Hindawi Publishing Corporation

Advances in Mechanical Engineering

Volume 2014, Article ID 651260, 8 pages
http://dx.doi.org/10.1155/2014/651260

Research Article
Drag Reduction of a Pipe Flow Using Nata de Coco Suspensions

Satoshi Ogata,1 Gunawan,2 Jun Warashina,1 and Yanuar2

1
Department of Mechanical Engineering, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji-shi, Tokyo 192-0397, Japan
2
Department of Mechanical Engineering, University of Indonesia, Kampus UI, Depok 16424, Indonesia

Correspondence should be addressed to Satoshi Ogata; ogata-satoshi@tmu.ac.jp

Received 12 May 2014; Revised 29 July 2014; Accepted 30 July 2014; Published 27 August 2014

Academic Editor: Jinjia Wei

Copyright 2014 Satoshi Ogata 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.

The addition of drag-reducing agents to reduce pipe friction loss has attracted attention as a method to conserve energy. In addition
to reducing drag, these agents are required to have a low environmental load and conserve natural resources. Therefore, naturally
occurring biopolymer additives, which are considered to have a low environmental load, have recently received much attention.
Here we focused on nata de coco, a type of biopolymer that exhibits low mechanical degradation, and found that it reduced drag by
up to 25% at a concentration of 50 ppm. With respect to the drag reduction (DR) mechanism, we investigated the relation between
DR phenomena and the fiber structure of nata de coco by visualization. As a result, we found that the DR effect appeared only when
a network of nata de coco fibers was formed in the suspension. In addition, DR increased as the size of the network of nata de coco
fibers increased.

1. Introduction above, the addition of a drag-reducing agent (typically a

There has been growing interest in research on energy
consumption with the aim of resolving environmental
issues. Furthermore, technological development is expected
to encourage energy and environmental solutions. Conse-
quently, in association with energy conservation techniques,
the addition of a drag-reducing agent to reduce pipe friction
loss has attracted attention since the discovery that adding
polymers to a turbulent flow reduced the pressure drop [1].
For instance, one study investigated the effect of friction drag
reduction (DR), caused by the injection of drag-reducing
agents into a pipe flow in a heat-transport system, on energy
consumption [2]. The authors found that this reduction of
pressure loss in a pipeline significantly reduced energy con-
sumption because it reduced the cost of operating pumps in
the system. Drag-reduction techniques also have the potential
to enable smaller and more efficient fluid flow devices.
Many techniques can be used to reduce flow resistance.
In particular, the method of adding a drag-reducing agent
to the liquid has attracted attention because it may provide a
low-cost solution that uses an existing device. As mentioned
polymer [3, 4]) to the fluid base has been shown to achieve
the highest reduction in flow resistance. DR by adding
surfactants [57] has been applied to regional heating systems
and air-conditioning systems in buildings. However, because
the additives used in these techniques are composed of
synthetic materials, careful disposal of the drained solutions
is required to prevent them from contaminating rivers and
soil. For this reason, such drag-reducing additives are used
only in closed-loop pipelines. In contrast, naturally occurring
biopolymer additives have recently received considerable
attention because they are considered to have a low environ-
mental load. Accordingly, much consideration was given to
the drag-reducing mechanism of biopolymer by a number of
studies [811]. However, the DR mechanism of biopolymers
remains to be determined. Moreover, there are mechanical
degradation problems.
In recent years, we have focused on nata de coco, which is
a type of biopolymer that is easily available and inexpensive.
We have found that it reduces drag by up to 11% at a
concentration of 400 ppm in research using a syringe pump
[12]. Furthermore, nata de coco suspensions show an effect

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016

2 Advances in Mechanical Engineering

Figure 1: Commercial production of nata de coco.

Figure 2: Micrograph of dried nata de coco (35).

similar to type-B DR [13], which is seen in fully extended
polymer molecules. However, there are still many unex-
plained phenomena of nata de coco suspensions associated
with DR.
Therefore, the purpose of this study is to reveal the drag-
reducing behavior of nata de coco suspensions in a pipeline-
flow loop by pressure loss measurements and visualization.
Furthermore, nata de coco is used for reinforcement in
composite materials owing to its high strength and complex
network structure. Therefore, nata de coco has the potential
to exhibit low mechanical degradation because of the high
mechanical strength of its filaments. In other words, nata de
coco has the potential to resolve the problem of mechanical
deterioration as well.
2. Experimental Setup and Procedure
2.1. Nata de Coco Suspensions. The tested nata de coco
suspensions were made from commercial sources. First, the
nata de coco was washed with tap water and then immersed in
tap water for 24 h to remove sugar (Figure 1). Second, nata de
coco was pressed at 10 ton/m2 and then dried in a refrigerator.
Figure 2 shows the dried nata de coco. The mass of dried nata
de coco was 0.26% compared to that of commercial nata de
coco before it was dried. This suggested that cellulose, which
is a major component of nata de coco, was extracted from
the commercial product because nata de coco contains over
99% water and less than 1% cellulose. Finally, nata de coco
suspensions were prepared by mixing the dried nata de coco
with tap water in a blender. The blender was operated at a rate
of 11,000 rpm with a mixing time, blender , of 1 to 10 min. The
suspension concentration, , was determined by the mass
ratio of dried fiber to the amount of total tap water in the
pipeline.
2.2. Experimental Apparatus. Figure 3(a) shows a schematic
of the experimental apparatus used in this study. The total
length of the pipeline, which was made of stainless steel, was
5100 mm, and the experimental apparatus consists of a tank,
pump, injection device, pressure transducer, and flow meter.
Four liters of tap water were used as a solvent for nata de coco
suspensions in this study. The suspensions were circulated
by a pump (25SCD 5.25: Ebara Technologies, Inc.), which

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016

Advances in Mechanical Engineering 3

Syringe pump Pressure transducer

Flow

Nata de coco suspension

1020 mm 300 mm 20 70 930 300

3
Valve
1 1
Flowmeter Test section

25

15
p

Pump Tank Flow direction

(a) (b)

Figure 3: (a) Experimental apparatus. (b) Details of test section and injection port.

1.5 the entrance length. Nata de coco suspensions were injected

through an injection port. The pressure drop was measured
1.4
by injecting additives into the circulating tap water. The mean
flow rate of the injection was 0.7 mm/s, which was about
0.05% of the mean flow rate in the pipe. The concentration
Viscosity (mPas)

1.3 of the nata de coco suspension in the pipeline was adjusted to

1.2
1.1
1 The viscosity of the test fluids was measured by means of
102
Shear rate (1/s)
50 ppm
Tblender = 1.5 min (DR = 20%) Tblender = 10 min (DR = 0%)
Tblender = 3 min (DR = 6%) Water
Figure 4: Viscosity of nata de coco suspension.
was controlled by an inverter (Sysdrive 3G3MV: Omron Elec-
tronics, Inc.). The pressure difference was measured using a
Validyne differential pressure transducer (DP15, 0.25% F. S.
accuracy). The pressure transducer data was recorded by a
data logger (NR-1000: Keyence Japan). The pressure drop was
determined by averaging 50 data points measured at intervals
of 100 ms. The flow rate was measured by an electromag-
netic flow meter (COPA-XE: Nippon Flow Cell Co., Ltd.).
Figure 3(b) shows a detailed schematic of the pipe from the
injection port to the measurement part for the pressure drop
(the test section). The test section had an inner diameter of
15 mm, and the length with the pressure drop was 300 mm.
The temperature of the suspension was maintained within
1.0 C of the target temperature (20 C) by using a constant-
temperature reservoir (LTB-125: As One Corporation). The
flow was fully developed because the injection port (inner
diameter: 3 mm) was set at a point 1020 mm from upstream
pressure port (inner diameter: 1 mm) in consideration of
values of = 12.5, 25, 50, and 100 ppm (the corresponding
concentrations before injection were 625, 1250, 2500, and
5000 ppm).
3. Results and Discussion
103 a cone-plate viscometer (HAKKE Reostress 600). Figure 4
shows the viscosity of nata de coco suspensions at =
50 ppm. The viscosity of the nata de coco suspensions shows
the shear thinning behavior and increases with decreasing
blender (or with increasing DR, as will be shown in Figure 9).
Figure 5 shows the effect of the injected nata de coco on
DR. To quantitatively estimate the amount of DR, the exper-
imental results are expressed in terms of the DR coefficient:
DR = 100%, (1)
where and denote the friction coefficients for tap water
and nata de coco suspensions, respectively.
The horizontal axis represents the injection volume ,
which is the total injection volume of nata de coco suspension
(625, 1250, 2500, and 5000 ppm); the vertical axis represents
DR. This figure shows that DR was 0% when the tap water
was injected into the pipe. Therefore, the injection of tap
water has minimal influence on DR. However, DR increased
with an increase in the total injection volume and concen-
tration of nata de coco suspensions. The concentrations of
the suspensions in the pipe after injection were 12.5, 25,
and 50 ppm. From these experimental results, it can be seen
that the DR effect occurs immediately after injection, but
the DR value requires time to become stable. However, it is
believed that if one waits a long time to stabilize the DR value,

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016

4 Advances in Mechanical Engineering

30

20

DR (%)
10

0 0.0001 0.0002 0.0003 0.0004

V (m3 )

50 ppm 12.5 ppm

25 ppm 0 ppm

Figure 5: Effect of injection on DR.

40 40

Cw = 100 ppm Re = 11000

Re = 11000 Tblender = 1.5 min
30 30
DR (%)

20
DR (%)

20

10 10

0 0

0 5 10 0 50 100
H (hour) Cw (ppm)

Figure 6: Effect of degradation on DR. Figure 8: Effect of concentration on DR.

0.045
= 0.3164 Re0.25
0.04
30
Cw = 50 ppm
0.035
Re = 11000
20

0.03
DR (%)

0.025 10

10000 20000 0
Re
100 ppm 12.5 ppm 0 2 4 6 8 10
50 ppm 0 ppm Tblender (min)
25 ppm
Figure 9: Effect of the operating time of the blender on DR.
Figure 7: Friction factor of nata de coco suspensions.

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016

Advances in Mechanical Engineering
(a) Before experiment (50 ppm)
Figure 10: Change of the fiber form (blender = 1.5 min, DR = 20%).
mechanical degradation occurs. Therefore, the effect of this
degradation has been investigated.
Figure 6 shows the results of measuring the DR effect
in an experiment conducted over a long duration. The
horizontal axis represents the experimental time , and the
vertical axis represents the DR. This figure shows that the DR
was maintained at almost a constant value for approximately
14 h. Under the flow rate used in this experiment, nata
de coco circulated through the pump about 10,000 times.
Therefore, it was revealed that the nata de coco fiber exhibits
low mechanical degradation. Hereafter, all figures are plotted
from data taken only after the experiment has been running
for an hour, at which time it is considered that the DR values
have become sufficiently stable.
Figure 7 shows the experimental results for the friction
coefficient. The dashed line in the figure was obtained by
using the Blasius equation for turbulent flow. The measured
friction coefficient of tap water was within 1% of the value
calculated using the Blasius equation. The nata de coco
suspensions had lower friction coefficients than tap water for
each concentration (100 ppm). The DR increased with the
increasing concentration of the suspensions, and the slope of
this increase was parallel to the line of the Blasius equation.
Figure 8 shows the effect of concentration on DR at 0
100 ppm in the case when the flow rate in the pipe is constant
(Re = 11000) and blender is 1.5 min. The figure shows that
DR increased with the increasing concentration and reached
values up to 25% more than its original value when the
concentration was increased to 50 ppm. At a concentration
of more than 50 ppm, DR was nearly constant.
We researched the relation between the scattering time
in the blender and DR because DR was significantly affected
by scattering time during the experiment. Figure 9 shows the
effect of scattering time in the blender, blender , on DR. blender
on the horizontal axis is the mean of the scattering time in
the blender when the nata de coco suspensions were prepared.
DR was observed for scattering times of 13 min, although a
change in DR was not observed in the case of long scattering
times (more than 5 min). This shows that DR is significantly
affected by the scattering time in the blender. In the case
Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016
5
(b) After experiment (50 ppm)
where the scattering time was less than 1 min, experiments
could not be conducted owing to the insufficient dispersion
of nata de coco fibers.
To investigate the mechanisms of DR, we observed fiber
formation at scattering times of 1.5 and 10 min. Micrographs
of nata de coco suspensions were taken with a high-speed
microscope (VW-9000 and VH-Z35, Keyence, Japan). A
droplet of the nata de coco suspension on a glass plate
was measured at a magnification of 35. Figure 10 shows
micrographs of nata de coco fibers (a) before and (b) after
experiments in the case of 20% DR at a blender value
of 1.5 min. After the experiments, the fibers were tangled,
and a large network of fibers was formed. Figure 11 shows
micrographs of nata de coco fibers in the case of 0% DR
at a blender value of 10 min. Significant differences were not
observed between the micrographs taken (a) before and (b)
after the experiments. These results show that a large network
of fibers had formed when DR occurred in spite of the
small difference between Figures 10(a) and 11(a). Why the
DR differs without an obvious difference in the initial state
is unclear. We surmised that this was because the difference
was too small to be measured by the microscope. Therefore,
we next observed microregions with a scanning electron
microscope (SEM) (VHX-D500/D510: Keyence, Japan).
Figures 12 and 13 show the observation results. Given
the nature of SEM observations, moisture has been removed
from the samples. In other words, the micrographs show
fibers in their solid state. No apparent difference in mesoscale
networks is observed between the initial states at blender =
1.5 min and blender = 10 min. However, small-scale networks
are far more common at blender = 1.5 min than at blender =
10 min. After the experiment, the fiber network at blender =
1.5 min becomes bundled to form a large network. In contrast,
the network at blender = 10 min retained a large number
of holes. We found that the initial state of the fiber has a
significant impact on forming a large network. Therefore, we
suggest that a large network of nata de coco fibers may play
an important role in DR. However, evaluating the network
quantitatively is difficult because we measured its length and
diameter approximately from Figures 10 and 11.
6 Advances in Mechanical Engineering

(a) Before experiment (50 ppm) (b) After experiment (50 ppm)

Figure 11: Change of the fiber form (blender = 10 min, DR = 0%).

(a) Before experiment (50 ppm) (b) After experiment (50 ppm)

Figure 12: Observation by SEM (blender = 1.5 min, DR = 20%).

Figure 14 shows the results of measuring the size of the 4. Conclusions

fiber formed after the experiment. We measured the length
and diameter of the fiber from the micrographs. Elongated
white objects were considered fibers, whereas objects that
could not be determined were not counted. If the network
structure was branched as in Figure 10(b), we measured only
the maximum length of the network. The diameter was
measured at the largest part in the network. The horizontal
axis shows the fiber diameter, (mm), and the vertical axis
shows the fiber length, L (mm). The solid line in the figure
shows the average value of the fiber aspect ratio. DR increased
with increasing fiber diameter and length, as noted in Figures
14(a), 14(b), and 14(c), while significant differences were not
observed in fiber size before the experiment. Furthermore, it
was found that DR decreased in the case where large fibers
were not formed at long values of blender . However, much
the same aspect ratio was observed in each experiment. From
these results it can be seen that DR was observed in the case
where a large network of nata de coco fibers was formed, while
in the case where a large network was not formed, no DR was
observed.
In the present work, we investigated the drag-reducing
behavior of nata de coco suspensions in a pipeline-flow loop.
We observed the following findings:
(1) DR by use of nata de coco suspension exhibited low
mechanical degradation,
(2) DR increased with increasing concentration and
reached 25% when the concentration was increased
to 50 ppm,
(3) DR was observed only when a network of nata de coco
fibers was formed in the suspension,
(4) DR increased with increasing network size of nata de
coco fibers.
Nata de coco is considered a candidate drag-reducing
agent in addition to existing polymer and surfactant solu-
tions. Moreover, nata de coco can be applied to full-scale
experiments for long durations because it has low mechan-
ical degradation. Therefore, nata de coco is highly suitable
for application to actual equipment devices that must be
enhanced with a long-lasting DR effect.

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016

Advances in Mechanical Engineering 7

(a) Before experiment (50 ppm) (b) After experiment (50 ppm)

Figure 13: Observation by SEM (blender = 10 min, DR = 0%).

20 Aspect ratio = 44 20

Aspect ratio = 48
L (mm)
L (mm)

10 10

Tblender = 10 min Tblender = 3 min

DR = 0% DR = 6%

0 0
0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5
D (mm) D (mm)

Before experiment Before experiment

After experiment After experiment
(a) Fiber size (blender = 10 min, DR = 0%) (b) Fiber size (blender = 3 min, DR = 6%)

20 Aspect ratio = 49
L (mm)

10

Tblender = 1.5 min

DR = 20%
0
0 0.1 0.2 0.3 0.4 0.5
D (mm)

Before experiment
After experiment
(c) Fiber size (blender = 1.5 min, DR = 20%)

Figure 14: Effect of fiber size on DR.

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016

8 Advances in Mechanical Engineering

Conflict of Interests
The authors declare that there is no conflict of interests
regarding the publication of this paper.

References
[1] B. A. Toms, Some observations on the flow of linear polymer
solutions through straight tubes at large Reynolds numbers, in
Proceedings of the 1st International Congress on Rheology, vol. 2,
pp. 135141, North Holland, The Netherlands, 1949.
[2] K. Gasljevic and E. F. Matthys, Field test of a drag-reducing
surfactant additive in a hydronic cooling system, in Proceedings
of the ASME Fluids Engineering Division Summer Meeting
(FEDSM 96), vol. 2, pp. 249260, 1996.
[3] J. L. Lumley, Drag reduction by additives, Annual Review of
Fluid Mechanics, vol. 1, pp. 367384, 1969.
[4] P. S. Virk, Drag reduction fundamentals, AIChE Journal, vol.
21, no. 4, pp. 625656, 1975.
[5] J. J. Wei, Y. Kawaguchi, F. C. Li et al., Reduction and turbulence
characteristics in sub-zero temperature range of cationic and
zwitterionic surfactants in EG/water solvent, Journal of Turbu-
lence, vol. 10, article N10, 2009.
[6] F.-C. Li, Y. Kawaguchi, K. Hishida, and M. Oshima, Inves-
tigation of turbulence structures in a drag-reduced turbulent
channel flow with surfactant additive by stereoscopic particle
image velocimetry, Experiments in Fluids, vol. 40, no. 2, pp.
218230, 2006.
[7] J. L. Zakin, B. Lu, and H. Bewersdorff, Surfactant drag
reduction, Reviews in Chemical Engineering, vol. 14, no. 4-5, pp.
253320, 1998.
[8] P. R. Kenis, Drag reduction by bacterial metabolites, Nature,
vol. 217, no. 5132, pp. 940942, 1968.
[9] M. W. Rosen and N. E. Cornford, Fluid friction of fish slimes,
Nature, vol. 234, no. 5323, pp. 4951, 1971.
[10] J. W. Hoyt, Drag reduction in polysaccharide solutions, Trends
in Biotechnology, vol. 3, no. 1, pp. 1720, 1985.
[11] G. R. Petersen, G. A. Nelson, C. A. Cathey, and G. G. Fuller,
Rheologically interesting polysaccharides from yeasts, Applied
Biochemistry and Biotechnology, vol. 20-21, no. 1, pp. 845867,
1989.
[12] S. Ogata, T. Numakawa, and T. Kubo, Drag reduction of bacte-
rial cellulose suspensions, Advances in Mechanical Engineering,
vol. 2011, Article ID 528373, 6 pages, 2011.
[13] P. S. Virk and D. L. Wagger, Aspects of mechanisms in type B
drag reduction, in Structure of Turbulence and Drag Reduction,
pp. 201213, International Union of Theoretical and Applied
Mechanics, 1990.

Downloaded from ade.sagepub.com at University of Indonesia on May 2, 2016