proanthocyanidins despite variations in pH conditions during mashing (Table 3).

Regarding the staling properties of the produced worts, the volatile aldehyde composition was affected by the pH conditions during mashing with varying trends depending on the compound (Figure 7). Again, the sweet wort sample produced at mashing-in pH of 4.5 was significantly different in volatile aldehyde levels when compared to other samples. In general, the levels of aldehydes found in this study are comparable to pilot-scale brewing trials,^[49,66] indicating good reproducibility irrespectively of the experimental scale. Considering that the malt used for the analysis had relatively high LOX-activity, very low levels of trans-2-nonenal were observed in the sweet worts, most likely due to a high mashing-in temperature at 63 °C, leading to immediate thermal inactivation of LOX. A greater formation rate for furfural in acidified conditions aligns with Yu and Zhang.^[67] The varying trends observed among Strecker aldehydes could suggest different mechanisms behind their formation. While the saturated Strecker degradation aldehydes (2-methylpropanal, 2-methylbutanal, 3-methylbutanal) may be formed by oxidative pathways, then methional and phenylacetaldehyde seem to fall into the category of the products of non-oxidation pathways.^[24] In addition, substrate-dependence could be linked with the formation pathways for phenylacetaldehyde, methional and furfural, as higher levels of these carbonyls coincided with the higher content of methionine, phenylalanine (and amino acids in general), which was not the case for saturated Strecker aldehydes. However, matrix effects may eclipse the absolute levels of volatile aldehydes present in wort, as carbonyl compounds were found to bind with other compounds (e.g., cysteine) especially during early stages of the brewing process^[66] and the stability of the adducts is pH-dependent.^[68]

The rate of radical formation was found to increase with the pH levels of the sweet worts and a strong linear correlation was demonstrated between the pH of the final sweet wort and the rate of radical formation determined by ESR (Figure 4). Additionally, the ESR results were supported by the analysis of the rates of oxygen consumption (Figure 5). The two methodologies assess the oxidative properties under different temperature conditions: 60 °C for the ESR spectroscopy and ambient temperature for the oxygen consumption rate. The agreement between the ESR spin trapping results and the oxygen consumption data implies that the observed increasing rates of oxidation with pH are inherent properties of the worts, and not due to artifacts caused by the analytical methods. Furthermore, Kunz et al.^[69] previously established a positive correlation between rates of radical formation determined by ESR and the pH in beer. It is not clear whether the observed effect is solely attributable to the pH itself or if other matrix factors, resulting from mashing at varying pH, are also involved. To investigate this further, the ESR experimental set up was extended with a supplementary set of samples, where pH in the final sweet wort were adjusted, as schematically presented in Figure 6. The results showed that, in terms of radical formation, the pH effect per se is predominant in the alkalinized sweet worts with a final pH of 5.7. No significant differences were observed in rates of radical formation when comparing the high pH wort samples irrespective of the pH during mashing (Figure 6). However, the rates of radical formation differed among the worts adjusted to a final pH of 4.6 indicating a combined effect of both pH and the inherent matrix composition. Notably, the sweet wort produced under acidified conditions (mashing-in pH of 4.5) exhibited a significantly lower rate of radical formation compared to worts with similar pH made by the normal and alkalinized mashing. This suggests that the acidified wort produced with an initial mashing pH of 4.5 exhibits enhanced antioxidative properties, partially attributed to the changes in the wort composition.

Mash pH had a substantial impact on the levels of transition metal ions in the sweet wort, which is in accordance with previous studies^[23,70,71] (Figure 3). Ferrous, Fe(II) and cuprous, Cu(I), ions, serving as electron donors, mediate the production of reactive oxygen species (ROS) and radicals through the Fenton and Haber-Weiss reactions^.[25,26,72] While the involvement of manganese Mn(II) has received comparatively less investigation, recent studies have focused on its potential role as a metal catalyst.^[23,37,73] One would anticipate a higher rate of radical formation in the samples with the highest contents of transition metals, thus in the wort samples produced at mashing-in pH of 4.5. Nonetheless, the negative correlations observed between the rates of radical formation determined by ESR and both the ferric reducing potentials and transition metal ion contents are unexpected. Several factors could potentially contribute to this phenomenon. The evaluation of radical formation by ESR is carried out under aerobic conditions where oxygen is not a limiting factor. In addition, in this study Pilsner-type malt was applied, while other investigations^{[23,38,71,74]} have been based on blends of Pilsner-type malt and specialty malts (such as roasted or Caramalt), where the malt roasting lead to higher levels of iron and manganese in sweet worts and lower levels of copper.^[38,75] Irrespective of the mash pH conditions, the FRAP of the produced sweet wort samples were markedly low (Table 3). The ferric reducing potential assessed through the FRAP method for beer samples generally falls within the range of 1 to 8 mM Fe(II),^[76,77] indicating that the FRAP values documented in this investigation are at the lower end of this range. However, the reducing potential may affect oxidative reactions in opposite ways: (1) prooxidative effect by enhanced reduction of trace levels of Fe(III) to Fe(II) further leading to free radical generation and (2) antioxidative properties due to the ability of reducing substance being oxidized instead of other compounds. The reducing power of polyphenols are expected to increase with pH as they begin to be deprotonated forming more easily oxidized phenolate ions and thereby become more reducing. However, we observed the highest FRAP reducing potential in the wort produced at the lowest mashing-in pH. Van Mieghem et al.^[71] have shown that iron is mainly present as Fe(II) in beer, while a mixture of Fe(II) and Fe(III) is present in wort, indicating less reducing conditions, which is in agreement with low FRAP values measured in the wort. While polyphenols and reductones are known for their reducing properties,^[69,78,79] they do not appear to fully account for the observed FRAP reducing potentials.

Conclusions

This study highlights the significance of the buffering capacity of natural wort. Even when attempts are made to disturb the buffering system by acidification or alkalization at mashing-in, the pH tends to return to its original state and align with the natural, unadjusted mash pH. As a result of buffering capacity, the produced wort samples display similar biochemical and staling properties, regardless of pH adjustment at mashing-in within the range 5.0 to 6.0. The performance of the mashing process and the characteristics of the resulting sweet wort are primarily influenced by the time/temperature factor, with pH playing a secondary role.

However, mashing with the very low pH of 4.5 at mashing-in, results in the production of sweet wort, which stands out as an outlier with several attributable downsides i.a. lower extract yield, higher dextrin content, protein precipitation, higher wort cloudiness, higher solubilization of transition metal ions, higher reducing potential. Conversely, higher content of amino acids in such produced worts could potentially compensate low level of amino acids when brewing with adjuncts, which are poor or lacking in assimilable amino nitrogen.

Finally the results of this study indicate that the rate of radical formation does not necessarily rely on the reducing potential and the content of transition metal ions. Regarding the mash pH effect, sweet worts produced from 100% Pilsner-type of malt behave differently in terms of their oxidative properties than sweet worts produced from malt blends with speciality malts, which could result from higher proportion of ferric Fe(III) ions in relation to ferrous Fe(II) ions. Further research elucidating the behavior of pro-oxidative and antioxidative compounds (especially iron species, reductones, polyphenols) and their reactivity under different pH conditions is necessary for better understanding of the underlying chemistry behind oxidative reactions occurring during mashing.

Acknowledgements

The authors would like to thank Henriette Erichsen from the Department of Food Science, Ingredient, and Dairy Technology, University of Copenhagen, for valuable analytical support in this study.

Disclosure statement

The authors declare there is no conflict of interest.

Table 3. Wort color, total polyphenols, flavonoids, proanthocyanidins and FRAP reducing potential of the sweet worts produced at varying mash pH conditions. Capital letters in the superscripts (A, B, C) indicate statistically distinguished groups by one-way analysis of variance (ANOVA) with a Tukey post hoc test at p < 0.05.

pH at mashing-in [-]	4.42 ± 0.06 ^E	5.01 ± 0.01 ^D	5.23 ± 0.03 ^C	5.43 ± 0.05 ^B	5.99 ± 0.03 ^A
pH at mashing-off [-]	4.60 ± 0.05 ^E	5.26 ± 0.04 ^D	5.36 ± 0.04 ^C	5.44 ± 0.01 ^B	5.66 ± 0.02 ^A
Color [EBC]	16.66 ± 2.99 ^A	11.11 ± 0.51 ^B	10.96 ± 0.64 ^B	11.19 ± 1.06 ^B	12.10 ± 1.12 ^B
Total Polyphenol Content [mg/L]	134.89 ± 17.92 ^B	165.12 ± 11.10 ^A, B	166.34 ± 4.32 ^A, B	172.58 ± 15.93 ^A	162.24 ± 7.70 ^A, B
Flavanoids [mg/L]	49.65 ± 2.47 ^A	46.15 ± 1.15 ^A, B	42.51 ± 2.68 ^B, C	41.05 ± 2.74 ^B, C	37.28 ± 3.43 ^C
Proanthocyanidins [mg/L]	109.20 ± 11.81 ^A, B	123.40 ± 9.30 ^A	98.81 ± 3.97 ^B	109.37 ± 10.07 ^A, B	90.56 ± 2.99 ^B
FRAP reducing potential [mM Fe^2+]	2.30 ± 0.14 ^A	1.67 ± 0.21 ^B	1.52 ± 0.22 ^B	1.60 ± 0.12 ^B	1.48 ± 0.09 ^B

pH measurements were carried out at 63 °C. Data represent mean and standard deviations (n = 3). Capital letters in the superscripts (A, B, C) indicate statistically distinguished groups by one-way analysis of variance (ANOVA) with a Tukey post hoc test at p < 0.05.

Additional information

Funding

The project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No. 722166.