Hindawi

Journal of Food Quality

Volume 2022, Article ID 5279177, 12 pages
https://doi.org/10.1155/2022/5279177

Research Article
Optimisation of Malting Parameters for Quinoa and Barley:
Application of Response Surface Methodology

Esther Tatenda Chawanda,1 Shepherd Manhokwe ,1 Talknice Z. Jombo,1

Desmond T. Mugadza,1 Michael Njini,1 and Pepukai Manjeru2
1
Midlands State University, Department of Food Science and Nutrition, P. Bag 9055, Gweru, Zimbabwe
2
Midlands State University, Department of Agronomy and Horticulture, P. Bag 9055, Gweru, Zimbabwe

Correspondence should be addressed to Shepherd Manhokwe; manhokwes@msu.ac.zw

Received 25 June 2021; Accepted 30 May 2022; Published 25 June 2022

Academic Editor: Daniel Cozzolino

Copyright © 2022 Esther Tatenda Chawanda et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

Quinoa (Chenopodium quinoa Willd) is a nutritious pseudocereal that is more stress-tolerant compared with traditional cereals. It
is an excellent example of a climate-smart crop that is more resilient to climate change compared with barley. The purpose of the
study was to investigate the optimum malting conditions required to produce quinoa malt using barley as a control. Response
surface methodology (RSM) was used to investigate the influence of the two malting parameters steeping time and germination
time on Brix (wort extract), diastatic power (DP), and free amino nitrogen (FAN) of the malt. The temperature was set at 15°C
during the steeping process. Steeping time ranging from 12 to 48 hours and germination time ranging from 24 to 96 hours were
designed using a central composite design (CCD). The kilning temperature for all malts was 65°C. For quinoa malt, there was a
notable weak positive correlation between germination time and Brix (r � +0.119). However, there was a strong positive cor-
relation between steeping time and diastatic power (r � +0.893). A similar trend was noted for barley with a weak positive
correlation between germination time and Brix (r � +0.142). A strong positive correlation was also recorded between steeping time
and diastatic power (r � +0.897) during the malting of barley. There was a relatively stronger correlation between steeping time
and FAN (r � +0.895) than germination time and FAN (r � +0.275) in quinoa malt. The optimum values for the malting of barley
were 47.68 hrs steeping time and 82.55 hrs germination time with a desirability value of 1.00. The responses for the optimised
barley malt were 8.25°Bx, 162.28 mg/L, and 271.69°L for Brix, FAN, and diastatic power, respectively. To produce quinoa malt with
Brix, FAN, and diastatic power of 8.37°Bx, 165.60 mg/L, and 275.86°L, respectively, malting conditions of 47.69 hrs steeping time
and 95.81 hrs germination time are required. It was noted that quinoa is a very good candidate for producing high-quality malt for
the brewing process.

1. Introduction quality of the manufactured products [3]. Whilst barley is an

important source of enzymes and fermentable sugars used
Barley (Hordeum vulgare L) is the primary cereal used in the by yeasts in brewing, alternative cereals prove to be more
production of malt in the world. It is a monocotyledon that economic, with an added advantage of sustainability [1].
belongs to the grass family Poaceae [1]. The major use of Malting is the controlled germination of cereals, causing
barley malt is in the brewing and distilling industries, with a biochemical changes within the grain, followed by kilning.
lesser amount utilized in food products [2]. The quality of Steeping increases the moisture content from 12 to at least
the beer and the cost-effectiveness of the brewing process are 40%, whilst kilning ensures product stability [4]. Mashing
influenced by the chemical composition, brewing tech- manipulates temperature profiles, interspersed with rest
niques, and technological indices of the barley malt. The periods to provide optimum conditions for enzyme catalysis.
barley malt quality then becomes critical in determining the The process of mashing breaks down complex nutrient
2 Journal of Food Quality

components (e.g., starch) from the grain by enzymatic Experimental data on the application of RSM in the malting
hydrolysis, resulting in simple fermentable molecules (e.g., of quinoa are scarce. This study seeks to explore the optimal
sugars and amino acids) [5]. malting conditions for quinoa malting using RSM, as an
Malting quality is an important parameter for barley alternative to barley in beer brewing.
grain used in the brewing process [6]. Important variables in
determining the malt quality include extract, viscosity, di- 2. Methodology
astatic power, Kolbach index, free amino nitrogen, wort
β-glucan content, and soluble protein [6, 7]. The ability to A study to investigate the relative contributions of two
hydrolyse starch to simple sugar during the germination of predictor factors (steeping time and germination time) to
barley is the diastatic power. The enzymes that contribute to the quality of barley and quinoa malt was conducted using
this malt diastatic system include α-amylase, β-amylase, RSM. Design Expert 7 (Stat-Ease Corporation, Minneapolis,
limit dextrinase, and α-glucosidase. The diastatic power is USA) was used to construct a central composite design
directly proportional to the yield and quality of beer brewing (CCD). Thirteen (13) runs of barley malting and 13 runs of
[7]. The solubility and filtration speed of the malt wort are quinoa malting were conducted as outlined in the CCD for
represented by viscosity, with a low viscosity indicating the variables shown in Table 1. The range of values for the
high-quality malt [6]. FAN measures the nitrogenous independent variables was defined by the operational values
compounds in the wort that can be utilised by yeast during used in the commercial production of barley malt. The
fermentation [8]. The level of FAN is determined primarily response variables for the malting process included extract
by the enzyme carboxypeptidase, a very heat-resistant en- (Brix value), FAN, and diastatic power.
zyme that is present in abundant quantities in most malts.
Quinoa (Chenopodium quinoa Willd) is a pseudocereal
2.1. Raw Material Collection. The barley was collected from a
that has been cultivated for thousands of years and has been
commercial maltster in Zimbabwe. The cultivated quinoa
rediscovered as a super grain for its health-promoting
was collected from Midlands State University Farm in
benefits [9]. It is a climate-smart crop, adapted to low
Kwekwe. The sampling, cleaning, transportation, and
rainfall, high temperatures, and different soil conditions.
analysis were done under controlled conditions to minimise
Consumers are becoming more health-conscious with a
contamination.
preference for functional foods. There is a public interest in
the replacement of common cereal grains (maize, wheat)
with more nutritious grains such as quinoa [10]. Quinoa 2.2. Micromalting. The steeping temperature for both qui-
seeds reveal a total absence of gluten and high levels of fatty noa and barley was 15°C. Both barley and quinoa were
acids, vitamins, minerals, antioxidants, dietary fibres, and steeped under the conditions prescribed in the CCD with
proteins [11]. Gluten found in wheat, barley, and rye causes variations in wet and dry conditions as shown in Table 2. The
celiac disease (CD). The symptoms of celiac disease include samples were allowed to germinate at 15°C [33]. After
damage to the intestinal epithelial cells (mucosa) in ge- germination, the samples were kilned at 65oC for 24 hrs and
netically susceptible individuals [12–14]. Brewing, using the malt was analysed following the standard methods de-
gluten-free ingredients such as quinoa, could produce a scribed in European Brewery Convention (EBC) Analytica
gluten-free beer suitable for gluten-sensitive individuals. [34]. All the samples were subjected to the same kilning
Response surface methodology (RSM) is a collection of conditions reducing the moisture content to 4.5–7%
mathematical and statistical techniques that is useful for the moisture within 24 hrs.
approximation and optimisation of stochastic models [15].
Experiments for fitting a predictive model involving several
continuous variables are known as response surface ex- 2.3. Chemicals and Reagents
periments [16]. The main applications of an experimental 2.3.1. Diastatic Power. The following chemicals and reagents
design are screening and optimisation [17–22]. The design of were used: starch, G.R. Merck 1252; sulphuric acid (H2 SO4),
experiments is intended to reduce the number of experi- G.R. Merck Titrisol 9981 (IM); acetic acid (HCOOH), G.R.
ments with a wide range of combinations of independent 96%, Merck 90062; sodium acetate trihydrate
variables [23]. Recently, RSM has been employed for opti- (NaC2H3O2–3H2O), G.R. Merck 6267; sodium hydroxide
misation of processes in food science and technology, (NaOH), G.R. Merck Titrisol 9956 (M); iodine, G.R. Merck
material engineering, chemistry, and chemical engineering Titrisol 9910 (0.05 M); sodium thiosulphate (Na2S2O3
[24]. RSM has been used in the optimisation of different food 5H2O), G.R. Merck Titrisol s9950 (0.1 M); thymolphthalein,
processes such as extraction, drying, blanching, enzymatic G.R. Merck 8175; ethyl alcohol (C2H4 OH), G.R. Merck 983;
hydrolysis and clarification, production of microbial me- and iodine (I2), G.R. Merck 4761.
tabolites, and formulation [25–27]. RSM was used to eval-
uate the capability of Phormidium valderianum on
biodegradation and decolorisation of distillery spent wash 2.3.2. Free Amino Nitrogen. The following chemicals and
[28]. Response surface methodology has been applied in reagents were used: glycine, Merck GR 4201; di-sodium
malting studies as reported by several researchers [29–31]. hydrogen orthophosphate (Na2HPO4.12H2O), Merck GR
RSM has also been used to study the influence of three 6579; potassium dihydrogen orthophosphate (KH2PO4),
malting parameters on the quality of proso millet malt [32]. Merck GR 4873; ninhydrin (indane-trione hydrate,
Journal of Food Quality 3

Table 1: Design summary of independent variables (factors) and their actual values.
Factor Name Units Type Coded low Coded high Mean Std. Dev.
A Steeping time hrs Numeric 12.00 48.00 30.00 14.70
B Germination time hrs Numeric 24.00 96.00 60.00 29.39
Key: Std. Dev. � standard deviation, hrs � hours.

Table 2: Steeping time with variations in wet and dry conditions.

Steeping time (hrs) Underwater (hrs) Dry (hrs) Underwater (hrs) Dry (hrs) Underwater (hrs) Dry (hrs)
4.54 2.54 2.00
12.00 8.00 6.00
30.00 6.00 16.00 6.00 2.00
48.00 4.00 18.00 4.00 12.00 4.00 6.00
55.46 10.46 18.00 12.00 10.00 3.00 2.00

C9H6O4), Merck AP 006762.100; D(-)fructose (laevulose), 2 hrs. The malt was ground using a Bühler Miag disc mill.
Merck 5323; potassium iodate, Merck GR 5051; ethanol, 96% The ground samples were placed in sampling bottles. The
Merck 971; pentachlorophenol, Merck 277392R; Kieselguhr mass of the weighing dish with lid was determined, and
(standard brewery grade); nitric acid, RG concentrated; approximately 5 g of the samples were weighed. The mass of
sodium hydroxide 2.0 M (AR) reagent; and hydrochloric the dish, lid, and sample was recorded. The samples were
acid 1.0 M (AR) reagent. dried at 105°C for 3 hours until a constant mass was
recorded. The sample and dish were cooled in desiccators.
3. Malt Characteristics The moisture content was determined as a percentage using
the following equation:
3.1. Moisture Content. Moisture content was determined
after driving out moisture by heat from a ground sample for

∗
W2 − W3 􏼁
%Moisture content � Loss for mass on drying∗ 100% � 􏼢 􏼣 100%, (1)
W2 − W1 􏼁

where W1 � mass of the empty dish and lid; W2 � mass of the colour, and pH value of the wort produced from the mashing
sample, dish, and lid before drying at 105°C; and W3 � mass stage. The vials for the prepared samples were thoroughly
of the sample, dish, and lid after drying at 105°C. rinsed and then filled up to about 8–10 ml. A printed sheet
with the results done by the microprocessor was produced
for traceability. Free amino nitrogen was determined
3.2. Micromashing. Twenty-six samples (barley malt and according to the European Brewery Convention (EBC)
quinoa malt) were analysed using a programmable mashing method by colour absorbance at 570 nm and reported as mg/
bath. The mashing process was carried out in the mash bath L with glycine as standard. Diastatic power indicates the total
with stirrers set to run at 80–100 rpm, at 45°C for 30 minutes. enzymatic power of malt, both α amylase and β amylase. It
The temperature of the mash bath was raised approximately was determined by measuring the breakdown of starch, and
1°C per minute; hence, the mash temperature of 70°C was this breakdown was measured by reaction with potassium
achieved exactly 25 minutes after the commencement of the ferricyanide and expressed in degrees Lintner (°L). This
heating. Distilled water preheated to 70°C was added to the reagent reacts with the reducing sugars produced by the
malt in the mash bath. The mash was held at 70°C for exactly action of the enzyme. This was determined by measuring the
60 minutes, and at the end of the mashing period, the mash amount of a buffered starch solution, which was degraded by
was cooled to 20°C for 10–15 minutes. The mass of the a 5 cm3 aliquot of an extract of the malt.
contents was adjusted to 450 g ± 0.01 g, by the addition of
distilled water. The mash was thoroughly stirred and filtered
into 500 cm3 conical flasks, returning the first 100 cm3 to the 4. Results
funnel. The filtrates were used to determine extract (Brix),
diastatic power, and free amino nitrogen. Table 3 shows the overall results of the malt parameters that
were analysed for barley grains, which were used as controls.
The grained grains were steeped and germinated at different
3.3. Mash Analysis. The Alcolyzer Beer Analysing System intervals. A maximum Brix value of 8.10°Bx was recorded in
(DMA-4500M; Anton Paar, Austria) was used to measure this design. There are notably low values recorded in Run 4
the density, original extract, real extract, calories, turbidity, (4.54 hrs steeping time; 60 hrs germination time) for Brix,
4 Journal of Food Quality

Table 3: A central composite design (CCD) for barley malting using response surface methodology.
Std Run Block Steeping time (hrs) Germination time (hrs) Brix (°Bx) FAN (mg/L) Diastatic power (°L)
1 11 Block 1 12.00 24.00 6.64 123.20 152.00
2 13 Block 1 48.00 24.00 8.09 160.50 264.00
3 12 Block 1 12.00 96.00 6.80 127.90 156.90
4 8 Block 1 48.00 96.00 8.10 161.50 268.00
5 4 Block 1 4.54 60.00 4.58 98.10 132.30
6 2 Block 1 55.46 60.00 7.94 162.00 270.00
7 7 Block 1 30.00 9.09 7.17 118.60 160.40
8 3 Block 1 30.00 110.91 8.05 150.90 258.40
9 10 Block 1 30.00 60.00 8.02 145.10 220.40
10 9 Block 1 30.00 60.00 7.98 142.90 215.90
11 6 Block 1 30.00 60.00 8.01 144.00 219.20
12 5 Block 1 30.00 60.00 8.00 142.30 211.10
13 1 Block 1 30.00 60.00 8.04 143.90 209.70

FAN, and diastatic power with 4.58°Bx, 98.10 mg/L, and recorded between steeping time and diastatic power
132.30°L, respectively. A strong positive correlation exists (r � +0.897) during the malting of barley. There was a rel-
between the FAN and diastatic power. atively stronger correlation between steeping time and FAN
The results obtained from the analysis of variance (r � +0.895) than germination time and FAN (r � + 0.275) in
(ANOVA) based on experimental data are presented in quinoa malt.
Table 4. The model F value of 14.44 implies the model is The contour and surface plots of the response functions
significant. There is only a 0.14% chance that an F value this are useful in understanding both the individual and the
large could occur due to noise. Statistical noise is unex- combined effects of the factors. Figure 1(a) shows the re-
plained variability within a data sample that is rendered lationship between steeping time and germination time with
meaningless by the existence of too many variables. P values the amount of sugar concentration (Brix) produced in the
less than 0.05 indicate model terms are significant. In this barley malt. The higher the steeping time and germination
case, A and A2 are significant model terms. Adequate time, the higher the concentration of sugar produced. This is
Precision measures the signal-to-noise ratio. A ratio greater shown as 8.10°Bx was obtained after steeping and germi-
than 4 is desirable. A ratio of 11.46 obtained in this analysis nating for longer hours, 48 hrs and 96 hrs, respectively. The
indicates an adequate signal. The model can be used to least concentration of Brix was obtained when barley was
navigate the design space. steeped for a shorter time (12 hrs) having 6.64°Bx. The re-
Table 5 shows the summary of the steeping and ger- lationship between steeping time and germination time with
mination time that were used during the malting process of the amount of sugar concentration (Brix) produced in the
quinoa grains. Run 8 (4.54 hrs steeping time; 60 hrs ger- quinoa malt is shown in Figure 1(b). The contours indicate
mination time) had the lowest values for all the responses higher response levels as the steeping time and germination
recorded in this study. The maximum values for Brix, FAN, time increases. A high concentration of the sugar in quinoa
and diastatic power were 8.11°Bx, 165.60 mg/L, and 268.70°L, malt (8.09°Bx) was recorded at a steeping time of 48 hrs and
respectively. germination time of 96 hrs. The trend noted in barley is also
The model for Brix from the two variables used in similar to that in the malting of quinoa with lower Brix
malting had a high F value (11.98) and low p value values recorded at lower malting variables.
(p � 0.0025) (Table 6), which indicated that the model was Figure 2(a) shows the relationship between the steeping
highly significant. In this case, A and A2 are significant time and germination time with the free amino nitrogen
model terms. The good correlation between the experi- obtained in barley malt. The FAN increases as the steeping
mental and predicted values was confirmed by the proximity and germination processes increase with a high FAN value
of the determination (R2) and adjusted determination (R2adj) (161.50 mg/L) recorded after a steeping time of 48 hrs and
values 0.8954 and 0.8207, respectively. An R2 value close to germination time of 96 hrs for barley malt. The relationship
unity and R2adj close to R2 ensure satisfactory fitting of the between steeping time and germination time with the FAN
model to the real system [35]. An adequate precision ratio of obtained in quinoa malt is shown in Figure 2(b). Coinci-
10.52 indicates an adequate signal. The established model is dentally, the highest FAN value recorded (163.00 mg/L) in
satisfactory and confirms response surface methodology is a the malting of quinoa was obtained from the same steeping
promising tool in the selection of malting variables. and germination time as in barley malting (48 hrs steeping
Table 7 shows that, for quinoa malt, there was a notable and 96 hrs germination time).
weak positive correlation between germination time and Figure 3(a) shows the diastatic power of barley malt after
Brix (r � +0.119). However, there was a strong positive steeping and germinating the grains at different intervals.
correlation between steeping time and diastatic power For barley malt, as steeping time and germination increase
(r � +0.893). A similar trend was noted for barley with a so does the diastatic power in the malt with 268.00°L being
weak positive correlation between germination time and recorded at 48 hrs/96 hrs compared with 152.00°L at 12 hrs/
Brix (r � +0.142). A strong positive correlation was also 24 hrs. Figure 3(b) shows that low diastatic power (152.40°L)
Journal of Food Quality 5

Table 4: ANOVA results for quadratic modelling of Brix content during barley malting.
Source SS df MS F value p value
Model 11.2800 5 2.2600 14.4400 0.0014 Significant
A-steeping time 7.0300 1 7.0300 45.0200 0.0003
B-germination time 0.2501 1 0.2501 1.6000 0.2463
AB 0.0056 1 0.0056 0.0360 0.8549
A2 3.9900 1 3.9900 25.5100 0.0015
B2 0.0466 1 0.0466 0.2985 0.6018
Residual 1.0900 7 0.1562
Lack of fit 1.0900 3 0.3639 727.8100 <0.0001
Pure error 0.0020 4 0.0005
Cor total 12.37 12
SS: sum of squares; MS: mean square.

Table 5: A central composite design (CCD) for quinoa malting using response surface methodology.
Std Run Block Steeping time (hrs) Germination time (hrs) Brix (°Bx) FAN (mg/L) Diastatic power (°L)
1 12 Block 1 12.00 24.00 6.23 124.40 152.40
2 9 Block 1 48.00 24.00 8.06 162.40 261.50
3 10 Block 1 12.00 96.00 6.50 129.60 158.00
4 13 Block 1 48.00 96.00 8.09 163.00 265.30
5 8 Block 1 4.54 60.00 3.17 99.50 134.90
6 7 Block 1 55.46 60.00 8.11 165.60 268.70
7 11 Block 1 30.00 9.09 7.10 122.10 162.10
8 2 Block 1 30.00 110.91 8.05 153.80 260.00
9 1 Block 1 30.00 60.00 8.03 144.20 211.90
10 5 Block 1 30.00 60.00 8.00 143.40 218.90
11 6 Block 1 30.00 60.00 8.03 145.30 216.60
12 4 Block 1 30.00 60.00 7.95 144.50 215.30
13 3 Block 1 30.00 60.00 7.97 143.20 213.60

Table 6: ANOVA results for quadratic modelling of Brix content during quinoa malting.
Source Sum of squares Df Mean square F value p value
Model 21.2400 5 4.2500 11.9800 0.0025 Significant
A-steeping time 13.5400 1 13.5400 38.1900 0.0005
B-germination time 0.3376 1 0.3376 0.9525 0.3616
AB 0.0144 1 0.0144 0.0406 0.8460
A2 7.3100 1 7.3100 20.6100 0.0027
B2 0.0229 1 0.0229 0.0646 0.8067
Residual 2.4800 7 0.3545
Lack of fit 2.4800 3 0.8254 644.8300 <0.0001
Pure error 0.0051 4 0.0013
Cor total 23.72 12

Table 7: Correlation coefficient (r) between independent and response variables during malting.
Barley malt Quinoa malt
Response variable Steeping time Steeping time
Germination time Germination time
Brix (oBx) +0.142 +0.754 +0.119 +0.755
FAN (mg/L) +0.280 +0.880 +0.275 +0.895
Diastatic power (°L) +0.317 +0.897 +0.326 +0.893

was obtained at a shorter steeping time (12 hrs) and ger- predicted values as most of the points are located in the
mination time. The diastatic power increased as the time of range of 7.06–8.36°Bx.
steeping increased (r � +0.893) with 265.30°L being recorded The plot of the residuals versus normal probability is
at 48 hrs as shown in Figure 4. shown in Figure 6. The residuals must be normally dis-
Figure 5 compares actual values of Brix in 5(a) barley tributed for the results of an ANOVA to be valid, and thus
malt and 5(b) quinoa malt with the predicted values ob- the normal probability plot should resemble a straight line to
tained from the statistical process model. The figure dem- indicate the underlying error distribution is normal [36]. It
onstrates good agreement between the experimental and can be observed from Figure 6 that the normality
6 Journal of Food Quality

Design-ExpertⓇ Software
Brix
Brix 96.00
Design Points
8.1

4.58 78.00

X1 = A: steeping time

B: germination time
X2 = B: germination time
7.19604
60.00 6.80377 7.58831 5
7.98059

42.00 6.41149

24.00
12.00 21.00 30.00 39.00 48.00
A: steeping time
(a)
Design-ExpertⓇ Software Brix
96.00
Brix
Design Points
8.11

3.17 78.00

X1 = A: steeping time
B: germination time

X2 = B: germination time
6.40488 6.93355 7.46223 7.9909
60.00 5

5.87621

42.00

24.00
12.00 21.00 30.00 39.00 48.00
A: steeping time
(b)

Figure 1: Contour plot for steeping time, germination time, and Brix value of (a) barley malt and (b) quinoa malt.

assumption is relatively satisfied as the points in the plot reaches its optimum, which could be either a maximum or a
form a fairly straight line. So it can be concluded that the minimum of the developed function [36]. The numerical
empirical model is adequate to describe the malting process optimisation was performed as facilitated in Design-Expert 7
as described in this study. software. RSM was used to determine the optimum process
The residuals versus the experimental run order checks for parameters that yield high malting characteristics. For nu-
lurking variables that may have influenced the response during merical optimisation, the goals (none, maximum, minimum,
the experiment. It is important to note that the runs are within target, or range) should be set for both the independent and
the control limit as shown in both Figures 7(a) and 7(b). response variables where all goals are combined into one
Process optimisation involves the determination of the desirable function [37]. Maximum values were set for the
values of the design parameters at which the response responses (i.e., Brix, FAN, and diastatic power). For barley,
Journal of Food Quality 7

Design-ExpertⓇ Software
FAN 164
162
147.5
98.1
X1 = A: steeping time
131
X2 = B: germination time

FAN
114.5

98

96.00
78.00
B: 48.00
ge 60.00
rm 39.00
in 42.00 30.00
ati
on 21.00 time
tim 24.00 eping
e 12.00 A: ste

(a)
Design-ExpertⓇ Software
FAN
165.6

99.5
X1 = A: steeping time
166
X2 = B: germination time
149.25

132.5
FAN

115.75

99
96.00

12.00 78.00
21.00
e
tim

60.00
n

30.00
io

A:
at

ste
in

epi 42.00
rm

ng ti 39.00
ge

me
B:

48.00 24.00
(b)

Figure 2: Response surface plots of steeping time and germination time on free amino nitrogen of (a) barley malt and (b) quinoa malt.

the numerical optimisation established that the desirability range of 99.5–165.6 mg/L, and diastatic power in the range of
function was maximised at a Brix in the range of 4.58–8.1°Bx, 134.9–268.7°L. However, the diastatic power is higher in
FAN in the range of 98.1–162 mg/L, and diastatic power in quinoa than in barley at similar optimum steeping
the range of 132.3–270°L. The optimum values (Table 8) for conditions.
the malting of barley were 47.68 hrs steeping time and
82.55 hrs germination time with a desirability value of 1.00. 5. Discussion
To produce quinoa malt with Brix, FAN, and diastatic power
of 8.37°Bx, 165.60 mg/L, and 275.86°L, respectively, malting 5.1. Effect of Steeping Time and Germination Time on Brix
conditions of 47.69 hrs steeping time and 95.81 hrs germi- Content in Barley and Quinoa Malt. Good modification
nation time are required. For quinoa, the numerical opti- requires the grains to remain in the compartment for 4–5
misation established that the desirability function was days under controlled temperature, and this is the degree to
maximised at a Brix in the range 3.17–8.11°Bx, FAN in the which enzymes break down the endosperm. This can be
8 Journal of Food Quality

Design-ExpertⓇ Software
Diastatic power
270

132.3 280
X1 = A: steeping time
242.5
X2 = B: germination time
205

Diastatic power
167.5

130

48.00
39.00
A:

30.00
ste

24.00
21.00
ep

42.00
60.00
ing

12.00 78.00 time

ination
tim

96.00
B: germ
e

(a)
Design-ExpertⓇ Software
Diastatic power
268.7

134.9
X1 = A: steeping time 280
X2 = B: germination time
242.5
Diastatic power

205

167.5

130

24.00
42.00 e
48.00 tim
39.00 60.00 n
io
30.00 at
A: st 78.00 in
eepin 21.00 rm
g tim 12.00 96.00 ge
e B:
(b)

Figure 3: Contour plots for steeping time and germination time on diastatic power of (a) barley malt and (b) quinoa malt.

shown by a lower Brix value, where quinoa was 3.17°Bx and The higher Brix value was obtained when the grains were
barley was 4.58°Bx as they were subjected to short steeping steeped for 48 hrs and 96 hrs of germination with 8.09°Bx
time and germination time (4.54 hrs steeping and 60 hrs quinoa and 8.1°Bx barley. Quinoa had slightly lower Brix
germination). The texture of the endosperm affects water values than barley. This may be attributed to the high protein
uptake and ultimately enzyme synthesis within the endo- content in quinoa grains (13.8% to 16.5%, with an average of
sperm [4]. This means there was no modification as less 15%) than in barley grains (between 9% and 15% dry matter)
water was absorbed to activate the enzymes within the [10, 39, 40]. The surfaces and internal structures of malt may
grains. As the steeping time and germination time increase influence the rate of enzyme hydrolysis with undegraded
so does the Brix value within the grains [38]. During the proteins limiting the movement of enzymes in the starchy
steeping process, the absorbed water actives the naturally endosperm. The uneven pattern of modification results in
existing enzymes within the grains to break down starch poor-quality wort with low extract, higher viscosity, and
granules into simpler fermentable sugar in the endosperm. slow wort separation (filtration) [41, 42].
Journal of Food Quality 9

Design-ExpertⓇ Software
270
Correlation: 0.893
Color points by
Run
235
13

Diastatic power
1
200

165

130

4.54 17.27 30.00 42.73 55.46

A:steeping time

Figure 4: Correlation between steeping time and diastatic power during quinoa malting.

Design-ExpertⓇ Software Design-ExpertⓇ Software

Brix Brix
Predicted vs. Actual Predicted vs. Actual
Color points by value of Color points by value of
8.30 8.40
Brix: Brix: 2
8.1 8.11
4.58 7.35 3.17 7.08
Predicted

Predicted
6.40 5.75

5.45 4.43

4.50 3.10

4.58 5.50 6.41 7.33 8.25 3.17 4.47 5.76 7.06 8.36
Actual Actual

(a) (b)

Figure 5: Actual vs predicted values of Brix in the malting process.

Design-ExpertⓇ Software Normal Plot of Residuals Design-ExpertⓇ Software Normal Plot of Residuals
Diastatic power DP
Color points by value of Color points by value of 99
99
Diastatic power: DP:
95 95
Normal % Probability
Normal % Probability

270 90 268.7 90
132.3 80 80
70 134.9 70
50 50
30 30
20 20
10 10
5 5
1 1

-1.83 -0.96 -0.09 0.78 1.64 -1.74 -0.86 0.03 0.91 1.79
Internally Studentized Residuals Internally Studentized Residuals

(a) (b)

Figure 6: Normal probability plot of residuals for (a) barley diastatic power and (b) quinoa diastatic power.

5.2. Effect of Steeping Time and Germination Time on Free of barley proteins are globulin, glutelin, prolamin, and al-
Amino Nitrogen in Barley and Quinoa. During malting and bumin [8]. As steeping time and germination times prolong,
mashing, the proteinases are critical because several aspects more water is absorbed and used to activate enzymes within
of the brewing process are affected by the soluble proteins, the grains. This increases protein modification during the
peptides, and/or amino acids that they release. As the dia- germination process. Proteolytic modification promotes the
static enzymes are proteins, their levels are directly related to development of enzymes required to modify starch in the
protein concentration in the grain. The four distinct classes endosperm [8]. The FAN is an indicator suitable for the
10 Journal of Food Quality

Design-ExpertⓇ Software Residuals vs. Run Design-ExpertⓇ Software Residuals vs. Run
DP Diastatic power
3.00 3.00
Color points by value of Color points by value of

Internally Studentized Residuals

Internally Studentized Residuals

DP: Diastatic power:
268.7 1.50 270 1.50
134.9 132.3

0.00 0.00

-1.50 -1.50

-3.00 -3.00

1 3 5 7 9 11 13 1 3 5 7 9 11 13
Run Number Run Number

(a) (b)

Figure 7: Residuals vs run plot for diastatic power in the malting process for (a) quinoa and (b) barley.

Table 8: Optimisation of the malting conditions of barley and quinoa grains.

Samples Steeping time (hrs) Germination time (hrs) Brix (°Bx) FAN (mg/L) Diastatic power (°L) Desirability
Barley 47.68 82.55 8.25 162.28 271.69 1.00 Selected
Quinoa 47.69 95.81 8.37 165.60 275.86 1.00 Selected

Table 9: Summary of FAN values from malting studies using RSM.

Grain Model, design FAN References
Buckwheat malt RSM, central composite design 144.26 mg/L [50]
Sorghum malt RSM, face-centred cube design 117.80 mg/100 g [51]
Rice wort RSM, central composite design 357.00 mg/L [52]
Quinoa malt RSM, central composite design 165.60 mg/L This study
Proso millet wort RSM, face-centred design 365.00 mg/100 ml [32]

prediction of the viability of yeast and the efficiency of being the major contributor to diastatic power generation
fermentation [42]. A low FAN content delays the ageing [53]. High levels of diastatic power recorded in this study
process, whilst a very high FAN content affects the sensory indicate there is more concentration of diastase enzymes that
properties of the beer and its microbiological quality. are capable of converting starch into simpler sugars that will
Quinoa had a slightly higher amount of FAN be consumed by yeast cells during the fermentation process
(163.00 mg/L) than barley (161.50 mg/L) after steeping for and alcohol will be produced [53]. Gluten-free grains often
48 hrs and germinated for 96 hrs. In wort, FAN content require prolonged germination times compared with barley
should range from 120 to 200 mg/L to form approximately [13]. From Figure 4, the strong positive correlation indicates
21–22% of the soluble nitrogen [43–45]. Therefore, the steeping time should be longer as more enzymes necessary
values recorded in this study were within the expected for the breakdown of starch will be activated, thereby
range. The higher FAN content of quinoa (283 mg/L) was converting more starch into simpler sugars.
reported in a study [12], with FAN in the range of
170–200 mg/L reported elsewhere [46–48]. In a related 6. Conclusion
study, the FAN content of triticale malt (a cross-breed of
wheat and rye) was lower (170 mg/l) in comparison with The results indicated that optimum values can be obtained in
barley malt (210 mg/L) [49]. The filterability of the beer the malting of quinoa that can satisfy the malt quality re-
depends on the concentration of high-molecular-weight quired for brewing. To produce quinoa malt with Brix, FAN,
protein and directly influences the foam stability, taste of and diastatic power of 8.37°Bx, 165.60 mg/L, and 275.86°L,
beer, and storage quality. Comparative values of FAN in respectively, malting conditions of 47.69 hrs steeping time
malt and wort are shown in Table 9. and 95.81 hrs germination time are required. Moreover, the
diastatic power is higher in quinoa than barley at similar
optimum steeping conditions. Therefore, quinoa has a po-
5.3. Effect of Steeping Time, Germination Time, and Diastatic tential application in the malting process for the production
Power on Barley and Quinoa. Gluten-free malt has a com- of gluten-free beer.
paratively similar diastatic power (265.30°L compared with
268.00°L for barley) as indicated in Figure 3. For malting, the Data Availability
recommended diastatic power should be above 100°L. The
activity of one enzyme group, the β-amylases, is regarded as Data are available on request.
Journal of Food Quality 11

Conflicts of Interest gluten-free products,” Food Research International, vol. 110,

pp. 62–72, 2018.
The authors do not have any conflicts of interest to declare. [15] D. C. Montgomery, “Introduction to statistical quality con-
trol,” in Design and Analysis of Experiments, pp. 457–461,
Acknowledgments John Wiley & Sons, Chichester, UK, 2001.
[16] A. Dean, D. Voss, and D. Draguljić, “Response surface
The authors thank the commercial maltster in Zimbabwe for methodology,” in Design and Analysis of Experiments-
providing technical assistance in the analysis of barley and Springer, Cham, Switzerland, 2017.
quinoa malts. Technical assistance provided by technicians [17] G. Hanrahan and K. Lu, “Application of factorial and re-
in the Department of Food Science and Nutrition at Mid- sponse surface methodology in modern experimental design
and optimization,” Critical Reviews in Analytical Chemistry,
lands State University is worth mentioning.
vol. 36, no. 3-4, pp. 141–151, 2006.
[18] M. H. Mahmoudian, M. Fazlzadeh, M. H. Niari et al., “A novel
References silica supported chitosan/glutaraldehyde as an efficient sor-
bent in solid phase extraction coupling with HPLC for the
[1] Y. J. Kok, L. Ye, J. Muller, D. S.-W. Ow, and X. Bi, “Brewing
determination of Penicillin G from water and wastewater
with malted barley or raw barley: what makes the difference in
samples,” Arabian Journal of Chemistry, vol. 13, no. 9,
the processes?” Applied Microbiology and Biotechnology,
pp. 7147–7159, 2020.
vol. 103, no. 3, pp. 1059–1067, 2019.
[2] B. Zhou, Z. Jin, P. Schwarz, and Y. Li, “Impact of genotype, [19] Y. Rashtbari, S. Hazrati, A. Azari, S. Afshin, M. Fazlzadeh, and
environment, and malting conditions on the antioxidant M. Vosoughi, “A novel, eco-friendly and green synthesis of
activity and phenolic content in US malting barley,” Fer- PPAC-ZnO and PPAC-nZVI nanocomposite using pome-
mentation, vol. 6, no. 2, 2020. granate peel: cephalexin adsorption experiments, mecha-
[3] A. Kochevenko, Y. Jiang, C. Seiler et al., “Identification of QTL nisms, isotherms and kinetics,” Advanced Powder Technology,
hot spots for malting quality in two elite breeding lines with vol. 31, no. 4, pp. 1612–1623, 2020.
distinct tolerance to abiotic stress,” BMC Plant Biology, vol. 18, [20] A. Azari, R. Nabizadeh, A. H. Mahvi, and S. Nasseri, “Inte-
no. 1, p. 106, 2018. grated Fuzzy AHP-TOPSIS for selecting the best color re-
[4] M. Gupta, N. Abu-Ghannam, and E. Gallaghar, “Barley for moval process using carbon-based adsorbent materials: multi-
brewing: characteristic changes during malting, brewing and criteria decision making vs. systematic review approaches and
applications of its by-products,” Comprehensive Reviews in modeling of textile wastewater treatment in real conditions,”
Food Science and Food Safety, vol. 9, no. 3, pp. 318–328, 2010. International Journal of Environmental Analytical Chemistry,
[5] G. G. Stewart, “Biochemistry of brewing,” in Biochemistry of vol. 2020, pp. 1–16, 2020.
Foods, pp. 291–318, Academic Press, San Diego, CA, USA, [21] A. Azari, M. Yeganeh, M. Gholami, and M. Salari, “The su-
2013. perior adsorption capacity of 2,4-dinitrophenol under ul-
[6] Y. Fang, X. Zhang, and D. Xue, “Genetic analysis and mo- trasound-assisted magnetic adsorption system: modeling and
lecular breeding applications of malting quality QTLs in process optimization by central composite design,” Journal of
barley,” Frontiers in Genetics, vol. 10, p. 352, 2019. Hazardous Materials, vol. 418, Article ID 126348, 2021.
[7] S. T. Cu, T. J. March, S. Stewart et al., “Genetic analysis of [22] A. Azari, R. Nabizadeh, A. H. Mahvi, and S. Nasseri,
grain and malt quality in an elite barley population,” Mo- “Magnetic multi-walled carbon nanotubes-loaded alginate for
lecular Breeding, vol. 36, no. 9, p. 129, 2016. treatment of industrial dye manufacturing effluent: adsorp-
[8] A. Hill and G. Stewart, “Free amino nitrogen in brewing,” tion modelling and process optimisation by central composite
Fermentation, vol. 5, no. 1, p. 22, 2019. face-central design,” International Journal of Environmental
[9] G. Li, F. Zhu, G. Mo, and Y. Hemar, “Supramolecular Analytical Chemistry, vol. 2021, pp. 1–21, 2021.
structure of high hydrostatic pressure treated quinoa and [23] S. A. Younis, W. I. El-Azab, N. S. El-Gendy et al., “Application
maize starches,” Food Hydrocolloids, vol. 92, pp. 276–284, of response surface methodology to enhance phenol removal
2019. from refinery wastewater by microwave process,” Interna-
[10] A. Vega-Gálvez, M. Miranda, J. Vergara, E. Uribe, L. Puente, tional Journal of Microwave Science and Technology, vol. 2014,
and E. A. Martı́nez, “Nutrition facts and functional potential Article ID 639457, 12 pages, 2014.
of quinoa (Chenopodium quinoa willd.), an ancient Andean [24] M. J. K. Bashir, S. S. A. Amr, S. Q. Aziz, N. C. Aun, and
grain: a review,” Journal of the Science of Food and Agriculture, S. Sethupath, “Wastewater treatment processes optimization
vol. 90, no. 15, pp. 2541–2547, 2010. using response surface methodology (RSM) compared with
[11] E. Pereira, C. Encina-Zelada, L. Barros, U. Gonzales-Barron, conventional methods: review and comparative study,”
V. Cadavez, and I. C. F. R. Ferreira, “Chemical and nutritional Middle-East Journal of Scientific Research, vol. 23, no. 2,
characterization of Chenopodium quinoa Willd (quinoa) pp. 244–252, 2015.
grains: a good alternative to nutritious food,” Food Chemistry, [25] Y. Mahmoud and J. Seid, “Applications of response surface
vol. 280, pp. 110–114, 2019. methodology in the food industry processes,” Food and
[12] B. Meo, G. Freeman, O. Marconi, C. Booer, G. Perretti, and Bioprocess Technology, vol. 10, pp. 413–433, 2017.
P. Fantozzi, “Behaviour of malted cereals and pseudo-cereals [26] N. Jafari and J. Seid, “Food process modeling and optimi-
for gluten-free beer production,” Journal of the Institute of zation by response surface methodology,” Mathematical and
Brewing, vol. 117, no. 4, pp. 541–546, 2011. Statistical Applications in Food Engineering, CRC Press, Boca
[13] A.-S. Hager, J. P. Taylor, D. M. Waters, and E. K. Arendt, Raton, FL, USA, pp. 181–203, 2020.
“Gluten free beer - a review,” Trends in Food Science & [27] S. Worku Kidane, “Application of response surface meth-
Technology, vol. 36, no. 1, pp. 44–54, 2014. odology in food process modeling and optimization,” Re-
[14] K. A. Scherf, H. Wieser, and P. Koehler, “Novel approaches sponse Surface Methodology in Engineering Science,
for enzymatic gluten degradation to create high-quality IntechOpen, London, UK, 2021.
12 Journal of Food Quality

[28] R. Ravikumar, K. Renuka, V. Sindhu, and K. B. Malarmathi, [43] P. Kofroň, R. Skoblı́k, J. Enge, and M. Sekora, “Testing of
“Response surface methodology and artificial neural network malting barley–variety bojos,” Kvasny Prumysl, vol. 52, no. 6,
for modeling and optimization of distillery spent wash pp. 179–184, 2006.
treatment using Phormidium valderianum BDU 140441,” [44] V. Psota, O. Dvořáčková, M. Nečas, and M. Musilová, “Barley
Polish Journal of Environmental Studies, vol. 22, no. 4, varieties registered in the Czech Republic after harvest 2017,”
pp. 1143–1152, 2013. Kvasnyprumysl, vol. 64, no. 3, pp. 102–110, 2018.
[29] S. Balasubramanian, J. Kaur, and D. Singh, “Optimization of [45] V. Psota, “Free amino nitrogen in sweet wort made from
weaning mix based on malted and extruded pearl millet and barley varieties tested in the Czech Republic,” Kvasnyprumysl,
barley,” Journal of Food Science & Technology, vol. 51, no. 4, vol. 65, no. 4, pp. 142–148, 2019.
pp. 682–690, 2014. [46] M. Zarnkow, T. Geyer, B. Lindemann et al., “The use of re-
[30] N. Gangopadhyay, M. B. Hossain, D. K. Rai, and sponse surface methodology to optimise malting conditions
N. P. Brunton, “Optimisation of yield and molecular weight of of quinoa (Chenopodium quinoa L.) as raw material for
β-glucan from barley flour using response surface method- gluten-free food and beverages,” Brauwiss, vol. 60, pp. 118–
ology,” Journal of Cereal Science, vol. 62, pp. 38–44, 2015. 126, 2007.
[31] Z. Hailu and B. Mekonnen, “Optimization of local wort and [47] M. Zarnkow and S. Kreisz, “Beer and Innovative Drinks Based
fermented beer from barley as substitute raw material for on Malted Cereals and Pseudo-cereals,” in Proceedings of the
ICC Jubilee Conference, Vienna, Austria, 2005.
Ethanol production using response surface methodology,”
[48] M. Zarnkow, M. Kessler, F. Burberg, S. Kreisz, and W. Back,
International Journal of Food and Nutrition Research, vol. 3,
“Gluten free beer from malted cereals and pseudocereals,” in
p. 33, 2019.
Proceedings of the 30th European Brewery Convention, Prague,
[32] M. Zarnkow, M. Keßler, F. Burberg, W. Back, E. K. Arendt,
Czech Republic, 2005.
and S. Kreisz, “The use of response surface methodology to
[49] M. Cioch-Skoneczny, M. Zdaniewicz, and A. Pater, “Impact of
optimise malting conditions of proso millet (Panicum mil-
triticale malt application on physiochemical composition and
iaceum L.) as a raw material for gluten-free foods,” Journal of profile of volatile compounds in beer,” European Food Re-
the Institute of Brewing, vol. 113, no. 3, pp. 280–292, 2007. search and Technology, vol. 245, pp. 1431–1437, 2019.
[33] D. E. Briggs, C. A. Boulton, P. A. Brookes, and R. Stevens, [50] B. P. N. Phiarais, B. D. Schehl, J. C. Oliveira, and E. K. Arendt,
Brewing Science and Practice, Woodhead Publishing, Boca “Use of response surface methodology to investigate the ef-
Raton, FL, USA, 2004. fectiveness of commercial enzymes on buckwheat malt for
[34] E. Brewery Convention, Analytica-EBC, Fachverlag Hans brewing purposes,” Journal of the Institute of Brewing,
Carl, Nürnberg, Germany, 2005. vol. 112, no. 4, pp. 324–332, 2006.
[35] D. Nahemiah, N. Iro, and H. B. Mamudu, “Application of [51] A. Hassani, M. Zarnkow, and T. Becker, “Influence of malting
response surface methodology (RSM) for the production and conditions on sorghum (Sorghum bicolor (L.) Moench) as a
optimization of extruded instant porridge from broken rice raw material for fermented beverages,” Food Science and
fractions blended with cowpea,” International Journal of Technology International, vol. 20, no. 6, pp. 453–463, 2013.
Nutrition and Food Sciences, vol. 5, no. 2, pp. 105–116, 2016. [52] A. Kongkaew, U. Usansa, and C. Wanapu, “Optimization of
[36] K. Anupam, J. Sikder, S. Pal, and G. Halder, “Optimizing the wort production from rice malt using enzymes and barley
cross-flow nanofiltration process for chromium (VI) removal malt,” African Journal of Biotechnology, vol. 11, no. 42,
from simulated wastewater through response surface meth- pp. 9941–9949, 2012.
odology,” Environmental Progress & Sustainable Energy, [53] T. S. Gibson, V. Solah, M. R. G. Holmes, and H. R. Taylor,
vol. 34, no. 5, pp. 1332–1340, 2015. “Diastatic power in malted barley: contributions of malt
[37] M. Jain, M. Verma, and P. Jain, “Process optimization for parameters to its development and the potential of barley
producing cowpea added instant kheer mix using response grainbeta-amylase to predict malt diastatic power,” Journal of
surface methodology,” Journal of Nutritional Health & Food the Institute of Brewing, vol. 101, no. 4, pp. 277–280, 1995.
Engineering, vol. 1, no. 5, 2014.
[38] T. M. Wolvever, “Relationship between proteinase activity
during malting and malt quality,” Journal of the Institute of
Brewing, vol. 108, pp. 371–376, 2009.
[39] E. Kordialik-Bogacka, P. Bogdan, K. Pielech-Przybylska, and
D. Michałowska, “Suitability of unmalted quinoa for beer
production,” Journal of the Science of Food and Agriculture,
vol. 98, no. 13, pp. 5027–5036, 2018.
[40] M. S. Izydorczyk and M. Edney, “Barley: Grain-Quality
Characteristics and Management of Quality Requirements,”
in Cereal Grains, pp. 195–234, Woodhead Publishing, Boca
Raton, FL, USA, 2017.
[41] R. M. Sá and G. H. Palmer, “Assessment of enzymatic en-
dosperm modification of malting barley using individual
grain analyses,” Journal of the Institute of Brewing, vol. 110,
no. 1, pp. 43–50, 2004.
[42] J.-L. Runavot, B. Bakan, N. Geneix et al., “Impact of low
hydration of barley grain on β-glucan degradation and lipid
transfer protein (LTP1) modifications during the malting
process,” Journal of Agricultural and Food Chemistry, vol. 59,
no. 15, pp. 8256–8264, 2011.