Comparison of the impact on the performance of small-scale mashing with different proportions of unmalted barley, Ondea Pro, malt and rice

Transcription

1 Research article Received: 28 September 2015 Revised: 4 February 2016 Accepted: 17 March 2016 Published online in Wiley Online Library: 27 April 2016 (wileyonlinelibrary.com) DOI /jib.325 Comparison of the impact on the performance of small-scale mashing with different proportions of unmalted barley, Ondea Pro, malt and rice C. M. Cooper, 1 D. E. Evans, 1,2 *A.Yousif, 3 N. Metz 4 and A. Koutoulis 1 The impact of using different combinations of unmalted barley, Ondea Pro and barley malt in conjunction with a 35% rice adjunct on mashing performance was examined in a series of small scale mashing trials. The objective was to identify the potential optimal levels and boundaries for the mashing combinations of barley, Ondea Pro, malt and 35% rice (BOMR) that might apply in commercial brewing. Barley and malt samples used for the trials were selected from a range of Australian commercial barley and malt samples following evaluation by small-scale mashing. This investigation builds on previous studies in order to adapt the technology to brewing styles common in Asia, where the use of high levels of rice adjunct is common. Mashing with the rice adjunct, combined with differing proportions of barley, Ondea Pro and malt, resulted in higher extract levels than were observed for reference mashing, using either 100% malt reference or 100% barley reference and Ondea Pro enzymes. Synergistic mashing effects between barley, Ondea Pro and malt were observed for mash quality and efficiency parameters, particularly wort fermentability. The optimum levels of barley in the grist (with the relative level of Ondea Pro ) were assessed to be in the range 45 55% when paired with 10 20% malt and 35% rice. When the proportion of malt was reduced below 10% of the grist, substantial reductions in wort quality were observed for wort quality parameters including extract, lautering, fermentability, free amino nitrogen and haze. Extension of this new approach to brewing with rice adjuncts will benefit from further research into barley varietal selection in order to better meet brewer s quality requirements for the finished beer. Copyright 2016 The Additional supporting information may be found in the online version of this article at the publisher s web-site. Keywords: Beer; barley; Ondea Pro ; malt; rice adjunct; wort quality; mashing 218 Introduction Beer is conventionally brewed with barley malt, often with the inclusion of a non-malted starch adjunct (1,2). The cost of malt has been reported as being approximately 3.5% of total beer production costs, with hops, energy, water, labour, packaging, capital, distribution and taxation costs making up the remainder (3). Despite the relatively low proportional cost of malt, there has been considerable effort within the brewing industry to produce beer from unmalted grains, included in the mash either as adjuncts or as a total malt replacement. Several recent publications have identified a range of drivers for brewing with unmalted grains (3,4). The drivers reported included avoiding the need to use low-quality malting facilities in developing countries, reduced energy and water costs associated with malting, and the avoidance of malt taxes in some countries. Steiner et al. (4) additionally referred to some brewhouse advantages for using unmalted barley in the mashing process, including the reduction of haze and of flavour staling precursors associated with lower nitrogen levels compared with conventional malt mashing. Rice, typically broken kernels, is commonly used in Asia for the production of alcoholic beverages both as an adjunct in brewing and for beverages where rice is the sole cereal source, including Japanese Saki and Chinese shaoshin-chue (5). In addition,very high rice adjunct beers (>75% rice) called happoshu are produced in Japan to reduce taxation, which is determined as a function of the malt threshold used (5,6).Yuet al.(7) have reported using rice adjunct as 35% of grist in their Chinese pilot brewing trials. Le Van et al. (8) reported that in South Asia rice is the most commonly used adjunct, and ratios of 30-50% in the grist are common in many countries. Typically the rice adjunct will be heated beyond the starch gelatinization point (75 80 C) for rice in combination with a liquefaction or amylase-type enzyme in the rice cooker (9). The advantages to brewing with a rice adjunct include lighter colour, improved beer colloidal stability based on reductions in protein and polyphenol levels, and the ability to use rice content for new product differentiation (8). It has also been reported that rice adjunct beers yield a light clean-tasting beer that * Correspondence to: Evan Evans, School of Biological Sciences, University of Tasmania, Private Bag 55, Hobart, Tasmania 7001, Australia. tassiebeerdr@gmail.com 1 School of Biological Sciences, University of Tasmania, Private Bag 55, Hobart, Tasmania 7001, Australia 2 The Tassie Beer Dr Consulting, 15 Rianna Rd, Lindisfarne, Tasmania 7015, Australia 3 Australian Export Grains Innovation Centre, 3 Barron-Hay Court, South Perth, Western Australia 6151, Australia 4 South East Premium Wheat Growers Association, PO Box 365, Esperance, Western Australia 6450, Australia J. Inst. Brew. 2016; 122: Copyright 2016 The

2 Impact on the performance of small-scale mashing is well suited to warm and humid climates (5).Furthermore,Goode et al. (3) reported that the use of rice in the production of American pale lager impacts flavour, colour and colloidal stability in a way that meets the traditional expectations of the market. An important issue associated with the use of rice adjuncts is the negative impact on fermentation associated with low wort free amino nitrogen (FAN) levels (8). Le Van et al. (8) reported improved fermentation outcomes with a 40% rice adjunct through either addition of a supplemental yeast extract or an exogenous proteinase. Getting the balance right in terms of wort fermentability, extract production and the production of beer with good sensorial qualities when using high levels of unmalted adjuncts and exogenous enzymes is a complex undertaking. The use of exogenous enzymes for brewing with reduced levels of malt has a rich history in beer research that dates back to the 1960s (10). Wiegandcolleagues developed a system called brew-n-zyme that required 15 25% malt in the grist to be feasible (11,12). In the 2000s Goode et al. (3) found that careful optimization of exogenous enzyme additions and mashing conditions allowed good quality wort to be produced without the use of malt. Kunz and co-workers (13) examined the oxidative stability of beer produced with varying barley content and the use of an enzyme mixture that included a protease as well as starch and cell wall degrading enzymes. The results indicated that brewing with enzymes and up to 50% barley produced a beer with increased anti-oxidative potential, good extract and attenuation levels. They also reported that lower FAN levels in wort and beer could lead to less Strecker aldehyde production during beer ageing. More recently, Ondea Pro has emerged as a suitable commercial enzyme mixture for barley brewing and brewing with other grains. The enzyme mixture includes an appropriate range of enzymes for the replacement of malt enzymes including for saccharification, cell wall degradation, amino acid release and lipid metabolism (4,10,14). In particular it has been noted that the inclusion of a lipase for the hydrolysis of triglycerides was a significant advance over previous endogenous enzyme formulations (4). Steiner et al. (4) examined the utility of Ondea Pro for the production of high-quality wort and beer. These authors concluded that the beer produced was not significantly different from conventionally brewed beer at an organoleptic level and that it had better foam stability despite a lower nitrogen content. They also suggested that more detailed investigations were required regarding the possible influence of barley quality on barley brewing with Ondea Pro. Evans et al. (10) extended that knowledge to show that high quality wort can be consistently made from mashing with barley and Ondea Pro. These authors also concluded that barley brewing with this enzyme mixture generally produced wort with slightly lower levels of extract and fermentability than obtained with conventional malt mashing. These differences could be minimized by using barley varieties with the high thermostable Sd2H β-amylase type (15). This enables extended β-amylase activity at higher temperatures, when α-amylase is still active, resulting in more extensive starch breakdown than would be possible for the intermediate Sd1 type varieties, which make up the majority of commercial malting barley varieties. There are also low thermostability Sd2L type varieties, which produce worts with lower fermentability. Ondea Pro has also been used to replace malt enzymes when mashing with up to 40% unmalted oats. Favourable results with respect to mash processability and wort quality were achieved with this grist mixture (16). However, it was also observed that, with levels of up to 20% oats, good-quality wort could be produced without the use of exogenous enzymes. Ondea Pro is currently being used for brewing commercial barley beers (10,14) and recently a Nordic Council of Minister s report described 100% barley brewing with this enzyme preparation as one of the best available new technologies for brewing (17).However,in Asia, where large amounts of rice adjunct are often used in brewing (18), the addition of barley without added enzymes is limited in practice owing to the already reduced malt enzyme proportion in the grist. In addition, brewers are sensitive to issues regarding colour and flavour matching for established brands. In summary, the literature suggests that: (a) good quality wort and beer can be produced with 100% barley and Ondea Pro (4,10,14); (b) good quality wort has been especially associated with Sd2H thermostable β-amylase in the barley (10); (c) Ondea Pro has been demonstrated to have utility when mashing with high levels of adjunct (16); and (d) the brewing industry recognizes a range of economic and sustainability opportunities in such practices (17).In order to extend that knowledge and build on the previous study by Evans et al. (10), this investigation examines the use of Ondea Pro with varying levels of barley and malt in the presence of rice adjunct at 35%. This combination has particular relevance to brewing in Asia, where high levels of rice adjunct are often used (7,18). The impact on brewing performance was assessed in a series of small scale mashing trials. Experimental Chemicals and reagents All chemical reagents were reagent or analytical grade. Megazyme (Bray, Ireland) enzyme kits were purchased from Deltagen Australia (Kilsyth, Australia). Ondea Pro was provided by Novozymes A/S (Bagsvaerd, Denmark). Barley, malt and rice samples Barley samples were obtained from the South East Premium Wheat Growers Association, while malt samples were sourced from Australian commercial maltsters. A commercial Australian medium grain white rice (Sun Rice) was used for the rice adjunct mashing trials. Barley and malt quality analyses Quality analysis data for barley and malt samples were provided by commercial suppliers and had been carried out where applicable according to Analytica EBC (19): moisture (method 4.2), total protein (method 4.3.1), Diastatic Power (DP, method 4.3.1) and Kolbach Index (KI method 4.9.1). These analyses were provided by the suppliers, using a level of replication that was standard for the commercial description of quality. Wort quality and filterability measurements The mashing performance parameters extract, fermentability (apparent attenuation limit AAL), FAN, protein, haze, viscosity and lautering were measured using the protocols outlined in Evans et al. (10). Each wort sample, obtained from a minimum of duplicate mashing samples, was analysed once for each sample of the duplicate except for the measurement of protein and FAN, which were analysed in either duplicate or triplicate. The BOMR lautering time was 12.5 min. Enzyme analyses Barley and malt samples were milled for enzyme extraction using a Cyclone Sample Mill (Udy Corporation, Fort Collins, USA) with a 1 mm mesh screen. 219 J. Inst. Brew. 2016; 122: Copyright 2016 The wileyonlinelibrary.com/journal/jib

3 C. M. Cooper et al. 220 The β-amylase, α-amylase, limit dextrinase and β-glucanase activities were determined using the appropriate Megazyme (Bray, Ireland) kits (Betamyl- 3, Ceralpha, Limit Dextrizyme and β-glucazyme, respectively) using a 96-well microplate method as has been previously described (10,20,21). The limit dextrinase and α-amylase assays were conducted as concordant duplicates; all other assays were conducted in triplicate. Mashing Small-scale mashing trials were conducted using a Bender and Hobein (Ismaning, Germany) mash bath. The barley and malt samples were milled using a Buhler Miag disc mill (Brunswick, Germany) set to a 0.7 mm grind. The rice, used in the barley, Ondea Pro, malt and rice (BOMR) mashing trials, was ground with a Buhler Miag disc mill set to grind at 0.2 mm. Each of the barleys and malts used in the combination trials were mashed separately, in either duplicate or triplicate, to establish the reference wort quality parameters. Samples of the malt were mashed using the Final 65 C malt protocol developed by Evans et al. (22). Briefly, this protocol uses a 0.7 mm disc milled malt (1:3 grist water, including 0.3 mm CaCl 2.2H 2 O) infusion mashed at 65 C for 1 h before ramping to 74 C with a final grist water ratio of 1:6. Barley samples were mashed with Ondea Pro using the protocol outlined in Evans et al. (10), which is briefly described below as a component of the BOMR mashing process. Mashing for the BOMR trials was performed in duplicate and was a modification of the aforementioned barley mashing protocol (10) using an initial rice gelatinization step, and the temperature profile as outlined in Fig. 1, as first used by Evans et al. (23). Accordingly, 150 ml of distilled water containing 0.3 mm CaCl 2.2H 2 O at room temperature (~20 C) was added to 17.5 g of rice in a mash pot with mixing. The temperature was brought to 80 C with constant stirring and held for 10 min before a reduction to 54 C. This was followed by the addition of 32.5 g of the barley/malt grist (0.7mm grind), in the ratios described in Table 1, and the temperature was held at 54 C for 30 min. The mash protocol then followed the detail as described in Fig. 1 and Evans et al. (10). Experimental design The mashing experiment, which used BOMR, was designed to examine the impact of changes in the proportion of barley (and Ondea Pro ), in a grist mixture of malt and rice, on wort quality and mash efficiency. Table 1. Grist ratios for barley, Ondea Pro, malt and rice treatments, control and references used for the BOMR mashing trials Grist treatment Barley (%) Ondea Pro a (%) Malt (%) Rice (%) Mash grist - water ratio : : : : : : : 3 Control : 3 Barley : 3 reference Malt reference : 3 a 100% represents 1000 μl of a 10% solution of Ondea Pro. Each addition of Ondea Pro gave a final value of 2 kg/tonne of barley in the grist. The changes to the quality parameters were studied using five different varietal combinations of barley, and malt as the proportion of barley in the grist mixture was increased with a corresponding decrease in the proportion of malt as outlined in Table 1. The selection of appropriate barley and malt samples for these mashing trials was achieved through first examining the grain characteristics and wort quality parameters of the available range of barleys (n = 6) and commercial or pilot malted Australian malts (n = 6). The wort quality parameters for either 100% barley or 100% malt were used to select the BOMR combinations, which then became the reference levels for comparison to the BOMR outcomes (see Table 1). Malt and barley samples for the BOMR trials were selected with the intention of selecting combinations that maximized the differences between malt and barley for characteristics such as extract, fermentability and DP enzyme levels. Each varietal combination was mashed in duplicate using the BOMR protocol. The level of barley in the mash was increased from 0 up to 65%, while malt was decreased from 65 to 0%. A constant grist proportion of 35% rice was used as this is the proportion that the major Chinese brewer Tsingtao had previously used in their pilot brewing trials (7). The grist proportions (Table 1) were designed to examine the contribution to mashing by barley and malt enzymes, while keeping the ratio of Ondea Pro to barley constant at 2 kg/tonne of barley. This level of enzyme was commensurate with the ratio previously used by Evans et al. (10) and is at the upper end of what has been recommended by Novozymes (the manufacturer). A control with 20% barley without Ondea Pro, 45% malt and 35% rice was also included for each barley and malt combination. Statistical analyses One-way analysis of variance (ANOVA) and regression data were obtained using R software (24). Fisher LSDs (least significant differences) were calculated from mean square errors obtained from the ANOVA analysis. BOMR treatments were considered significantly different when a mean was different from another mean by more than the LSD (p < 0.05). Results Barley and malt characteristics/wort quality parameters The barley and malt characteristics are presented in Table 2, and the wort quality parameters are shown in Table 3, for all of the commercial samples tested for inclusion in the barley, Ondea Pro, malt and 35% rice (BOMR) mashing trials. Overall, the malt and barley characteristics (Table 2) and wort characteristics (Table 3) were within the range expected for commercial barley and malted samples (10,25). The main principle applied for the selection of barley and malt for the BOMR mashing trials was divergence or similarity in the key quality characteristics of extract and fermentability (AAL) for the barleys and malts described in Tables 2, 3. For example, with regard to extract, the Buloke-2 malt and Hindmarsh barley combination had divergent extract levels of 75.4% for Hindmash barley and 80.5% for Buloke-2 malt. In contrast, for the Scope barley and Commander malt combination the levels of extract were more similar at 79.0 and 78.2%, respectively. In addition, the Hindmarsh barley used for these mashing trials had the highest level of β-amylase activity (1226 U/g d.w.), while Scope barley had the highest extract level (79.0% d.w.), but also the highest haze 8.15 ( EBC) of the barley samples. The Commander malt had a high level of β-amylase (921 U/g d.w.), a low level of α-amylase (170 U/g d.w.) activity, and the lowest limit dextrinase (247 U/kg d.w.) activity, wileyonlinelibrary.com/journal/jib Copyright 2016 The J. Inst. Brew. 2016; 122:

4 Impact on the performance of small-scale mashing Figure 1. Temperature profile for mashing with barley, Ondea Pro, malt and rice. (a = addition of rice at room temperature; b = addition of barley/ondea Pro /malt at 54 C). Table 2. Barley and malt quality characteristics assessed for the BOMR mashing trials Variety Moisture (%) Protein (%) KI (%) DP ( WK) α-amylase (U/g d.w.) Total limit dextrinase (U/kg d.w.) Total β-amylase (U/g d.w.) β-amylase Sd type β-glucanase (U/kg d.w.) Barley (n = 6) Gairdner n.d. n.d. < Sd1 72 Scope n.d. n.d. < Sd2H 55 Bass n.d. n.d. < Sd1 49 Buloke n.d. n.d. < Sd2H 65 Baudin n.d. n.d. < Sd1 73 Hindmarsh n.d. n.d. < Sd2H 60 Mean n.a. 62 LSD n.d. n.d. n.s. 40 n.a. n.s. Malt (n = 6) Buloke Sd2H 238 Buloke Sd2H 320 Buloke Sd2H 260 Gairdner Sd1 285 Commander Sd1 230 Buloke Sd2H 234 Mean n.a. 261 LSD n.d. n.d. n.d. n.d n.a. 27 n.d., not determined; n.a., not available, n.s., not significant; LSD, least square difference, p < compared with the other malts tested. In comparison with the Commander malt, the Buloke-2 malt had a much lower level of β-amylase (836 U/g d.w.), but substantially higher levels of α-amylase (220 U/g d.w.) activity and limit dextrinase (415 U/kg d.w.) activity. The KI of the Buloke-2 malt (45.3%) was also substantially higher than that of the Commander malt (37.0%). Barley and malt combinations for the trial were either both Sd2H type β-amylase or an Sd2H variety paired with an Sd1 variety. The data in Tables 2 and 3 allow a fuller comparison of the differences and similarities between the barleys and malts used for all five BOMR mashing trials. Barley, Ondea Pro, malt and rice mashing The data from two divergent barley and malt combinations: (i) Hindmarsh (Sd2H barley) and Buloke-2 (Sd2H malt); and (ii) Scope (Sd2H barley) and Commander (Sd1 malt) in Figures 2 4, were 221 J. Inst. Brew. 2016; 122: Copyright 2016 The wileyonlinelibrary.com/journal/jib

5 C. M. Cooper et al. Table 3. The wort quality parameters of the barleys and malts assessed for the BOMR mashing trials Variety Extract (% d.w.) AAL (%) Lauter a (ml) Viscosity (mpa s) Haze ( EBC) FAN (mg/l) Bradford prot (mg/l) ph Barley (n = 6) Gairdner Scope Bass Buloke Baudin Hindmarsh Mean LSD Malt (n = 6) Buloke Buloke Buloke Gairdner Commander Buloke Mean LSD a Malt mash lauter time 25 min and barley mash lauter time 12.5 min. AAL, Apparent attenuation limit; FAN, free amino nitrogen; LSD, least square difference, p < selected for presentation in this paper. The wort quality and brewing performance data for the two representative BOMR mashing trials are presented in Figs 2 4. The complete data set for the five combinations is presented in the Supporting Information section (Figs S1 S7). Extract and fermentability Importantly, the brewer desired parameter, extract yields, were higher than that obtained for the 100% barley or malt mashing alone references, at all levels of barley in the grist (Fig. 2) for all BOMR combinations. Two distinct patterns were observed for changes in the level of extract. Firstly, for the Hindmarsh barley/buloke-2 malt combination, the level of extract was found to progressively decrease from 84.9 to 81.0% as the barley levels were increased to the maximum proportion of 65% in the grist. This pattern was typical of all BOMR combinations (Supporting Information, Fig. S1) except for the Scope barley/commander malt combination. In the latter combination, the level of extract showed a small but significant increase from 83.5 to 84.1% when barley and Ondea Pro were initially added to the grist at a level of 15% barley. This slightly higher level was maintained until malt was completely removed from the grist (65% barley), at which point the extract level showed a significant decrease to 83.6%. The control mashes (no Ondea Pro ) showed substantially and significantly lower levels of extract than for all BOMR combinations tested indicating a significant Ondea Pro contribution to extract yield. It was noted that the level of extract for the control was substantially higher than that observed for barley or malt reference levels in all BOMR combinations (Fig. 2; Supporting Information, Fig. S1). Fermentability (AAL) increased to an apparent optimum when the barley levels in the grist were raised from zero up to a level of 45 55% barley in the mash, after which, they declined (Fig. 2). In some varietal combinations (e.g. Scope barley and Commander malt), the level of fermentability increased to a level above that observed for the reference barley or malt alone mashes. In other combinations (e.g. Hindmarsh barley and Buloke-2 malt), the level of fermentability did not reach the level of the 100% malt mashing alone reference. The fermentability increased above the level of the 100% malt mashing alone reference in the two BOMR combinations where the Sd1 (intermediate thermostability β-amylase) malt variety Commander (Fig. 2B; Supporting Information, Fig. S2) was used in combination with an Sd2H (high thermostability β-amylase) barley variety. Furthermore, the maximum fermentability percentages for all BOMR combinations were in a relatively narrow range of %, despite a greater variation for the 0% barley and 100% malt brewing fermentabilities, which were in the range %. The control demonstrated a significant and substantial lowering of fermentability when the enzyme was excluded from the mash, for all BOMR combinations. FAN and protein FAN and protein data for the two representative BOMR combinations are presented in Fig. 3. Initially, there were no significant decreases in FAN for all BOMR combinations, when the barley level was increased from 0 to 15% in the grist. At higher proportions of barley in the grist, decreases in FAN were significant and overall the rate of decline in FAN levels appeared to accelerate. FAN levels were maintained above the level of the barley mashing alone reference until the barley level reached 45 55% barley/ondea Pro in the grist. For all barley and malt combinations the BOMR FAN levels were consistently below the level observed for the 100% malt mash reference at all barley/ondea Pro levels (Fig. 3; Supporting Information, Fig. S3). A small decrease in FAN was observed for the Ondea Pro excluded control (20% barley, 45% malt, 35% rice). wileyonlinelibrary.com/journal/jib Copyright 2016 The J. Inst. Brew. 2016; 122:

6 Impact on the performance of small-scale mashing Figure 2. The impact on extract and apparent attenuation limit (AAL) achieved from two different barley and malt combinations mashed with a 35% rice adjunct. (A) Extract and (B) AAL, for the BOMR combinations (i) Hindmarsh (Sd2H barley) and Buloke-2 (Sd2H malt) and (ii) Scope (Sd2H barley) and Commander (Sd1 malt). LSD, Least square difference (p < 0.05). The dashed and dotted lines are the reference data that correspond to mashing with either 100% reference barley and Ondea Pro or 100% reference malt, respectively. The point at 20% barley is the control without Ondea Pro. The level of wort protein generated by varying the levels of barley and malt, with or without Ondea Pro, was relatively unresponsive (Fig. 3). Changes in protein levels as the percentage of barley was increased were not significant in four of the five barley/malt combinations (Fig. 3: Supporting Information, Fig. S4). Analysis of the Scope barley/commander malt combination did show significant differences for protein at some barley levels but overall these differences were small in magnitude. Figure 3. The impact on free amino nitrogen (FAN) and protein achieved from two different barley and malt combinations mashed with a 35% rice adjunct. (A) FAN and (B) protein, for the BOMR combinations (i) Hindmarsh (Sd2H barley) and Buloke-2 (Sd2H malt) and (ii) Scope (Sd2H barley) and Commander (Sd1 malt). LSD, Least square difference (p < 0.05). The dashed and dotted lines are the reference data which correspond to mashing with either 100% reference barley and Ondea Pro or 100% reference malt, respectively. The point at 20% barley is the control without Ondea Pro. 223 J. Inst. Brew. 2016; 122: Copyright 2016 The wileyonlinelibrary.com/journal/jib

7 Viscosity, haze and lautering The viscosity, haze and lautering data of the two representative BOMR mashing combinations are presented in Fig. 4. The largest impacts on these parameters were observed when either barley and Ondea Pro were initially added to the grist or when malt was excluded from the grist. The largest reduction in viscosity occurred when the barley level in the grist was increased from 0 to 15%. This was followed by smaller decreases as the percentage of barley in the grist was increased. All of the grist mixtures that contained barley and Ondea Pro showed lower levels of viscosity C. M. Cooper et al. than was observed for the 100% barley reference mash. Viscosity levels were also significantly lower than observed when Ondea Pro was excluded in the control grist. A statistically significant reduction in haze occurred when the barley levels in the grist were increased from 0 to 15%. When the barley levels were further increased up until 45% barley in the grist, the changes in haze were either marginally different or not significantly different. However, increases above 45% in the barley level resulted in significant increases in haze and the largest changes in wort haze were observed when malt was completely removed from the grist. The levels of haze in the grist mixture 224 Figure 4. The impact on FAN and protein achieved from two different barley and malt combinations mashed with a 35% rice adjunct. (A) Viscosity, (B) haze and (C) lautering, for the BOMR combinations (i) Hindmarsh (Sd2H barley) and Buloke-2 (Sd2H malt) and (ii) Scope (Sd2H barley) and Commander (Sd1 malt). LSD, Least square difference (p < 0.05). The dashed and dotted lines are the reference data which correspond to mashing with either 100% reference barley and Ondea Pro or 100% reference malt, respectively. The point at 20% barley is the control without Ondea Pro. wileyonlinelibrary.com/journal/jib Copyright 2016 The J. Inst. Brew. 2016; 122:

8 Impact on the performance of small-scale mashing containing 65% barley/35% rice returned to a level close to that observed for the 100% barley alone, reference mash. By comparison much smaller changes, but significant in four out of five BOMR combinations (Supporting Information, Fig. S6), in the haze levels were observed for the control samples where Ondea Pro was excluded from the grist. Initially, a significant increase in the lautering volumes was observed when the barley levels were increased from 0 to 15% in the grist for all BOMR combinations (Fig. 4; Supporting Information, Fig. S7). The lautering volumes either slightly increased or stayed the same, up to a level of 55% barley, after which there was a significant reduction in lautering capacity when malt was removed from the grist. The lautering volumes were also higher than the levels observed for either the malt or barley/ondea Pro alone reference baselines until malt was removed from the grist. In the absence of Ondea Pro significant reductions in lautering volumes were observed. The sudden decrease in lautering efficiency observed when malt was completely removed from the grist was mirrored to some extent by the increase in haze at higher concentrations of barley. Discussion Recent research has demonstrated that barley brewing using Ondea Pro yields good quality beer (4,10,14). In this study,the possibility of adapting Ondea Pro for brewing beer with high barley and rice adjunct levels was examined. A key aim was to identify the optimal levels and boundaries for the mashing combinations in the BOMR trials. High-quality barleys and malts were used for these mashing trials and their quality characteristics (Table 2) were within the ranges previously reported by Evans et al. (10). The barley wort quality parameters (Table 3) were also similar (10), except for the FAN levels, which were at the lower end of the previously recorded range with one value recorded below the range at 44 mg/l. This was perhaps not surprising given the limited number of barleys and malts available for this investigation (n = 6) compared with the 137 barley samples in Evans et al. (10). The wort parameters for the malts (Table 3) were also within the previously determined ranges except for some small differences for Bradford protein and ph levels. Analysis of the BOMR mashing data identified that an optimal barley level of 45 55% in the grist was consistently apparent through examination of several wort quality parameters. This was especially evident through consideration of wort extract and fermentability (Fig. 2), and FAN (Fig. 3), which are important measures that guide brewers in their choice of grain for brewing (26). The extract levels were consistently higher than that observed for either the 100% barley brewing with Ondea Pro or 100% malt mashing references (Table 3). In addition, the extract levels of wort produced in the BOMR mashing trials with barley levels >55% in the grist were higher than the maximum value of 81.0% reported for barley mashing with Ondea Pro of 137 barley varieties by Evans et al. (10). This underlines the desirability, in terms of extract yield, of including rice in the grist, particularly in the presence of Ondea Pro. Overall, the levels of extract found in these mashing trials at all barley levels compared with levels that have been previously reported for rice adjunct brewing. Yu et al. (7) reported values of 81.0 and 83.1% for pilot scale rice adjunct brewing with the Canadian LOX1-less variety Polar Star and AC Metcalfe malts, respectively. Evans et al. (23) found that, for small-scale mashing trials with rice adjunct brewing, extract levels varied from approximately 81 to 85% d.w. with different malt barley varieties, which were higher extract percentages than those recorded for the individual 100% malt references without rice. Fermentability peaked at 45% barley in the grist but decreased significantly after malt was completely removed from the grist (Fig. 2). The peak AAL levels, in the narrow range of %, were above the levels previously observed by Evans et al. (10) for 100% barley brewing with Ondea Pro (mean = 80.7% for 137 samples) and were more commensurate with levels observed for malt brewing (mean = 90.5% for 93 samples). This level of fermentability occurred despite variations in the 100% malt reference wort fermentability (Table 3) and the relatively large differences in the constituent barleys and malts in terms of key parameters such as extract, fermentability and the DP enzymes (Tables 2, 3). Consequently, it may not be possible to extrapolate malt fermentability as a definitive indicator of BOMR wort fermentability in this mashing system as it would be for conventional mashing. However, it is likely that the use of lower quality barley or malt in the BOMR system will incur brewing yield and wort quality penalties. In addition, FAN levels (Fig. 3) remained above the level observed for brewing that variety of barley with Ondea Pro alone (100% barley reference), up until the optimum barley level. This added further support for the notion that there was a definite advantage associated with the inclusion of malt in the mash. The observations of changes to haze and lautering (Fig. 4) were also consistent with a benefit from including some malt in the grist. The finding that wort quality was improved through the inclusion of 10 20% malt in the grist is comparable with the observations from early forms of barley brewing, where 15 25% malt in the grist was required (10).The brew-n-zyme barley brewing system devised by Wieg (11,12) used 20% malt for mashing. More recent research (3,13), prior to the emergence of Ondea Pro for 100% commercial barley brewing (4,10), also highlighted the advantages of using a combination of malt and barley for brewing with enzymes. In addition, the use of rice as a high starch adjunct, which does not contribute enzymatically to the mashing process, appears to require malt enzymes, particularly for improved fermentability, maintenance of FAN levels, haze reduction and efficient lautering. The haze levels in particular require further detailed analysis. The low wort haze levels produced by the control and 65% malt grist (both without any added Ondea Pro ) when compared with the 100% malt reference, indicated that malt enzymes may be more important for haze reduction than the inclusion of Ondea Pro (Fig. 4, Bi and Bii). The apparent importance of malt in the BOMR mash is consistent with the observation that the changes in wort quality parameters may have involved a synergism in the action of enzymes from the barley, Ondea Pro and malt for this brewing method. Significant improvements were observed, when the data was compared with a control mash without enzyme addition, for all quality parameters except protein (Figs 2 4). Whilst Ondea Pro had a significant role in the BOMR mashing system it was observed that AAL, haze and viscosity showed large decreases in quality once malt was removed from the grist (Figs 2, 4). Furthermore, the absence of linear changes to the parameters as the level of barley (plus Ondea Pro enzymes) was increased in the grist indicated a level of enzyme interaction. Synergistic interactions between enzymes in malt for fermentability and lautering (27) and at a molecular level in saccharification (28) have been reported. Therefore, it is very likely that the enzymes from the three different sources are working in a complementary way to improve the functional outcomes observed in these mashing trials. 225 J. Inst. Brew. 2016; 122: Copyright 2016 The wileyonlinelibrary.com/journal/jib

9 226 The use of good-quality malt with this brewing style may lead to benefits for brewers based on the synergies observed. The low FAN levels observed in these mashing trials were consistent with observations that rice contributes little FAN to wort (1) and the generally lower levels of FAN reported for 100% barley brewing when compared with traditional malt mashing (10). However, the absence of reduction in fermentability observed in these trials, at barley levels of up to 55%, was evidence that the lower FAN levels are sufficient for yeast nutrition. This observation may be related to either the availability of specific amino acids produced in the wort (14) or other types of assimilable nitrogen including ammonium ions and small peptides (29). According to Aastrup (14), barley brewed with Ondea Pro yields a wort containing a higher proportion of absorbable amino acids based on the original classification by Jones and Pierce (30). In addition, barley wort has a much lower level of proline than malt wort (14). Proline is not absorbed by yeast cells during fermentation due to a range of metabolic reasons (29). Hence a higher level of metabolizable amino acids in the wort may help explain the observation that fermentability did not begin to decrease until wort FAN dropped below the level observed for mashing with barley alone. Other factors may be contributing to fermentability. Evans et al. (10) suggested that the higher levels of free fatty acids normally found when brewing barley with Ondea Pro may ameliorate the impact of low FAN levels on yeast nutrition. An alternative solution for overcoming the nitrogen limitation associated with the low FAN levels was identified by Lei et al. (31) whereby problems with nitrogen limitation associated with high gravity worts were ameliorated through supplementation of wort with proteases during the fermentation process. These authors found that proteinase supplementation increased FAN levels during fermentation, improved fermentability, and produced a beer with little impact on beer foam stability and with good flavour outcomes. Pilot scale trials are needed to determine whether the results for small scale-fermentability, where an excess of yeast is used, are transferable to a brewery process. Further laboratory or pilot scale experiments may also provide a more detailed understanding of the ratio of barley to malt required to maximize production of a good quality beer for specific brewers. Furthermore, whilst the wort quality observations indicated that consistent quality outcomes might be possible with a wide range of barley and malt combinations some differing trends were observed that might indicate an advantage to combining particular barley and malt varieties. In particular, an understanding of the reasons for the improved trend in extract level for the Scope barley and Commander malt combination as the barley level was increased, might be useful in guiding varietal choice. Evans et al. (10) found that for 100% barley mashing with Ondea Pro, β-amylase thermostable Sd2H type barley varieties yielded higher extract and fermentability outcomes when compared with the less thermostable Sd1 and Sd2L type varieties. However, the Sd2H Scope barley combined with Sd1 Commander malt in these mashing trials appeared to synergistically maintain extract in contrast to the decrease for other malt barley combinations. Future experiments may be useful to guide brewers regarding the choices of barley and malt combinations that provide benefits through specific synergies. Overall this investigation gives support and guidance to the potential uptake of the Ondea Pro technology by brewers that use high levels of rice adjunct. The results are a guide for both further experimental work, pilot scale and brewhouse trials tailored to specific brewer s requirements. The blending of barley, malt and Ondea Pro in rice adjunct brewing had a synergistically beneficial impact on several wort quality parameters, most particularly extract. The study also indicated that, when using 35% rice adjunct, the optimum level of barley and malt in the grist is in the ranges and 10 20%, respectively. A key observation that supported this conclusion was that fermentability increased until barley reached the 45 55% level in the grist. In addition, at these levels of barley malt, FAN levels stayed above that observed when mashing with barley alone, up until about 45% barley. Furthermore, significant improvements in haze and lautering were observed with 10 20% malt. The observation that extract levels for all barley and malt combinations remained within a commercially acceptable range as the barley levels were increased was also supportive of further development for rice adjunct brewing with malt, barley and Ondea Pro. Acknowledgements The Australian Grains Research and Development Corporation ( grants UT00017 and UT00023) provided funding to support this research. Regina David de Julian and Noel Bautista (Novozymes) are thanked for their technical advice and provision of Ondea Pro. Pilot Malting Australia, Joe White Malting, Barrett Burston Malting are thanked for the provision of malt samples while South East Premium Wheat Growers Association is thanked for the provision of barley samples. Thanks to Michelle Williams, Andrew Measham and the Tasmanian Institute of Agriculture, University of Tasmania for technical assistance. References C. M. Cooper et al. 1. Goode, D. and Arendt, E. (2006) in Developments in the supply of adjunct materials for brewing, Brewing: New Technologies, (Bamforth, C. Ed.), pp , Woodhead, Cambridge. 2. Stewart, G. G. (2006) in Adjuncts, Handbook of Brewing (Priest, F. G. and Stewart, G. G. Eds.) 2nd ed., pp , Taylor and Francis, Boca Raton, FL. 3. Goode, D. L., Wijngaard, H. H., and Arendt, E. K. (2005) Mashing with unmalted barley-impact of malted barley and commercial enzyme (Bacillus spp.) additions, Tech. Q. Master Brew, Assoc. Am. 42, Steiner, E., Auer, A., Becker, T., and Gastl, M. (2012) Comparison of beer quality attributes between beers brewed with 100% barley malt and 100% barley raw material, J. Sci. Food Agric. 92, Yoshizawa, K. and Ogawa, Y. (2004) in Rice in brewing, Rice: Chemistry and Technology (Champage, E. T., Ed.) 3rd ed., pp , AACC Press, St Paul, MN. 6. Nakatani, K. (2007) Beer in Japan present and future trends. The story of Happoshu and the Third way, Brew. Distill. Int. 3, Yu, J., Huang, S., Dong, J., Fan, W., Huang, S., Liu, J., Chang, Z., Tian, Y., Hao, J., and Hu, S. (2014) The influence of LOX-less barley malt on the flavour stability of wort and beer, J. Inst. Brew. 120, Le Van, V. M., Strehaiano, P., Nguyen, D. L., and Taillandier, P. (2001) Microbial protease or yeast extract: Alternative additions for improvement of fermentation performance and quality of beer brewed with a high rice content, J. Am. Soc. Brew. Chem. 59, Bradee, L., Duensing, W., Halstad, S., Klimovitz, R., and Laidlaw, A. (1999) in Adjuncts, The Practical Brewer: A Manual For the Brewing Industry, (McCabe, J. T. Ed.) 3rd ed., pp , Master Brewers Association of the Americas, Wautosa, WI. 10. Evans, D., Redd, K., Haraysmow, S., Elvig, N., Metz, N., and Koutoulis, A. (2014) The Influence of malt quality on malt brewing and barley quality on barley brewing with Ondea Pro, compared by small-scale analysis, J. Am. Soc. Brew. Chem. 72, Wieg, A. (1970) Technology of barley brewing, Process Biochem. 5, Wieg, A., Hollo, J., and Varga, P. (1969) Brewing beer with enzymes, Process Biochem. 4, Kunz, T., Müller, C., Mato-Gonzales, D., and Methner, F.-J. (2012) The influence of unmalted barley on the oxidative stability of wort and beer, J. Inst. Brew. 118, wileyonlinelibrary.com/journal/jib Copyright 2016 The J. Inst. Brew. 2016; 122:

10 227 Impact on the performance of small-scale mashing 14. Aastrup, S. (2010) Beer from 100% barley, Scand. Brew. Rev. 67, Eglinton, J., Langridge, P., and Evans, D. (1998) Thermostability variation in alleles of barley beta-amylase, J. Cereal Sci. 28, Schnitzenbaumer, B. and Arendt, E. K. (2014) Effect of unmalted oats (Avena sativa L.) on the quality of high-gravity mashes and worts without or with exogenous enzyme addition, Eur. Food Res, Technol. 238, Grøndahl Kristiansen, A., Lau Johansen, K., Mou, C. and Johansson, C.-G. (2011) Input to TWG on BAT candidates for breweries. Available from: (last accessed March 2016) 18. Davey, M. J., Landman, K. A., McGuinness, M. J., and Jin, H. N. (2002) Mathematical modeling of rice cooking and dissolution in beer production, AIChE J. 48, European Brewery Convention (1998) Analytica EBC, Verlag Hans Carl Getranke Fachverlag, Nurnberg. 20. Evans, D. E. (2008) A more cost- and labor-efficient assay for the combined measurement of the diastatic power enzymes β-amylase, α-amylase, and limit dextrinase, J. Am. Soc. Brew. Chem. 66, Evans, D. E., Dambergs, R., Ratkowsky, D., Li, C., Harasymow, S., Roumeliotis, S., and Eglinton, J. K. (2010) Refining the prediction of potential malt fermentability by including an assessment of limit dextrinase thermostability and additional measures of malt modification, using two different methods for multivariate model development, J. Inst. Brew. 116, Evans, D. E., Goldsmith, M., Dambergs, R., and Nischwitz, R. (2011) A comprehensive revaluation of small-scale congress mash protocol parameters for determining extract and fermentability, J. Am. Soc. Brew. Chem. 69, Evans, D. E., Collins, H., Eglington, J., and Wilhelmson, A. (2005) Assessing the impact of the level of diastatic power enzymes and their thermostability on the hydrolysis of starch during wort production to predict malt fermentability, J. Am. Soc. Brew. Chem. 63, R Core Team (2015) R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria. Available from: Wang, J., Zhang, G., Chen, J., and Wu, F. (2004) The changes of β-glucan content and β-glucanase activity in barley before and after malting and their relationships to malt qualities, Food Chem. 86, Evans, D. E., Goldsmith, M., Redd, K. S., Nischwitz, R., and Lentini, A. (2012) Impact of mashing conditions on extract, its fermentability, and the levels of wort free amino nitrogen (FAN), β-glucan, and lipids, J. Am. Soc. Brew. Chem. 70, Evans, D. E. (2012) The impact of malt blending on lautering efficiency, extract yield, and wort fermentability, J. Am. Soc. Brew. Chem. 70, Fujii, M. and Kawamura, Y. (1985) Synergistic action of α-amylase and glucoamylase on hydrolysis of starch, Biotechnol. Bioeng. 27, Stewart, G. G., Hill, A., and Lekkas, C. (2013) Wort FAN Its characteristics and importance during fermentation, J. Am. Soc. Brew. Chem. 71, Jones, M., and Pierce, J. (1964) Absorption of amino acids from wort by yeasts, J. Inst. Brew. 70, Lei, H., Zhao, H., and Zhao, M. (2013) Proteases supplementation to high gravity worts enhances fermentation performance of brewer s yeast, Biochem. Eng. J. 77, 1 6. Supporting information Additional supporting information may be found in the online version of this article at the publisher s web site. J. Inst. Brew. 2016; 122: Copyright 2016 The wileyonlinelibrary.com/journal/jib