JFS E: Food Engineering and Physical Properties

Small and Large Deformation

Rheology for Hard Wheat Flour Dough
as Influenced by Mixing and Resting
Y.-R. KIM, P. CORNILLON, O.H. CAMPANELLA, R.L. STROSHINE, S. Lee, AND J.-Y. SHIM

ABSTRACT: The effects of mixing and resting on the physicochemical properties of doughs prepared with strong
and weak hard wheat flours were investigated, specifically concerning aspects related to their rheological behavior

E: Food Engineering &

Physical Properties
and molecular mobility. Small deformation dynamic tests showed that, during the initial resting period, the complex
modulus G* decreased and phase angle decreased for undermixed dough, whereas overmixed dough showed oppo-
site trends. G* values for optimally mixed dough did not vary during the resting period investigated. This was more
obvious for the strong dough. Large deformation tests more clearly showed differences among optimal, under-, and
overmixed dough, and also between doughs prepared with strong and weak flour. Optimally mixed dough exhibited
the highest peak stress and strain for both samples. In addition, the peak stress of dough prepared with the strong
flour was higher than that of dough prepared with weak flour. Inconsistent results between small and large defor-
mation tests implied that small and large deformation tests reflected different structural aspects of dough. NMR
measurements were performed to estimate the relaxation properties of the sample upon resting. Decreased water
mobility during resting, indicated by decreasing T 1 relaxation time, was possibly attributed to increasing molecular
interactions caused by continued hydration. Evidence of additional molecular interactions created by mixing was
also observed.
Keywords: hard wheat flour dough, mixing time, NMR, resting time, rheological properties

Introduction logical responses of dough to these actions govern its breadmaking

F rom the 1st step in the breadmaking process (blending of flour

and water with other ingredients) to the final step (baking),
the ingredients used undergo a number of physical and chemi-
cal changes. In fact, upon mixing, dough properties continuously
change simply due to the passage of time (Faubion and Faridi 1986).
The type and extent of physicochemical changes that wheat flour
dough experiences during each step profoundly influence finished
baked products (Ramkumar and others 1996). Each step in bread-
making has optimum processing parameters, that is, an optimum
water absorption, an optimum mixing time, and an optimum rest-
ing time, to name a few of the numerous processing parameters
used during breadmaking. Exceeding or undercutting these val-
ues can significantly affect the properties of the resulting dough,
and, most important, the quality of the final product (Faubion
and Faridi 1986). Therefore, optimum conditions must be prop-
erly determined, and understanding of how they can be achieved
is essential to control the breadmaking process. During breadmak-
ing, dough undergoes different degrees of stress and deformation,
which range from severe stretching and molecular reorganization
during mixing to mild expansion during fermentation. The rheo-
MS 20070328 Submitted 5/3/2007, Accepted 10/17/2007. Author Kim is with
Center for Agricultural Biomaterials and Dept. of Biosystems & Biomateri-
als Science and Engineering, Seoul Natl. Univ., Seoul, 151-742, Korea. Au-
thor Cornillon is with Danone Vitapole, 15 Ave. Galilee, 92350 Le Plessis-
Robinson, France. Authors Campanella and Stroshine are with Dept. of
Agricultural and Biological Engineering, Purdue Univ., West Lafayette, IN
47905, U.S.A. Author Lee is with Dept. of Food Science & Technology, Se-
jong Univ., Seoul, 143-747, Korea. Author Shim is with Dept. of Food &
Biotechnology, Food and Bio-industrial Research Center, Hankyong Natl.
Univ., Ansung, 456-749, Korea. Direct inquiries to author Kim (E-mail:
yongro@snu.ac.kr).
properties (Abdelrahman and Spies 1986).
To overcome the limitations of empirical measurements, such as
mixograph, farinograph, alveograph, and extensograph (Bloksma
1978), many researchers have looked at fundamental rheological
properties. However, it is fair to say that still there is not a unique
fundamental rheological technique that could completely replace
empirical methods mainly due to the fact that dough systems are
far more complex than synthetic polymers for which fundamen-
tal methods are routinely utilized. Nevertheless, the results from
fundamental measurements have provided valuable information
on the relationship between chemical and rheological properties
of complex dough system and how they are affected by processing
conditions, for example, mixing (Zheng and others 2000).
There are still conflicting results regarding relationships between
fundamental rheological properties and processing variables. For
instance, in the case of mixing, some researchers (Dreese and oth-
ers 1988; Mani and others 1992) reported a decrease in storage
modulus with increasing mixing time, whereas others (Bohlin and
Carlson 1980; Navickis 1989; Zheng and others 2000) found an in-
crease with increasing mixing time. The literature on the effects of
resting time on fundamental rheological properties also shows con-
flicting results. A limited number of studies have examined this ef-
fect in combination with mixing time (Abdelrahman and Spies
1986; Mani and others 1992; Lindborg and others 1997). However,
only 2 or 3 different resting times were used, making it difficult to
observe clear trends that enable the understanding of the phys-
iochemical changes occurring during resting. More intensive and
systematic monitoring of this process is needed, and experiments
should also be designed to study the combined effect of resting and
mixing.


C 2007 Institute of Food Technologists Vol. 73, Nr. 1, 2008JOURNAL OF FOOD SCIENCE E1
doi: 10.1111/j.1750-3841.2007.00599.x
Further reproduction without permission is prohibited
 Dough influenced by mixing and resting . . .
The important roles of water during dough processing (Hoseney
1994; Cherian and Chinachoti 1997; Piazza and Schiraldi 1997) have
prompted researchers to investigate the relationship between wa-
ter mobility and functionality of dough using qualitative and quan-
titative NMR analyses (Leung and others 1979, 1983; Richardson
and others 1986; Belton and others 1995; Cherian and Chinachoti
1997; Grant and others 1999; Kim and Cornillon 2001). Results have
provided useful information about physicochemical changes oc-
curring in wheat flour dough during processing. It is expected that
this and future research will provide new insights to aid dough pro-
cessing and to improve the quality of the final products derived
from dough.
The overall goal of this research was to study the physicochemi-
cal changes occurring in dough samples prepared with hard wheat
flour of different strengths. Specifically changes occurring during
E: Food Engineering &
mixing and resting were studied. It is expected that this study will
Physical Properties
provide a clearer understanding of structural changes during dough
processing, which still remain uncertain. It should provide a knowl-
edge base necessary to determine the influence of important pro-
cessing parameters such as mixing time and resting time on differ-
ent flours.
Materials and Methods
Materials
Two flour samples were produced in the Hard Winter Wheat
(HWW) Quality Laboratory at Kansas State Univ. using wheat from
9 experimental hard winter wheat lines. The samples came from
Kansas, Colorado, Oklahoma, and Nebraska. The flours from these
wheats were individually evaluated for milling, rheological, and
baking quality, and then especially blended to produce 2 flours with
different strengths for the purposes of this research. Mixing prop-
erties for the 2 samples are given in Table 1. Sample number 401
had weak mixing properties and sample number 402 had strong
mixing. The blending was carefully conducted to produce 2 flour
samples having as little difference as possible in protein content
(11.6% compared with 12.0%) and water absorption requirements
(62.3% compared with 62.9%) but considerable differences in opti-
mum mixing time (2.88 compared with 5.75 min) and mixing toler-
ance (1 compared with 5).
Dough preparation
To rule out the effects of other ingredients, only flour and dis-
tilled water were used to prepare dough samples. Optimum mixo-
graph water absorption (14% moisture basis) was used for dough
formulation throughout the experiment. A 100 g mixograph mixer
(National MFG. Co., Lincoln, Nebr., U.S.A.) was used for dough mix-
ing. The optimum mixing times for both flour samples were deter-
mined from the obtained mixograms. These optimum times were
2 min 53 s for the weak flour (sample 401) and 5 min 45 s for the
strong flour (sample 402). Under- and overmixing times were de-
termined from these optimum mixing times, and a ratio to their
optimum mixing time was maintained the same for both samples.
Accordingly, mixing times for under- and overmixed dough were
Table 1 --- Mixing properties of hard wheat flour samples.
Sample MC PC MIXO MIXO MIXO TOL
nr (%) (%) WA (%) MT (min) Scale (0 to 6)
401 13.3 11.6 62.3 2.88 1
402 13.5 12.0 62.9 5.75 5
MC = moisture content; PC = protein content; MIXO WA = mixograph water
absorption; MIXO MT = mixograph mixing time; MIXO TOL = mixograph mixing
tolerance.
E2 JOURNAL OF FOOD SCIENCEVol. 73, Nr. 1, 2008
determined as 1 and 5 min for sample 401, and 2 and 10 min for
sample 402. Mixing was conducted at room temperature.
Rheological measurements
A ViscoTech mechanical spectrometer (ReoLogica Instru-
ments, Lund, Sweden) with a plateplate system was used for mea-
suring the rheological properties of the dough samples. Operation,
including temperature control and data handling, was conducted
using a PC-based software provided by the spectrometer manu-
facturer. Serrated plates of 25-mm dia were used to prevent slip-
page during measurement. Before dough samples were loaded into
the rheometer, approximately 2 g of each sample was placed be-
tween 2 parallel aluminum plates and slowly pressed, minimizing
normal stress built up while pressing, to make the sample thick-
ness uniform (2.5 mm) as fully described in Kim (2001). To prevent
dough samples from sticking to the aluminum plates, a thin layer
of mineral oil was applied to the contact area of the plates. The pre-
pressed dough samples were loaded between the serrated plates
in the instrument, and after the desired gap (2 mm) was achieved
the samples were trimmed to get approximately the same diameter
(25 mm) as the upper plate. To prevent moisture loss during mea-
surements, mineral oil was applied around the open edge of the
sample. The gap between the 2 plates was maintained at 2 mm dur-
ing the measurement. All the measurements were performed at a
constant bottom plate temperature of 25 C, which was controlled
using an electrical heater and an air cooler connected to the bot-
tom plate. Dynamic oscillatory tests were performed at a strain of
0.5% and a constant frequency of 1 Hz. The complex modulus (G )
and the phase angle () as well as storage (G ) and loss (G ) moduli
were obtained from these experiments. For the large deformation
test, stress growth was monitored over time at a constant shear rate
of 0.1/s (the maximum shear rate at the rim of the plate). The strain
(the maximum strain at the rim of the plate) was calculated by mul-
tiplying the shear rate by time. From the stress growth curves, the
peak stress and the strain at the peak were determined (Figure 1).
Immediately after mixing, the samples were covered with plas-
tic wrap to prevent moisture loss. During the 3 h of resting, small
pieces of dough were sampled, and small (dynamic oscillatory
tests) and large (stress growth curve) deformation tests were per-
formed on each sample. A resting period much longer than the
actual resting time used in bakeries (approximately 30 min) was
selected for scientific purposes. The 1st measurements were con-
ducted 5 min after mixing. After the 1st and 2nd (15 min after mix-
ing) measurements, sampling was conducted every 15 min. The
whole procedure was repeated for duplicated experiments.
Figure 1 --- Typical stress growth curve for large deforma-
tion test.
Dough influenced by mixing and resting . . .
NMR measurements
NMR relaxation times were measured using a Maran bench-
top NMR spectrometer (Resonance Instruments Ltd., Witney, U.K.)
running at 15 MHz having a variable temperature control system.
PC-based NMR software, provided by the instrument manufac-
turer, was used for pulse sequencing and data acquisition. For every
set of experiments, about 3 g of dough sample was used. Freshly
mixed dough samples were placed into 18 mm (O.D.) glass NMR
tubes, which were then placed into the NMR probe inside the mag-
net module. The spin-lattice (T 1 ) relaxation time was continuously
measured at every 4 to 5 min during the 3 h of resting period us-
ing the inversion recovery pulse sequence (180 - -90 ). The relax-
ation delay was 1 s. The signal acquisition was averaged over only
4 scans due to the long acquisition time required to measure T 1 . A
set of 20 logarithmic scale time spacings between the 180 and 90
pulses was selected. The intervals between the 180 and 90 pulses
varied between 100 s and 2 s. The obtained T 1 curves were fitted
to exponential decay equations using data fitting software (WinFit,
Resonance Instruments Ltd.) to calculate corresponding relaxation
time, T 1 . Three independent resting experiments were performed
for each sample and the obtained data were averaged. To prevent
moisture loss during measurements, the top surface of the sample
was covered with a circular piece of plastic wrap and the NMR tube
was sealed with a rubber stopper.
Results and Discussion
Small deformation test
Figure 2 shows the complex modulus (G ) and the phase angle
of samples 401 (Figure 2A) and 402 (Figure 2B) mixed for differ-
ent periods during 3 h of resting. The strong flour 402 consistently
showed higher G and lower phase angle than the weak flour 401, as
expected. The rheological changes occurring during resting could
be divided into 2 periods: (1) before 30 to 45 min; (2) after 30 to
45 min. After 30 to 45 min of resting, the trends of G and phase an-
gle changes were similar regardless of mixing time and sample dif-
ferences, even though some data showed less significant changes.
G decreased and the phase angle increased gradually after 30 to
45 min of resting. In order to ensure the trends after 30 to 45 min of
resting, linear regression was performed for the G and phase an-
gle data measured after 45 min of resting. This confirmed the de-
crease in G (negative slope) and the increase in the phase angle
(positive slope) with increasing resting time at the 0.05 significance
level (data not shown). Among samples of different mixing times,
overmixed dough samples showed the least significant change. The
decreasing G and increasing phase angle after 30 to 45 min of
resting are probably due to the stress relaxation occurring in the
gluten structure. Ewart (1977) proposed that, during resting, the ex-
tended and unfolded protein molecules revert to the compact state
of lower free energy with the help of disulfide bond interchange.
Protein concatenation becomes more random, and the stress is re-
laxed. It could also be explained by the loops and trains model
E: Food Engineering &
proposed by Belton and others (Belton and others 1994, 1995; Bel-
Physical Properties
ton 1999). During mixing, the number of loops (proteinwater hy-
drogen bonds) decreases and that of trains (proteinprotein hydro-
gen bonds) increases due to the stretching of the protein network.
During resting, the structure relaxes to drive the system to more fa-
vored conformational equilibrium of those loops and trains.
The observed trends on the rheological properties during the ini-
tial 30 to 45 min of resting were more complicated. Consistently,
undermixed dough showed a rapid decrease of G and a rapid in-
crease of the phase angle, whereas overmixed dough showed totally
opposite trends. These trends were more obvious for samples made
of the strong flour (sample 402). The optimally mixed dough ap-
peared to be more stable to changes during the resting of the 1st
30 to 45 min, showing less changes in G and phase angle as com-
pared to those for under or overmixed dough. Abdelrahman and
Spies (1986) found similar trends. Using reconstructed flour with
water absorption of 54%, they found that the storage and loss mod-
uli of the undermixed dough decreased during 30 min of resting,
whereas the behavior of overmixed dough was opposite. Strain of
10%, which is probably higher than that corresponding to the lin-
ear viscoelastic range for dough (generally less than 1%), was used.
Because of the potential nonlinear effects, the absolute values of
Figure 2 --- Changes of the
complex modulus and phase
angle of weak flour dough sample
401 (A) and strong flour dough
sample 402 (B) during resting.
Vol. 73, Nr. 1, 2008JOURNAL OF FOOD SCIENCE E3
 Dough influenced by mixing and resting . . .

storage and loss moduli reported by these authors could deviate showed a similar amount of variation to that of optimally mixed
from true values, but a relative comparison including increasing or dough.
decreasing trends should be meaningful. The same research also For the strong flour sample 402 (Figure 3B), the optimally mixed
found that the closer the dough to its optimum water absorption dough (mixing time of 5.75 min) showed much less variation in
and optimum mixing time, the fewer changes in the measured vis- G , compared to both under-(mixing time of 2 min) and overmixed
coelastic parameters during 30 min of resting. G values illustrated (mixing time of 10 min) dough during the initial period of resting,
in Figure 2 were redrawn to better visualize the effect of resting but similar small variations for longer resting periods.
time (Figure 3). For the weak flour sample (sample 401), Figure 3A
shows that optimally mixed dough (mixing time of 2.89 min) exhib- Large deformation test
ited smaller variation in G than undermixed dough (mixing time
As described previously, 2 parameters from the stress growth
of 1 min). However, G of overmixed dough (mixing time of 5 min)
curve, the peak stress and the strain at the peak, were measured
(Figure 1). These parameters were plotted as a function of resting
time and are presented in Figure 4. Stress growth curves for under-
mixed dough samples did not show any noticeable peak. For this
reason, strain at peak data for undermixed dough is omitted in Fig-
E: Food Engineering &
Physical Properties

ure 4. In this case, the maximum attainable stress in the experiment

Figure 3 --- Effect of mixing time on the complex modulus
of weak flour dough sample 401 (A) and strong flour dough ing period, Overmixed dough prepared with sample 402 showed a
sample 402 (B) during resting.
was recorded instead of the peak stress.
Dough prepared with the strong flour (sample 402) consistently
showed a higher peak stress than that observed in doughs prepared
with the weak flour (sample 401), except when the dough was un-
dermixed. Optimally mixed dough showed the highest peak stress
followed by over- and undermixed doughs for both samples. Dif-
ferences observed in the measured rheological parameters of un-
dermixed dough from the other two were greater in the strong flour
compared to the weak flour. Only after 3 h of resting, optimally and
overmixed dough showed similar peak stress values. Conversely,
during the whole resting period, the strain at the peak of opti-
mally mixed dough was consistently higher than that of overmixed
dough for both samples. It appears that the 2 parameters measured
in these tests could be used to evaluate conditions of optimum
mixing.
The general trend in the change of the peak stress during rest-
ing was similar to that observed in G (Figure 2). After the initial
30 to 45 min of resting period, the peak stress of optimally and
overmixed dough gradually decreased during the remaining rest-
less significant change of peak stress after the initial 30 to 45 min

Figure 4 --- Changes of peak shear

stress and strain at the peak of
weak flour dough sample 401 (A)
and strong flour dough sample
402 (B) during resting.

E4 JOURNAL OF FOOD SCIENCEVol. 73, Nr. 1, 2008

Dough influenced by mixing and resting . . .
of resting period, which agreed well with the result for G . The de-
creasing peak stress also could be related to the relaxation phe-
nomenon of dough as discussed previously. A fast increase in the
peak stress during the initial 30 to 45 min of resting was also ob-
served for optimally and overmixed dough, similarly as observed
from dynamic rheological properties obtained at a low strain (Fig-
ure 2). The undermixed dough samples did not exhibit any distinct
change in the peak stress during the whole resting period. This is
possibly due to the low degree of dough development.
The strain at the peak for both overmixed dough samples 401
and 402 showed a similar extent of initial decrease for the first 30
to 45 min of resting, and then did not show any noticeable change
during the remaining resting period. However, during the first 30
to 45 min of resting, the strain at the peak for the optimally mixed
dough showed an apparent difference between samples. While the
weak flour 401 showed a large decrease of strain at the peak dur-
ing the first 30 to 45 min of resting, it was less evident for the
strong flour sample 402. None of the samples exhibited noticeable
changes in strain after 45 min of resting. Lindborg and others (1997)
performed large deformation tests to study flour samples of differ-
ent strength during 30 min of resting. As they used mixing times
similar or longer than the farinograph optimum mixing time, all
of their samples were either optimally mixed or overmixed. They
found the greater increase of the peak viscosity for longer mixed
dough during 30 min of resting. For the strong flour dough, the
strain at the peak viscosity decreased upon 30 min of resting in the
overmixed dough, but did not change for optimally mixed dough.
Comparing their finding with the above results for the optimally
and overmixed dough, both results agreed well. The above results
would imply that not only the absolute value of the peak stress and
the strain at the peak but also the trend of the strain at the peak
during resting might be related to the flour strength and mixing
degree.
Discussion on the rheological
behaviors of dough during resting
No clear explanation of the phenomena occurring during the ini-
tial 30 to 45 min of resting was found in the literature. However,
the results presented in this study provide evidence, which might
bring clues to explain these phenomena. As a possible explanation
for the rheological property changes of optimally and overmixed
dough during the initial 30 to 45 min of resting, several researchers
(Wang and others 1992; Weegels and others 1995, 1997a, 1997b;
Don and others 2005) reported the possibility of repolymerization,
of glutenin during resting, as measured by the amount of SDS-
insoluble protein. According to their results, the amount of SDS-
insoluble protein increased during resting, and the increase was
largest for the first 45 min. Piazza and Schiraldi (1997) indirectly
supported the structural recovery of dough during resting by mea-
suring biaxial elongation and tensile properties. In their results, an
increase in both the tensile stress at break and the tensile elastic
modulus was observed during resting. The changes were more sig-
nificant during the first 20 min and slow after that. It seems that,
if the partial structural recovery or repolymerization were occur-
ring during resting, these would proceed quickly during the ini-
tial period of this process. Based on the facts described previously
in addition to Ewart (1977, 1979) and Belton (1999)s hypotheses,
the following mechanism for dough resting could be drawn (Figure
5A). Immediately after mixing, the lack of mechanical fission makes
the disulphide interchange reactions dominate to promote repoly-
merization of the glutenin polymer. At the same time, the stretched
and aligned polymers tend to relax and return to a random orien-
tation. However, it appears that the faster rate of the disulphide in-
terchange as compared with glutenin polymer relaxation would re-
sult in an increase of the size of still aligned glutenin polymers.
This would promote cooperative secondary forces such as hydro-
gen bonding and van der Waals interactions among polymers, re-
sulting in the net increase of resistance to deformation. This would
cause an increase in the shear stress during the initial period of
resting. However, as the polymer orientation becomes more and
more random during longer resting periods, the secondary forces
become less and less effective to make cooperative interactions.
From this point the shear stress stays constant or even decreases
during the remaining resting period.
As previously discussed, the rheological behavior of undermixed
dough during the initial 30 to 45 min of resting was opposite to that
of optimally and overmixed dough in small deformation tests. Also,
the trend of the measured G in small deformation tests for under-
E: Food Engineering &
mixed dough did not agree with that of maximum stress obtained
Physical Properties
from large deformation tests. The maximum stress for undermixed
dough was lower than that of optimally and overmixed dough dur-
ing the whole resting period. However, G of undermixed dough
was much larger than that of optimally and overmixed dough im-
mediately after mixing. Higher G and lower phase angle of under-
mixed dough immediately after mixing could be explained by a lack
of dough hydration. It is believed that well-developed gluten struc-
ture increases resistance to deformation. However, another rheo-
logically active substance in wheat flour dough is starch, whose
content is 6 times higher than that of protein. In flour particles,
starch granules and proteins are tightly packed. During hydration
and mechanical abrasion, proteins are separated from flour par-
ticles and form a continuous matrix with starch granules embed-
ded in it. If mixing and hydration time is not sufficient, the result-
ing dough would resemble the structure of a discontinuous protein
network containing a number of unhydrated flour particles com-
posed of starch granules and unseparated proteins (Figure 5B). The
presence of rigid flour particles in undermixed dough could in-
crease solid-like behavior in small deformation tests (higher G and
lower phase angle). However, during resting, flour particles would
become hydrated and their rigid structure would be softened (Fig-
ure 5B). Due to the lack of developed protein structure in under-
mixed dough, the hydration of flour particles during resting would
directly lead to a decrease in G and an increase in phase angle.
On the other hand, the discontinuous protein network contain-
ing unhydrated flour particles in undermixed dough would pro-
vide a number of weak points when the sample is subjected to
a large deformation. For this reason, the maximum stress of un-
dermixed dough, which lacked in developed protein network, was
smaller than that of longer mixed dough. Also, absence of contin-
uous protein network in undermixed dough resulted in the lack
of strain hardening during stress growth, a phenomenon regularly
observed in dough systems (Amemiya and Menjivar 1992; Vanvliet
and others 1992). The discrepancy of small and large deformation
tests was also evidenced in Figure 6. It shows that even though
the maximum stress of both undermixed dough samples hardly
changed, the stress at a lower strain (yield stress values in stress
growth curve, inside circles) decreased during the initial 30 min of
resting, which is consistent with the trend observed during small
deformation tests. The initial stress at a lower strain and G values
for optimally and overmixed dough also showed good agreement in
their trends. It would be difficult to expect a perfect correlation be-
tween the initial stress and G , because the smallest strain recorded
for each stress growth curve was around 50% to 60%, which is much
greater than 0.5% strain used for small deformation tests. How-
ever, the fact that, after 5 min of resting, the initial stress for under-
mixed dough is higher than or at least similar to that of optimally or
Vol. 73, Nr. 1, 2008JOURNAL OF FOOD SCIENCE E5
 Dough influenced by mixing and resting . . .

overmixed dough regardless of their maximum stress values, sup- ate more practical information specifically related to optimum mix-
ports the above hypothesis that small and large deformation tests ing time and flour strength.
could reflect different structural aspects of dough depending on its
hydration status. Evolution of NMR relaxation
This study showed that small deformation tests were more time during mixing and resting
advantageous for understanding molecular interactions and mi- Nuclear magnetic resonance (NMR) is one of the most powerful
crostructure, whereas large deformation tests were useful to evalu- techniques for determining the molecular mobility and dynamic

Figure 5 --- Schematic diagram of

dough resting process for (A)
well-mixed dough and (B)
undermixed dough.
E: Food Engineering &
Physical Properties

Figure 6 --- Large deformation

behavior of weak flour dough
sample 401 (A) and strong
flour dough sample 402 (B)
during initial 30 min of
resting. Circles indicate
shear stress at small strains.

E6 JOURNAL OF FOOD SCIENCEVol. 73, Nr. 1, 2008

Dough influenced by mixing and resting . . .
molecular interactions. Relaxation of spins occurs because of the
interactions between individual spins (the spinspin relaxation, T 2 )
and between a spin and its environment (the spinlattice relax-
ation, T 1 ). During relaxation, resonant energy is exchanged among
spins and environments through the process of molecular motion.
In the fast molecular motion environment, such as a dough system,
T 1 and T 2 relaxation occurs in a similar way; that is, faster molec-
ular motion (higher molecular mobility) brings about slower relax-
ation (longer relaxation time). Hence, the analysis of the relaxation
time is a very powerful tool for determining changes in molecular
mobility (Hore 1995).
A decrease in water mobility during resting was clearly observed
by the T 1 relaxation time presented in Figure 7. This phenomenon
might be related to a continued hydration of dough after mix-
ing. In a system of restricted diffusion such as dough (Umbach
and others 1992; Ohtsuka and others 1994), relatively mobile water
molecules would migrate slowly into neighboring compartments,
such as starch and proteins, promoting a higher level of molecular
interactions between water and the macromolecules. As a result,
overall water mobility decreased, as indicated from the gradual de-
crease of T 1 during resting. Also, as mixing time increased, T 1 de-
creased for both samples. Each data point was an average value of 3
independent measurements (different sets of resting experiments)
and the standard deviation of the triplicates ranged from 0.03 to
0.76 ms for all the data points, showing quite small deviations. Even
though T 1 value differences were relatively small (a few ms) among
under, optimally, and over mixed dough, those differences should
be significant enough when the much smaller standard deviation
values were considered. The decrease of T 1 with mixing time was
attributed to a decrease in water mobility due to increased molec-
ular interactions during mixing. Interestingly, T 1 difference created
by different mixing time remained throughout the resting period.
It seemed that mechanical mixing created additional interactions
between water and other molecules, probably proteins, which was
unachievable from the simple hydration by water migration. Even
though mixing times of sample 402 is considerably longer than
those of sample 401, sample 402 showed greater T 1 relaxation time
at the beginning of the resting period. This implies that there might
be inherent compositional or chemical factors of flour, which af-
fect initial molecular interactions with water. Also, the rate of T 1
decrease during resting was quite different between 2 flour sam-
ples. T 1 of the strong flour dough (sample 402) (Figure 7B) reached
a minimum value within 100 min and then leveled off, whereas that
Figure 7 --- Evolution of T 1 relaxation
times of weak flour dough sample 401
(a) and strong flour dough sample 402
(b) during resting.
E: Food Engineering &
Physical Properties
of the weak flour dough (sample 401) (Figure 7A) decreased slowly
during the entire resting period of 200 min. At the end of the rest-
ing period, 2 flour samples showed similar T 1 values, especially for
optimally and overmixed dough.
Conclusions
F undamental rheological properties of hard wheat flour dough
samples of different strength were investigated as functions of
mixing and resting time. Both small and large deformation tests
consistently showed that stress built up during mixing gradually
relaxed during 3 h of resting period. However, during the first 30
to 45 min of resting, structure reformation was evident, probably
due to the repolymerization of the gluten polymer. The small de-
formation test was able to demonstrate a lack of flexibility in un-
dermixed dough, whereas the large deformation test more clearly
showed the differences between optimal and under-or overmixing,
and between strong and weak flour. From the above studies, it was
shown that small deformation tests were advantageous for under-
standing molecular interactions and microstructure, whereas large
deformation tests were useful to evaluate practical information re-
lated to the processing of dough, notably optimum mixing time and
flour strength. There was no obvious relationship between the evo-
lution of NMR relaxation time (T 1 ) and the rheological property
changes during resting. However, from the NMR relaxation data,
molecular interactions between water and other molecules, which
depended on mixing, resting, and flour type, could be understood
more clearly.
Acknowledgments
Dr. Okkyung Kim Chung and Bradford W. Seabourn of the Hard
Winter Wheat Quality Laboratory at Kansas State Univ. are grate-
fully acknowledged for providing flour samples and flour property
data. This work was partly supported by the Korea Research Foun-
dation Grant funded by the Korean Government (MOEHRD) (KRF-
2006-005-J04703), partly by Technology Development Program for
Agriculture and Forestry, Ministry of Agriculture and Forestry, Re-
public of Korea.
References
Abdelrahman A, Spies R. 1986. Dynamic rheological studies of dough systems. In:
Faridi H, Faubion M, editors. Fundamentals of dough rheology. St. Paul, Minn.: The
American Assn. of Cereal Chemists Inc.
Amemiya JI, Menjivar JA. 1992. Comparison of small and large deformation measure-
ments to characterize the rheology of wheat flour doughs. J Food Eng 16(1/2):91
108.
Vol. 73, Nr. 1, 2008JOURNAL OF FOOD SCIENCE E7
 Dough influenced by mixing and resting . . .
Belton PS. 1999. On the elasticity of wheat gluten. J Cereal Sci 29(2):1037.
Belton PS, Gil AM, Tatham A. 1994. 1 H NMR relaxation time studies of the hydration
of the barley protein c-hordein. J Chem Soc Faraday Trans 90(8):1099103.
Belton PS, Colquhoun IJ, Grant A, Wellner N, Field JM, Shewry PR, Tatham AS. 1995.
FTIR and NMR studies on the hydration of a high-Mr subunuit of glutenin. Int J Biol
Macromol 17(2):7480.
Bloksma AH. 1978. Rheology and chemistry of dough. In: Pomeranz Y, editor. Wheat:
chemistry and technology. St. Paul, Minn.: American Assn. of Cereal Chemists Inc.
Bohlin L, Carlson TL-G. 1980. Dynamic viscoelastic properties of wheat flour dough:
dependence on mixing time. Cereal Chem 57(3):1747.
Cherian G, Chinachoti P. 1997. Action of oxidants on water sorption, 2 H nuclear mag-
netic resonance mobility, and glass transition behavior of gluten. Cereal Chem
74(3):3127.
Don C, Lichtendonk WJ, Plijter JJ, van Vliet T, Hamer RJ. 2005. The effect of mixing
on glutenin particle properties: aggregation factors that affect gluten function in
dough. J Cereal Sci 41(1):6983.
Dreese PC, Faubion JM, Hoseney RC. 1988. Dynamic rheological properties of flour,
gluten, and gluten-starch doughs. II. Effect of various processing and ingredient
changes. Cereal Chem 65(4):3549.
Ewart JAD. 1977. Re-examination of the linear glutenin hypothesis. J Sci Food Agric
28:1919.
Ewart JAD. 1979. Gluten structure. J Sci Food Agric 30:48292.
E: Food Engineering &
Faubion JM, Faridi HA. 1986. Dough rheology: its benefits to cereal chemists. In: Faridi
Physical Properties
H, Faubion M, editors. Fundamentals of dough rheology. St. Paul, Minn.: The Amer-
ican Assn. of Cereal Chemists Inc.
Grant A, Belton PS, Colquhoun IJ, Parker ML, Plijter JJ, Shewry PR, Tatham AS, Well-
ner N. 1999. Effects of temperature on sorption of water gluten determined using
deuterium nuclear magnetic resonance. Cereal Chem 76(2):21926.
Hore PJ. 1995. Nuclear magnetic resonance. New York: Oxford Univ. Press.
Hoseney RC. 1994. Principles of cereal science and technology. 2nd ed. St. Paul, Minn.:
American Assn. of Cereal Chemists Inc.
Kim Y. 2001. Physicochemical properties of hard wheat flour dough influenced by
processing. Dept. of Food Science, West Lafayette, Ind.: Purdue Univ.
Kim YR, Cornillon P. 2001. Effects of temperature and mixing time on molecular mo-
bility in wheat dough. Lebensm Wiss Technol (Food Sci Technol) 34(7):41723.
Leung HK, Magnuson JA, Bruinsma BL. 1979. Pulsed nuclear magnetic-resonance
study of water mobility in flour doughs. J Food Sci 44(5):140811.
Leung HK, Magnuson JA, Bruinsma BL. 1983. Water binding of wheat-flour doughs
and breads as studied by deuteron relaxation. J Food Sci 48(1):959.
E8 JOURNAL OF FOOD SCIENCEVol. 73, Nr. 1, 2008
Lindborg KM, Tragardh C, Eliasson AC, Dejmek P. 1997. Time-resolved shear viscosity
of wheat flour doughseffect of mixing, shear rate, and resting on the viscosity of
doughs of different flours. Cereal Chem 74(1):4955.
Mani K, Eliasson AC, Lindahl L, Tragardh C. 1992. Rheological properties and bread-
making quality of wheat-flour doughs made with different dough mixers. Cereal
Chem 69(2):2225.
Navickis LL. 1989. Rheological changes of fortified wheat and corn flour doughs with
mixing time. Cereal Chem 66(4):3214.
Ohtsuka A, Watanabe T, Suzuki T. 1994. Gel structure and water diffusion phenomena
in starch gels studied by pulsed-field gradient stimulated echo NMR. Carbohydr
Polym 25(2):95100.
Piazza L, Schiraldi A. 1997. Correlation between fracture of semisweet hard biscuits
and dough viscoelastic properties. J Texture Stud 28(5):52341.
Ramkumar DHS, Bhattacharya M, Menjivar JA, Huang TA. 1996. Relaxation behavior
and the application of integral constitutive equations to wheat dough. J Texture
Stud 27(5):51744.
Richardson SJ, Baianu IC, Steinberg MP. 1986. Mobility of water in wheat-flour sus-
pensions as studied by proton and o-17 nuclear-magnetic-resonance. J Agric Food
Chem 34(1):1723.
Umbach SL, Davis EA, Gordon J, Callaghan PT. 1992. Water self-diffusion coefficients
and dielectric-properties determined for starch-gluten-water mixtures heated by
microwave and by conventional methods. Cereal Chem 69(6):63742.
Vanvliet T, Janssen AM, Bloksma AH, Walstra P. 1992. Strain-hardening of dough as a
requirement for gas retention. J Texture Stud 23(4):43960.
Wang GIJ, Faubion JM, Hoseney RC. 1992. Studies of the breakdown and reformation
of SDS insoluble glutenin proteins with dough mixing and resting. Lebensm Wiss
Technol (Food Sci Technol) 25(3):22831.
Weegels PL, Orsel R, Vandepijpekamp AM, Lichtendonk WJ, Hamer RJ, Schofield JD.
1995. Functional properties of low M(R) wheat proteins.2. Effects on dough prop-
erties. J Cereal Sci 21(2):11726.
Weegels PL, Hamer RJ, Schofield JD. 1997a. Depolymerisation and re-polymerisation
of wheat glutenin during dough processing.2. Changes in composition. J Cereal Sci
25(2):15563.
Weegels PL, Hamer RJ, Schofield JD. 1997b. Functional properties of low M(R) wheat
proteins.3. Effects on composition of the glutenin macropolymer during dough
mixing and resting. J Cereal Sci 25(2):16573.
Zheng H, Morgenstern MP, Campanella OH, Larsen NG. 2000. Rheological properties
of dough during mechanical dough development. J Cereal Sci 32(3):293306.