Korea-Australia Rheology Journal

Vol. 18, No. 3, September 2006 pp. 127-131

On the compressibility of bread dough

1 1
Chunguang Wang, Shaocong Dai and Roger I. Tanner *

Mechanical & Electrical Engineering College, Inner Mongolia Agriculture University,

Hohhot, Inner Mongolia, P.R. China
1
School of Aerospace, Mechanical & Mechatronic Engineering, University of Sydney, Sydney, NSW 2006, Australia
(Received May 17, 2006, final revision received August 23, 2006)

Abstract
Few investigations of bread dough compressibility have been reported in the literature, despite the fact that
high compression stresses are often reached in processing. Here we report some experiments on the com-
pressibility of an Australian wheat bread dough under compressive stresses up to 5 MPa, and show that the
results are consistent with a mathematical model of bread dough containing entrained air. The implications
for tensile testing are also considered.
Keywords bread dough, compressibility, bulk modulus
:

1. Introduction cision scale. The mixing operation was conducted at a tem-

perature of 24 C and under ambient humidity in an air-
o

Bread dough is a soft viscoelastic solid, often considered conditioned laboratory. The rotation speed of the Dough-10
to be incompressible. However, in processing quite high g mixograph was measured to be 71 rpm. The mixing time
compressive stresses may be reached (~3 MPa) and com- to peak bread dough development was determined from the
pressibility may be important. There are some results in the mixing curve, which was 7 minutes for each mixing. After
literature, for example Glucklich and Shelef (1962), and mixing, a little Vaseline oil was smeared on the outside sur-
Bloksma (1957). However, Glucklich and Shelef used face of the bread dough sample, and the bread dough was
quite low pressures (~0.1 MPa) in their experiments, and put into a plastic bag and then the bag was sealed to pre-
Bloksma simply quoted their work. Charalambides et al. vent moisture loss. The bread dough sample relaxed for 45
(2002) in fitting their biaxial extension data, considered the min in the bag before the compression experiment. In
material as either incompressible or with a value of the every compression experiment the quantity of the bread
bulk modulus of 7.3 kPa. They did not do any experiments dough sample was equal to 62.8 g including wheat flour,
to determine compressibility directly. Given the paucity of distilled water and salt. The density of the dough was about
data available, we present here some new experiments up 1207 kg/m .3

to 5 MPa compression stress, plus a model of the bulk The bread dough sample was made as a cylinder with a
modulus. In this paper the bulk response of the dough is diameter of about 37 mm before the bread dough sample
regarded as elastic; further work would be needed to refine was put into the compression chamber of the compression
this picture. We also consider the implications for tensile unit that was designed for the project. The compression
testing. unit is shown in Fig. 1. The internal diameter of the steel
cylinder of the compression unit was 48 mm and the wall
2. Materials and methods thickness was 5 mm. All compression experiments were
performed with this unit.
2.1. Materials
The wheat flour was a brand of commercial Australian 2.2. Experimental method
flour, variety JANZ wheat, grown in 2001 at Narrabri, To begin, the chamber was filled with silicone oil (Dow
NSW, and milled in a Buhler experimental mill. The bread Corning grade 200). Before the compression experiments
dough was mixed in a Dough-10 g mixograph. 9.5 g of were conducted the air vent was opened and the air in the
wheat flour was mixed, together with 6 g distilled water compression chamber was expelled by pressing on the
and 0.2 g salt, as determined by using a digital high pre- plunger by hand. An Instron 5567 testing machine was
used to apply and measure the load. The maximum com-
*Corresponding author: rit@[Link] pression stress was 4.97 MPa, the plunger moved at 0.05
© 2006 by The Korean Society of Rheology mm/s, and the interval of information sampling was 5s.
Korea-Australia Rheology Journal September 2006 Vol. 18, No. 3 127
 Chunguang Wang, Shaocong Dai and Roger I. Tanner

Every experiment was carried out in triplicate, and the

results were averaged.
We show the effect of using silicone oil alone in the
chamber in Fig. 2(a). As can be seen, the percentage com-
pression (or volume strain) vs. pressure curve is repro-
ducible and is a nonlinear function of the pressure. We now
consider the case of dough.
Before placing the bread dough sample into the com-
pression chamber 30 ml of silicone oil was placed there,
and then the Vaseline-coated bread dough sample was put
into the silicone oil. The top surface of the bread dough
sample in the compression chamber was the same height as
or a little lower than the surface of the silicone oil. The
results are shown in Fig. 2(b); some scatter in the results is
evident, and the mean value of the three trials was used in
subsequent analysis.
Fig. 1. The compression unit. The steel walls are 5 mm thick and
the inner diameter is 48 mm. 3. Compressibility
The magnitude of the bread dough compressibility is
clearly not constant and depends on the pressure acting on
the bread dough. So, the pressure in the compression
chamber and the diminution of the total volume of the
bread dough and the silicone oil, as well as the diminution
of the volume of the silicone oil, were measured in the
experiments, and then the diminution of the volume of the
bread dough was calculated from the experimental results.
We used equation (1).
∆v = ∆vd + ∆vo (1)
where ∆v is the diminution of the total volume, mm , ∆vd3

is the diminution of the volume of the bread dough, mm 3

and ∆vo is the diminution of the volume of the silicone oil,

mm . In equation (1) we need, in principle, to consider the
3

expansion of the steel chamber. Using elastic cylinder the-

ory (Timoshenko, 1951) we calculated that the increase in
cylinder volume due to inner pressure was only of order
1 % of the total volume change of the samples under the
same pressure. Hence we have neglected the effect of cyl-
inder expansion as being outside the precision of the mea-
surements.
We assume that the diminution in volume of the silicone
oil can be computed using the results of Fig. 2(a) and the
fact that the silicone oil initial volume is 30 ml. The initial
volume of the dough was 52 ml, and hence the change in
dough volume can be found by subtraction. Results are
shown in Table 1.
The formula for the percentage volume strain is the fol-
lowing.
Fig. 2. (a) The compression stress as a function of the percentage γ = ∆-----v- × 100 (2)
V
volume strain of the silicone oil. (b) The compression
stress as a function of the percentage volume strain of the where γ is the percentage volume strain, ∆v and V are the
bread dough and silicone oil. diminution of the volume and the initial volume respec-
128 Korea-Australia Rheology Journal
 On the compressibility of bread dough

Table 1. The diminution of volume (total, silicone oil and bread dough)
Diminution of the Stress/(MPa)
volume/(mm3) 0.02 0.04 0.06 0.08 0.10 0.40 0.60 0.80 1.00 1.20
Total (∆v) 533 1016 1413 1742 2015 3667 4039 4280 4446 4567
Silicone oil (∆vo) 42 79 108 134 161 406 524 643 740 817
Bread dough (∆vd) 491 937 1305 1608 1854 3261 3515 3637 3706 3750
Diminution of the Stress/(MPa)
volume/(mm3) 1.40 1.60 1.80 2.00 2.50 3.00 3.50 4.00 4.50 4.97
Total (∆v) 4667 4750 4825 4891 5037 5165 5283 5396 5499 5597
Silicone oil (∆vo) 885 943 998 1048 1167 1275 1378 1475 1570 1663
Bread dough (∆vd) 3782 3807 3827 3843 3870 3890 3905 3921 3929 3934

7.56%. Thus the compressibility of the bread dough begins

to lessen with the increase of stress when the compression
stress is more than about 1 MPa.
4. Bulk modulus
In the experiments the initial volume of the bread dough
and the silicone oil were 52000 mm and 30000 mm ,
3 3

respectively. The bulk modulus of the bread dough was

calculated according to the experimental results in Table 1.
The calculated results for the bulk modulus K of the bread
dough are presented in Table 2.
The relationship curve between the bulk modulus and the
volume strain and the compression stress are given in Fig.
4 and Fig. 5, respectively.
Fig. 3. Showing the diminution of the volume of the bread dough The slope of the curve in Fig. 4 changes significantly
with the compression stress. with strain. The curve can be divided into two sections
according to the slope. The volume strain in the first sec-
tion is zero through about 6.8%. In this section the increase
tively, mm . The average bulk modulus K is just ∆p/γ.
3
of the bulk modulus is slow. When the volume strain of the
The relationship between the diminution of the volume bread dough is increased to 6.8% the bulk modulus
of the bread dough and the compressive stress is given in increases from the minimum value to about 10 MPa. In the
Fig. 3 and Table 1. second section, the percentage of the compression ratio of
When the compression stress is 1 MPa considerable the volume is more than 7.0%. In this section, for larger
compressibility of the bread dough is evident, about 7.13 %. strains, the bulk modulus of the bread dough increases very
But when the compressive stress reaches nearly 5 MPa the quickly, up to about 66 MPa, and the bread dough begins
change of the volume of the bread dough is only about to stiffen rapidly.

Table 2. The bulk modulus and the volume strain of bread dough
Stress/(MPa) 0.02 0.04 0.06 0.08 0.10 0.40 0.60 0.80 1.00 1.20
Volume strain/(%) 0.94 1.80 2.50 3.09 3.57 6.27 6.76 6.99 7.13 7.21
Bulk modulus/(MPa) 2.12 2.22 2.39 2.59 2.81 6.38 8.85 11.48 14.12 16.70
Stress/(MPa) 1.40 1.60 1.80 2.00 2.50 3.00 3.50 4.00 4.50 4.97
Volume strain/(%) 7.27 7.32 7.36 7.39 7.44 7.48 7.51 7.54 7.56 7.56
ulk modulus/(MPa) 19.2 21.9 24.6 27.1 33.6 40.1 46.6 53.1 59.5 65.7

Korea-Australia Rheology Journal September 2006 Vol. 18, No. 3 129

 Chunguang Wang, Shaocong Dai and Roger I. Tanner

modulus of the bread dough were developed. The corre-

lation of the average bulk modulus K with the compression
stress as well as the bulk modulus with the percentage vol-
ume strain are given as equations (3) and (4), respectively.
K = 12.81∆p + 1.379
R = 0.9999
2
(3)
K = 1.779 + 0.117γ + 5.47 × 10− e − 2 17 ( 1.05+5.614γ)

R = 0.9940
2
(4)
Here K is the bulk modulus of the bread dough, MPa, ∆p
is the compression stress, MPa, γ is the percentage volume
strain, and R is the correlation factor.
When the compression stress ∆p or the volume strain γ is
equal to zero the bulk moduli of the bread dough are found
to be 1.379 and 1.779 MPa respectively. These two values
are reasonably close. These results indicate that the two
Fig. 4. The relationship between the bulk modulus of the bread correlations to predict the bulk modulus of the bread dough
dough and the percentage volume strain of the bread are appropriate. The regression curves are given in Fig. 4
dough. and Fig. 5, respectively.
5. Prediction of the bulk modulus of dough
The dough mixing process inevitably entrains air, and so
we adopt a model which recognizes this. Let us suppose
that the dough initially contains a volume fraction x of air,
plus a volume fraction (1 − x) of solid dough. Suppose the
bulk modulus of the solid phase, in the absence of
entrained air, is k. The entrained air is considered to be at
atmospheric pressure initially, say p . Upon compression to
a pressure (p + ∆p), the volume strain in the solid is ∆------p ,
0

0
k
∆ p
and the volume of the air also changes, by a factor ----------------
p0 + ∆ p
times the initial volume, assuming isothermal compression.
Thus the total diminution in a unit volume of dough is:
Fig. 5. The relationship between the bulk modulus of the bread ( 1 – x ) ∆------p + x ---------------
∆p - (5)
k p0 + ∆ p
dough and the compressive stress experimental results.
The correlation (3) and the theoretical curve (6) are all in The mean bulk modulus (K) of the dough is then
good agreement. ∆p - , which gives
----------------------------------
volume strain

We also found that the average bulk modulus of the ---1- = (-------------
1 – x)- + ---------------
x - (6)
K k p0 + ∆ p
bread dough increased nearly linearly with the increase of
the compression stress, see Fig. 5. Using the data of Fig. 5, fitting the data at ∆p = 1 MPa
In order to find the functional relationship between the and ∆p = 5 MPa, we find k = 5830 MPa and x = 0.077.
bulk modulus of the bread dough and the compression Thus the solid phase is typical of a solid polymer; the
stress, as well as the percentage of the compression ratio of 7.7 % air entrained is typical of our experiments and those
the volume of the bread dough, we analyzed the functional of others (Bloksma 1971).
relationship between the (average) bulk modulus and the
compression stress, as well as the percentage of the com- 6. Conclusion
pression ratio of the volume of the bread dough, by a
regression method. Two correlations to predict the bulk the main purpose of this work is to measure and ana-
130 Korea-Australia Rheology Journal
 On the compressibility of bread dough

lyze the compressibility and the bulk modulus of the pressibility needs to be considered.
bread dough by direct experiment. based on the exper-
imental results the bulk modulus of the bread dough was Acknowledgments
calculated up to 5 mpa compressive stress. two corre-
lations to predict the bulk modulus of the bread dough This work was supported by the Australia Research
were developed. we find that the relationship between Council, and this support is gratefully acknowledged.
the compression stress and the bulk modulus of the bread
dough is nearly linear. the study indicates that the exper- References
imental results show an excellent agreement with a sim-
ple model that includes entrained air. clearly this may Bloksma A.H., 1957, A calculation of the shape of the alveo-
have implications for tensile and biaxial testing of grams of some rheological model substances. Cereal Chem-
dough, where large below-atmospheric pressures may istry 34, 126-136.
occur and for processing, where large positive pressures Bloksma, A.H., 1971, Baker’s dough , Ch. 14 in R. Houwink,
occur. for example, using the same dough, we found that [Link] Decker (eds) Elasticity, Plasticity and Structure of Mat-
ter, 3rd edn., Cambridge [Link]
at an elongation rate of 0.1 s− fracture occurred at a ten-
1

Charalambides, M.N., L. Wanigasooriya and J.G. Williams, 2002,

sile stress of about 35 kpa. thus a negative gauge pres- Biaxial deformation of dough using the bubble inflation tech-
sure of about 30% of this, say 0.01 mpa, exists in the nique. 2. Numerical modelling. Rheol. Acta 41, 541-548.
specimen. according to the simple entrainment model Glucklich, J. and L. Shelef, 1962, An investigation into the rheo-
used here, the bulk modulus then decreases; very high logical properties of flour dough. Studies in shear and com-
rates of extension would decrease it even further leading pression. Cereal Chemistry 39, 242-255.
to failure. hence, we conclude that the incompressible Timoshenko, S., 1951, Strength of Materials 2nd edn., vol 2. Van
model of dough is generally valid, but occasionally com- Nostrand, New York.

Korea-Australia Rheology Journal September 2006 Vol. 18, No. 3 131