Vous êtes sur la page 1sur 12

Bioprocess Biosyst Eng (2017) 40:439–450

DOI 10.1007/s00449-016-1712-8


Design and analysis of flow velocity distribution inside a raceway

pond using computational fluid dynamics
Ramakant Pandey1 • M. Premalatha1

Received: 24 August 2016 / Accepted: 25 November 2016 / Published online: 17 December 2016
Ó Springer-Verlag Berlin Heidelberg 2016

Abstract Open raceway ponds are widely adopted for Keywords Raceway pond  Microalgae  Mixing study 
cultivating microalgae on a large scale. Working depth of Turbulence modeling  Multiphase modeling
the raceway pond is the major component to be analysed
for increasing the volume to surface area ratio. The
working depth is limited up to 5–15 cm in conventional Introduction
ponds but in this analysis working depth of raceway pond
is considered as 25 cm. In this work, positioning of the Light intensity, distribution of light, gas injection, mixing
paddle wheel is analysed and corresponding Vertical and hydrodynamics/flow pattern are the main factors that
Mixing Index are calculated using CFD. Flow pattern along influence microalgal growth, and hence these factors are to
the length of the raceway pond, at three different paddle be understood for design and scale up [1]. Open ponds are
wheel speeds are analysed for L/W ratio of 6, 8 and 10, the oldest and simplest systems used for algal production,
respectively. Effect of clearance (C) between rotor blade of which raceway ponds have risen to be the predominant
tip and bottom surface is also analysed by taking four type studied, major issue in carrying out economic analysis
clearance conditions i.e. C = 2, 5, 10 and 15. Moving is that much of the information is not available on com-
reference frame method of Fluent is used for the modeling mercial size algal production. The available information is
of six blade paddle wheel and realizable k–e model is used limited for open ponds, and for PBR’s there is even less
for capturing turbulence characteristics. Overall objective information available for the investors [2]. One of the
of this work is to analyse the required geometry for major advantages of open ponds is that, they are easier to
maintaining a minimum flow velocity to avoid settling of construct and operate than most closed systems. But it
algae corresponding to 25 cm working depth. Geometry requires large land areas and, moreover, insufficient stir-
given in [13] is designed using ANSYS Design modular ring mechanisms causes lower mass transfer which results
and CFD results are generated using ANSYS FLUENT for in lower biomass productivity. The same growth-kinetics
the purpose of validation. Good agreement of results is modeling subroutines used in raceway pond could easily be
observed between CFD and experimental Particle image incorporated into a computational fluid dynamics internal
velocimetry results with the deviation of 7.23%. flow model of a photo-bioreactor, Therefore, a computa-
tional model can easily be developed to simulate these
systems [3].
Sunlight can penetrate only to a limited depth inside the
pond, hence the ponds are usually kept shallow. Liquid in
& M. Premalatha the raceway is mechanically translated at about
0.15–0.4 m/s, which ensures mixing of the culture medium
Ramakant Pandey by creating turbulence [4]. The required amount of light is
available up to a depth of 25–30 cm inside the tank, the
Department of Energy and Environment, National Institute of working depth of raceway pond could be increased [5].
Technology, Tiruchirappalli 620 015, India Paddle wheels are the most efficient devices used for

440 Bioprocess Biosyst Eng (2017) 40:439–450

driving liquid in open ponds, for example aquatic pro- modeled as two zones, the first zone up to 250 mm from
duction and wastewater treatment [6, 7]. Change in the the bottom filled with water and the second zone as air for
design of mixing device needs to be considered for the remaining height i.e. 350 mm. A rotor with six blades,
increasing the working depth of raceway pond, since is fixed at one end of the raceway pond is used to provide
mixing ensures uniform light intensity distribution, suffi- the fluid motion.
cient CO2 transfer and maintaining uniform pH [8]. Mixing
is necessary to prevent sedimentation of algae, at the same Meshing
time to avoid the cell attachment to the reactor wall.
Mixing ensures that all the cells are equally exposed to the The generated mesh is a combination of tetrahedral, hex-
light and nutrients and also improves proper gas exchange ahedral, wedge and pyramid shapes mesh. The modeled
between the culture medium and the air [8–10]. The paddle geometry contains approximately 9 lakhs cells, of which
wheel helps in the creation of eddies, which helps in cir- 99% accounts for hexahedral and tetrahedral cells. 95% of
culating algae from the bottom layer to the upper layer. the cells have skewness factor less than 0.5.
Vertical mixing is more pronounced near the paddle wheel
due to the blade passage and near the bends due to the Governing equations
secondary flows [11]. A 50% increase in algae growth is
reported when doubling the flow velocity in raceway from The mathematical equations are taken from [14]. The
15 to 30 cm/s [12]. Small clearance gap prevents the equation for conservation of mass can be written as:
possible back flow of the liquid, and thus paddle wheel  !
plays a role of a positive displacement pump. Efficiency of þ r  q v ¼ 0; ð1Þ
paddle wheel depends upon various factors, such as num-
ber of blades, speed, diameter of paddle wheel, clearance t is the velocity vector.
where, ~
of blades from the bottom floor. The mixing intensity is The conservation of momentum equation is given by:
usually defined by the Reynolds number (Re) in literature !
oq !
[4]. High Re gives high mixing and vice versa. High Re þ r  q v v ¼ rp þ r:ðsÞ þ q g þ F ; ð2Þ
means high free stream velocity which causes turbulence.
To focus the significance of turbulence for algae growth, where, p is the static pressure and s is the stress tensor, ~
only vertical mixing is considered as effective way of ~
and F are the gravitational body force and external body
analysis, to describe the probability of an algal cell to get
the light [13], and to avoid settling of cells at pond floor.
Vertical mixing is defined as the cyclical movement of the
Multiphase modeling
algal cells between top surface and bottom layers of the
Volume of flow multiphase model of ANSYS FLUENT is
In this work, position of paddle wheel inside raceway
employed to perform two phase modeling suitable for
pond is analysed using the concept of Vertical Mixing
raceway pond. Single phase model is used in which, an
Index. Later on three different L/W ratio of raceway pond
arbitrary k scalar in phase-l (water), is denoted by /kl . The
with four clearance values are simulated by considering
three different speeds of paddle wheel. CFD is used as transport equation (Eq. 3) inside the volume of phase-l is
analysis tool and ANSYS FLUENT is used for modeling, shown below:
meshing, simulation and post processing. Studied literature !
al ql /kl þ r al ql ul /kl  al Ckl r/kl ¼ skl ; ð3Þ
also suggested the capabilities of CFD to predict the flow ot
behaviour inside the raceway pond [1]. !
where ql /kl and ul are the volume fraction, physical den-
sity, and velocity of phase-l, respectively. Ckl is diffusion
coefficient and skl is source term, respectively, in phase-l.
Computational methodology

Geometry Turbulence modeling

The dimensions of the raceway pond considered for the The realizable k–e model is used for this study, since the
analysis is 3900 9 650 9 600 mm3. Which is a case of L/ transport equation for dissipation rate e has been derived
W = 6 (Fig. 1b) for further L/W ratios length of the pond is from an exact equation for the transport of the mean-square
increased, keeping the width constant. To analyse bottom vorticity fluctuation which makes it superior over tradi-
clearance effect, depth of pond is varied. The geometry is tional k–e model. This model has been extensively

Bioprocess Biosyst Eng (2017) 40:439–450 441

Fig. 1 Position of paddle wheel in raceway pond

validated for a wide range of flows including rotational The model constants considered for analysis are:
homogeneous shear flows, free flows, jet and mixing lay- C1e ¼ 1:44; C2 ¼ 1:9; rk ¼ 1:0; re ¼ 1:2:
ers, channel and boundary layer flows, and separated flows.
Equations 4 and 5 shows the modeled transport equations
for k and e in the realizable k–e model: Moving reference frame modeling
o o o l ok
ðqkÞ þ ðqkuj Þ ¼ lþ t þ Gk þ Gb In moving reference frame, the equations of motion are
ot oxj oxj rk oxj
modified to incorporate the additional acceleration terms
þ qe  YM þ Sk ;
which occur due to the transformation of motion from the
ð4Þ moving reference frame to the stationary frame. The
and, computational domain is divided into multiple cell zones
  with well-defined interfaces between the zones.
o o o lt oe The transformed fluid velocities from the stationary
ðqeÞ þ ðqeuj Þ ¼ lþ þ qC1 Se
ot oxj oxj re oxj frame to the rotating frame is:
e2 e !
 qC2 pffiffiffiffiffi þ C1e C3e Gb þ Se ; m ¼!
r m ! u; r ð8Þ
k þ ve k
ð5Þ where !mr is the relative velocity, i.e. the velocity viewed
from rotating frame, ! m is the absolute velocity. The whirl
 velocity, i.e. the velocity due to moving frame ! ur is given
g k pffiffiffiffiffiffiffiffiffiffiffiffi by:
C1 ¼ max 0:43; ; g¼S ; S¼ 2Sij Sij :
gþ5 e !
u ¼x
r ~  ~; r ð9Þ
In above equation Gk represents the generation of tur-
Here x ~ is the angular velocity of the rotating reference
bulence kinetic energy due to mean velocity gradient,
frame relative to the stationary (inertial) reference frame
calculated as:
and ~r is the distance of the stationary control volume from
ouj the rotating reference frame.
Gk ¼ qu0i u0 j ; ð6Þ
Gb is the generation of turbulence kinetic energy due to
buoyancy, calculated as: Results and discussion
lt oT Validation of CFD model
Gb ¼ bgi ; ð7Þ
Prt oxi
where Prt is the Prandtl number for energy and gi is the Geometry given in the experimental work [13] is used for
component of gravitational vector in the ith direction. the validation purpose. In the reference experiment, author
YM represents the contribution of the fluctuating dilatation in has used delta wings to provide vertical velocity for better
compressible turbulence to the overall dissipation rate. C2 and mixing index. Particle image velocimetry (PIV) results
C1e are constants. rk and re are the turbulent Prandtl numbers without delta wing presented in Table 5.2 [13], is used for
for k and e, respectively. Sk and Se are user defined terms. the validation purpose.

442 Bioprocess Biosyst Eng (2017) 40:439–450

Table 1 Validation of CFD results with PIV experiment indicating that there is a positive flow gradient available at
Parameter PIV (exp.) CFD
the bottom surface of the raceway. From the Fig. 2a, it
could be inferred that, flow velocity for the entire distance
Mean velocity (m/s) 0.0152 0.016 from the rotor position along the downstream of the race-
way pond is better corresponding to the conditions L/
From Table 1, there is a difference of 7.23% between W = 6, RPM = 20, and bottom clearance of 5 cm. Fig-
CFD and PIV results. It could be inferred from the results ure 2b, c indicates that, as L/W increased to 8 and 10, the
that CFD over predicts the mean flow velocity. Validated maximum velocity attained by the fluid also increased to
CFD model is used for further flow visualization purpose. 0.24 and 0.3 m/s, respectively. However, the flow velocity
attained by the fluid in both the cases is reduced at a shorter
Position of paddle wheel distance from the rotor position compared to L/W = 6.

Figure 1 shows three different positions of paddle wheel Velocity distribution pattern
which are analysed for the better mixing index. Vertical
Mixing Index (VMI) is used as a parameter to quantify the A flow velocity ranging from 0.1 to 0.3 m/s needs to be
vertical mixing behaviour in the raceway. It is defined in maintained in the pond corresponding to a liquid depth of
Eq. 10 and is used to compare the mixing relationships 25 cm to ensure homogeneity of mixing and required sun-
with respect to the position of paddle wheel in the raceway, light penetration. Therefore, the velocity contours ranging up
Wa to 0.3 m/s are considered in this study to visualize the effects
VMI =  100 of L/W ratio, bottom clearance and the rotor speed. Total
height of the modeled geometry is 600 mm and working
where ð10Þ height (water level) of 250 mm from the bottom surface.
Wa ¼ ; The velocity profiles are shown at three different planes
s for analysis purpose (Fig. 3a), Plane 1 corresponds to the
level nearest to the water surface. Plane 2 corresponds to a
where, VMI = Vertical Mixing Index, Wa = average
level of 125 mm from the water surface, and plane 3 rep-
vertical velocity at the section, Ua = average stream
resents the level nearest to the bottom surface. The position
velocity at the section, Wp = vertical velocity at each grid
of the planes 1, 2 and 3 considered is the same for all cases.
point, p = grid point, S = no. of grid points.
Raceway pond is divided into four equal quarters for the
From Table 2, in the case of single rotor at the center the
analysis of flow patterns (Fig. 3b).
VMI is 28. By changing the position to the end the VMI is
Table 3 gives the parameters and the range of parame-
increased to 39.6. Hence, it is suitable to use rotor at the
ters considered for this study. L/W is the ratio of total
end position. When two rotors are tested, at end position
length to total width of the raceway pond. The clearance
the observed VMI was only 41.2, the increase in VMI is
(C), is the difference in distance between raceway pond
not sufficient to justify the two rotors since the power input
depth from the rotor axis and the rotor radius. N is the
is doubled.
speed of paddle wheel in rpm.
Velocity profile along the downstream of raceway
Velocity profile at L/W = 6
pond for the bottom plane
Comparison of velocity contours with respect to section
Figure 2 shows the velocity profile along the downstream
planes Figure 4a shows the velocity contour available at
of the raceway pond for the bottom most plane. Rotor
three planes with respect to depth, corresponding to a bottom
position is considered as zero position on the x-axis in
clearance of 2 cm and at the three respective speed of the
graph. Because of the friction offer by the bottom surface,
rotor. When rotor speed is increased from 10 to 20 rpm, the
velocity is very less but still it is not zero at any point,
velocity variation is observed only in quarter 1 and quarter 4
of plane 3. In rest of the quarters of plane 3, the velocity
Table 2 Vertical Mixing Index with respect to paddle wheel position variation is almost negligible. Whereas the increase in rotor
Sr. no. Rotor position VMI (%) speed has an effect on velocity contour in all four quarters of
the upper planes. Hence, velocity contour at bottom most
1 Center 28
plane is not much affected due to the change in rotor speed
2 End 39.6
corresponding to a clearance of 2 cm.
3 Two rotors at end 41.2
Figure 4b shows the contours corresponding to a
VMI between 20 and 50% is considered good for proper mixing bottom clearance of 5 cm and at three respective speed

Bioprocess Biosyst Eng (2017) 40:439–450 443

(a) 0.18 C2
C5 C10 C15
(b) 0.24 L/W=8
0.16 R10 R10 R10 R10 0.22 C2 C5 C10 C15
R15 R15 R15 R15 R10 R10 R10 R10
0.14 R20 R20 R20 R20 R15 R15 R15 R15
0.18 R20 R20 R20 R20
0.12 0.16
Velocity (m/s)


Velocity (m/s)
0.06 0.08


0.00 0.00
0 1 2 3 0 1 2 3 4
Length along the downstream of raceway Length along the downstream of raceway

(c) 0.30
0.28 L/W=10
0.26 C2 C5 C10 C15
R10 R10 R10 R10
R15 R15 R15 R15
R20 R20 R20 R20
Velocity (m/s)

0 1 2 3 4 5
Length along the downstream of raceway

Fig. 2 Velocity profile along the downstream of raceway pond for the bottom plane for L/W at 6, 8, 10

Table 3 Range of parameters-L/W ratio, bottom clearance and speed

of paddle wheel
Clearance L/W = 6 L/W = 8 L/W = 10
(C) (in cm) Speed of paddle wheel (N) (rpm)

2 10 15 20 10 15 20 10 15 20
5 10 15 20 10 15 20 10 15 20
10 10 15 20 10 15 20 10 15 20
15 10 15 20 10 15 20 10 15 20

variation of velocity in all the quarters of the raceway

When bottom clearance is increased to 10 cm (Fig. 4c),
increasing rotor speed resulted in a proper velocity distri-
Fig. 3 a Section plane position for contour visualization, b partition bution except for the second quarter of the bottom most plane
naming of raceway pond for analysis purpose which was not affected even at a rotor speed of 20 rpm.
With a bottom clearance of 15 cm (Fig. 4d) it could be
of rotor. At 10 rpm velocity variation is minimal with observed, that even with the higher rotor speed, velocity
respect to planes. It could be seen from the contours, as distribution is poor in second, third and fourth quarter of
the rotor speed is increased from 10 to 20 rpm, there is a bottom most plane.

444 Bioprocess Biosyst Eng (2017) 40:439–450

Fig. 4 Comparison of velocity contours with respect to section planes at a 2 cm clearance, b 5 cm clearance, c 10 cm clearance, d 15 cm

Comparison of velocity contours between different rotor avoid settling of algal species. Therefore, it can be
speeds Figure 5 shows the velocity contours for differ- stated that bottom clearance always plays a very
ent clearance values for three different rotor speed. As we important role for the proper velocity distribution inside
can see from the contour with the increase in rotor speed the raceway pond. It is desired to have less bottom
the velocity distribution gets better. Increasing bottom clearance to give more translation velocity to fluid
clearance from 2 to 5 cm results in a proper distribution coming from rotor, but it should not be less than 5 cm,
of velocity but as we change from 5 to 10 cm or more as shown in Fig. 5 since it will effect negatively on
velocity distribution shows very poor results in all four flow distribution.
quarter of raceway pond.
With the increase in rotor speed the suction rate also Comparison of velocity contours between different clear-
increases, therefore, velocity distribution will always ance Velocity distribution with respect to bottom clear-
shows good results in fourth quarter but second and ance is presented for three different rotor speeds (Fig. 6). It
third quarters are not getting enough flow velocity to could be seen that the increase in clearance it resulted in

Bioprocess Biosyst Eng (2017) 40:439–450 445

Fig. 5 Velocity contours with

varying rotor speed of 10, 15, 20
rpm for (i) c = 2 cm, (ii) c = 5
cm, (iii) c = 10 cm, (iv) c = 15

Fig. 6 Velocity contours for

bottom clearance of 2, 5, 10 and
15 cm for (i) 10 rpm, (ii)
15 rpm, (iii) 20 rpm

better distribution for 5 cm. The velocity distribution Velocity profile at L/W = 8
becomes poor beyond the clearance value 10 cm.
Comparing contours of Fig. 6, it could be inferred that Comparison of velocity contours with respect to the section
the bottom clearance of 5–10 cm is preferred, considering planes Figure 7a shows the contours corresponding to a
clearance less than 5 or greater than 10 will result in poor bottom clearance of 2 cm. Velocity variation is observed in
velocity distribution for (L/W) = 6. plane 3 (bottom most plane) when the rotor speed is increased

446 Bioprocess Biosyst Eng (2017) 40:439–450

from 10 to 20 rpm, still, variation occurring in quarter 1 and 4 third, and half of the fourth quarter of third plane are
only. In rest of the raceway, velocity variation is almost showing poor velocity distribution.
negligible and similar to L/W = 6. For upper planes velocity With bottom clearance of 15 cm (Fig. 7d), it can be
distribution is better as compared to the bottom plane. Similar observed that even with higher rotor speed there is a poor
results were observed in Fig. 4a. In both the cases, distribution velocity distribution in second, third and fourth quarter in
at bottom plane is not much effected by rotor speed. bottom most plane. Therefore, 15 cm clearance is not
Figure 7b shows contours for bottom clearance of 5 cm. As preferred for L/W = 8 or more.
rotor speed is increased from 10 to 20 rpm, variation of velocity
is noticed in all the quarters of the raceway pond. At 10 rpm Comparison of velocity contours between different rotor
variation is negligible with respect to the planes. But variation speeds Figure 8 shows the velocity contours for different
is getting better with 15 and 20 rpm rotor speeds. Similar clearance values, respectively, each at three different rotor
results were observed in Fig. 4b, only difference observed is in speeds. It could be observed from contours that the increase
second quarter due to the increased working length of raceway in rotor speed improves the velocity distribution. Increase
pond. The results indicate that for a fixed design and dimen- in bottom clearance from 2 to 5 cm results in a better
sions of rotor, L/W needs to be optimized to reduce the dead distribution of velocity and further change from 5 to 10 cm
zone of the pond or corresponding to the required L/W ratio, the or more, velocity distribution becomes very poor, in all
dimensions of the rotor needs to be selected. four quarter of the raceway pond.
In Fig. 7c, bottom clearance is increased to 10 cm and When working with L/D = 8 or more, bottom clearance
velocity variation is observed in all the quarters of third plays a very important role to provide required vertical
plane when the rotor speed is increased to 20 rpm, except mixing and minimum translational velocity to fluid to
the second quarter. At 10 and 15 rpm of speed second, avoid settling of algal species. But going beyond certain

Fig. 7 Comparison of velocity contours with respect to section planes at a 2 cm clearance, b 5 cm clearance, c 10 cm clearance, d 15 cm

Bioprocess Biosyst Eng (2017) 40:439–450 447

Fig. 8 Velocity contours with

varying rotor speed of 10, 15,
20 rpm for (i) c = 2 cm, (ii)
c = 5 cm, (iii) c = 10 cm, (iv)
c = 15 cm

Fig. 9 Velocity contours for

bottom clearance of 2, 5, 10 and
15 cm for (i) 10 rpm, (ii)
15 rpm, (iii) 20 rpm

clearance value, the effect of mixing rotor speed becomes raceway pond is showing poor velocity distribution, except
negligible or almost zero. Clearance less than 5 cm is not first quarter Fig. 9i. From Fig. 9ii, at 5 cm clearance
preferred and going beyond 5 cm is resulting in poor velocity distribution is better than other three cases. When
velocity distribution as shown in Fig. 8 and with 15 cm rotor speed is increased to 20 rpm Fig. 9iii, 5 and 10 cm
clearance value velocity variation is almost nil in three clearance are giving better velocity distribution than the
quarters of raceway pond except first quarter. Therefore, other two cases.
from the above results it could be inferred that for L/ Therefore, it could be inferred from the above results
W = 8, clearance should be greater than 5 cm and less than that for L/W = 8, clearance should be in between 5–10 cm
10 cm and rotor speed should be maintained around and rotor speed should be around 20 rpm.
20 rpm. Going beyond this rotor speed may cause cell
damage due to excessive turbulence near rotor region. Velocity profile at L/W = 10

Comparison of velocity contours between different clear- Comparison of velocity contours with respect to the section
ance Figure 9 shows the velocity distribution for clear- planes Figure 10a shows the velocity variation for 2 cm
ance values of 2, 5, 10 and 15 for the varying rotor speed of bottom clearance. As the rotor speed is increased from
10, 15 and 20 rpm, respectively. All the quarters of 10 rpm to 20 rpm, velocity variation is observed in third

448 Bioprocess Biosyst Eng (2017) 40:439–450

Fig. 10 Comparison of velocity contours with respect to section planes at a 2 cm clearance, b 5 cm clearance, c 10 cm clearance, d 15 cm

and fourth quarter but it is negligible in second quarter. As per Fig. 10d, velocity variation is almost negligible
Increase of flow path from L/W = 6 to 10, translational in all the four quarters of raceway pond except the region
velocity given by the rotor to the fluid is not enough for the nearer to the rotor. Therefore, it could be inferred that a
fluid to travel up to the end of the pond. Therefore, second clearance of 15 cm is not suitable irrespective of high rotor
and half of third quarter will suffer cell damage because of speed.
settling of algal cells.
Figure 10b shows the velocity distribution for 5 cm Comparison of velocity contours between different rotor
bottom clearance, out of the three rotor speeds, 20 rpm speeds Figure 11 shows the comparison of velocity
gives better results, but still that is not enough as there is no variation contour for three rotor speed at different bottom
velocity variation in second quarter of the third plane. In clearance conditions. It could be inferred from the con-
addition, in the fourth quarter of the top plane at 20 rpm, tours, that the conditions of 20 rpm rotor speed and 5 cm
velocity variation is negligible. Therefore, it could be clearance is giving better results. Still, it is insufficient to
inferred from the contours, that for L/W = 10, a clearance provide minimum required translational velocity of 0.1 m/s
of 5 cm is not giving required velocity variation even at to avoid settling.
20 rpm.
Figure 10c shows very poor velocity variation in sec- Comparison of velocity contours between different clear-
ond, third and fourth quarter of the bottom plane at three ance Figure 12 shows the comparison of velocity varia-
different rotor speed conditions. Upper planes show better tion by varying the bottom clearance and rotor speed. At
velocity variation with respect to the increase in the rotor 10 rpm, the velocity variation is negligible in all the
speed. At 20 rpm fourth quarter of top plane shows poor quarters except in the first quarter irrespective of any
velocity variation, It could be inferred (Fig. 10c) that nei- clearance. At 15 rpm, a clearance value of 5 cm, shows a
ther bottom plane nor top plane gives required velocity better velocity variation compared to other clearance val-
variation for L/W = 10 at 20 rpm. ues. As the rotor speed is increased to 20 rpm, maximum

Bioprocess Biosyst Eng (2017) 40:439–450 449

Fig. 11 Velocity contours with

varying rotor speed of 10, 15,
20 rpm for (i) c = 2 cm, (ii)
c = 5 cm, (iii) c = 10 cm, (iv)
c = 15 cm

Fig. 12 Velocity contours for

bottom clearance of 2, 5, 10 and
15 cm for (i) 10 rpm, (ii)
15 rpm, (iii) 20 rpm

variation is observed in 5 cm clearance and small variation rotor could be altered to avoid the occurrence of high
in first, third, and fourth quarter of 2 cm clearance value. velocity closed to the rotor.
But none of the contours are showing minimum required
velocity variation to avoid settling. Acknowledgements We greatly acknowledge Department of Sci-
ence and Technology (DST), Government of India, New Delhi for
funding this project (DST/IS-STAC/CO2-SR-12/07).

Conclusion References
PIV experiment gives closer values with CFD simulation 1. Bitog JP, Lee IB, Lee CG, Kim KS, Hwang HS, Hong SW, Seo
with a difference of 7.23%, as compared to ADV. For the IH, Kwon KS, Mostafa E (2011) Application of computational
fluid dynamics for modelling and designing photo-bioreactors for
fixed rotor design, results indicated that the L/W ratio value
microalgae production: a review. Comput Electron Agric
for the raceway pond could be between 6 and 7, increasing 76:131–147
thereafter affects the velocity distribution in the second and 2. Richardson JW, Johnson MD, Outlaw JL (2012) Economic
third quarter of the raceway pond which results in dead comparison of open pond raceways to photo bio-reactors for
profitable production of algae for transportation fuels in the
zones. The bottom clearance should be in between 5 and
Southwest. Algal Res 1:93–100
8 cm, the rotor speed of 20 rpm provides the sufficient 3. James SC, Boriah V (2010) Modeling algae growth in an open-
mixing for L/W of 6. The distance between the wall and the channel raceway. J Comput Biol 17:895–906

450 Bioprocess Biosyst Eng (2017) 40:439–450

4. Hreiz R, Sialve B, Morchain J, Escudié R, Steyer JP, Guiraud P 9. Carvalho AP et al (2006) Microalgal reactors: a review of
(2014) Experimental and numerical investigation of hydrody- enclosed system designs and performances. Biotechnol Prog
namics in raceway reactors used for algaculture. Chem Eng J 22:1490–1506
250:230–239 10. Lou HP, Al-Dahhan MH (2004) Analyzing and modeling of
5. Pandey R, Sahu A, Vasumathi K, Premalatha M (2015) Studies photobioreactors by combining first principles of physiology and
on light intensity distribution inside an open pond photo-biore- hydrodynamics. Biotechnol Bioeng 85:382–393
actor. Bioprocess Biosyst Eng 38:1547–1557 11. Richmond A, Vonshak A (1978) Spirulina Culture in Israel. Arch
6. Moulick S, Mal BC, Bandyopadhyay S (2002) Prediction of Hydrobiol 11:274–280
aeration performance of paddle wheel aerators. Aquacult Eng 12. Richmond A, Hu Q (2013) Handbook of microalgal culture:
25:217–237 biotechnology and applied phycology, 2nd edn. Wiley, Oxford
7. Ahmad T, Boyd CE (1988) Design and performance of paddle 13. Voleti RS (2012) Experimental studies of vertical mixing in an
wheel aerators. Aquacult Eng 7:39–62 open channel raceway for algae biofuel production. All graduate
8. Kommareddy AR, Anderson GA (2005) Mechanistic modeling of theses and dissertations, paper 1307
photobioreactor system. ASAE paper no 057007, ASAE, St 14. ANSYS FLUENT Theory guide
Joseph, Michigan