Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

Contents lists available at ScienceDirect

Chemical Engineering Research and Design

journal homepage: www.elsevier.com/locate/cherd

Passive thermal regulation approach for Algofilm ©

photobioreactor through phase change

Kashif Hussain Mangi a,b , Zakaria Larbi c , Jack Legrand a , Jérémy Pruvost a ,
El-Khider Si-Ahmed a,∗
a University of Nantes, ONIRIS, CNRS, GEPEA, UMR-6144, 37 Bd de l’université BP406, 44602 Saint-Nazaire, France
b Department of Chemical Engineering, Quaid-e-Awam University of Engineering, Sciences and Technology,
Nawabshah, Pakistan
c Laboratoire de Mécanique des Fluides Théorique et Appliquée, Département d’Energétique et de Mécanique des

Fluides, Faculté de Physique, USTHB, B.P. 32, El-Alia 16111, Algeria

a r t i c l e i n f o a b s t r a c t

Article history: This study deals with the development of a self-sustained passive thermal regulation tech-
Received 27 October 2020 nique in Algofilm © solar photobioreactor. The reported experimental investigation includes,
Received in revised form 6 February the evaluation of an average equivalent condensate film thickness, in indoor operating con-
2021 ditions, as function of the flowing liquid Tl and ambient Tamb temperatures, inclination angle
Accepted 10 February 2021  as well as air flowrate. For instance, an average condensate film thickness of 2.2 ± 0.2 mm
Available online 20 February 2021 at Tl = 50 ◦ C, for liquid flow rate of 0.08 l/s and  = 4◦ in steady state conditions was recorded.
Furthermore, a correlation estimating the condensation rate as function of the flowing liquid
Keywords: and ambient temperatures, was developed. A theoretical, heat and mass transfer model was
Algofilm © Photobioreactor proposed and validated with the indoor experimental results. The numerical predictions of
Passive thermal regulation the model adapted to a simulated sunny day were in good agreement with the experimen-
Evaporation tal recordings. This validation evidenced the model applicability to real solar conditions,
Condensation where an accurate evaluation of the condensed water is required for predicting the infrared
filtration.
© 2021 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction Open raceways and closed photobioreactors are mainly

considered for microalgae production. However, the open sys-
Microalgae are unicellular microorganisms that can convert tems are applicable to few microalgae species, whereas the
sunlight into biochemical energy via the process of photo- closed systems are relevantly preferred due to controlled and
synthesis. It is a promising feedstock for bulk commodities isolated operation as well as their capability to produce more
like chemicals, food, nutraceutical, pharmaceutical industries microalgae species (Benemann, 2013).
and bio-fuels (Litchman and Klausmeier, 2001; Spolaore et al., Numerous photobioreactors have been developed to pro-
2006; Axelsson et al., 2012; Draaisma et al., 2013). Microalgae is duce various strains of microalgae and optimize biomass
becoming a prominent technology for wastewater treatment productivity and photosynthesis efficiency (Pruvost et al.,
(Derakhshan et al., 2019; Xie et al., 2019; Yang et al., 2019; Luo 2006; Goetz et al., 2011; de Jesus and Maciel Filho, 2017;
et al., 2021), and has been considered as one of the sustain- Solimeno et al., 2017; Deprá et al., 2019).
able solutions towards CO2 control and capture (Al Ketife et al., The volumetric productivity enhancement of microalgae
2016) to counter the global warming and climatic change (Zhao cultivation depends on several factors, including solar radia-
et al., 2019). tion, temperature, nutrient feeding procedures, pH regulation,

∗
Corresponding author.
E-mail address: [email protected] (E.-K. Si-Ahmed).
https://doi.org/10.1016/j.cherd.2021.02.013
0263-8762/© 2021 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
412 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

mum temperature ranges, to achieve maximum microalgae

Nomenclature productivity, for different types of microalgae were 25, 25,
40 ◦ C respectively for psychrophilic (Asterionella formosa),
Accents mesophilic and thermophilic (Chaetoceros, Anacystis nidu-
x Average quantity lans) species.
The estimation of microalgae production is quite difficult
Greek symbols
in outdoor systems (Mata et al., 2010; Béchet et al., 2014).
ˇ Correction factor, –
For instance, the temperature cannot be easily controlled at
ı Film thickness, m
outdoor scale, because algae species experience significant
 Density, kg/m3
temperature change during varied weather conditions such
as cloudy conditions, Day/Night cycle and seasonal variations
Roman symbols
(Béchet et al., 2010; Pruvost et al., 2019).
ṁ Mass flux, kg/s
Overheating of photobioreactors is an unavoidable chal-
Dv Binary diffusion coefficient, m2 /s
lenge for microalgal mass production, resulting in low
H Enthalpy, J
photosynthetic efficiency and low biomass productivity
h Heat transfer coefficient, w/m2 K
(Pruvost et al., 2017, 2019). Microalgae production requires
k Thermal conductivity, w/m K
thermal regulations in outdoor conditions (Bechet, 2014;
L Length, m
Nwoba et al., 2019; Pruvost et al., 2019). Microalgae needs
Levap Latent heat, kJ/kg
the photo-synthetically active spectrum i.e. 400–700 nm for
Nu Nusselt number, –
biomass growth (Wondraczek et al., 2015; Baer et al., 2016).
p Partial vapor pressure, Pa
Therefore a substantial portion of the solar spectrum is
Q Heat flux, J/s
unwanted for these solar PBRs which includes infrared and
Ra Rayleigh number, –
ultraviolet ranges which are the basic cause of overheat-
Re Reynold’s number, –
ing and cell-damaging on microalgae culture (Nwoba et al.,
S Surface, m2
2019). Some recent researches have witnessed that microal-
T Temperature, K
gae culture temperature can be regulated through passive
t Time, s
evaporating cooling (PEC) and glazing which can filter infrared
W Width, m
spectrum (Béchet et al., 2010; Nwoba et al., 2016). Although PEC
U Velocity, m/s
is an active thermal regulation approach but still due to cost
Y Vapor mass fraction, –
and performance limitation, the use of PEC systems to main-
Superscripts tain the optimum culture temperatures is not economically or
0 Initial conditions environmentally sustainable. However, the glazing affects the
amb Ambient transmission of the visible spectrum. Therefore, an innovative
b Bottom technology for microalgal PBR that requires less, or no, fresh-
c Critical water to filter the infrared spectrum will be a game-changer
cond Condensation (Nwoba et al., 2019).
conv Convection On the passive thermal regulation for a semi-buried open
evap Evaporation raceway, Pruvost et al. (2019) has developed a thermal model to
fc Forced convection cultivate microalgae and achieved very good agreement with
ha Humid air experimental results for temperatures. However, the simul-
i Subsystem taneous heat and mass transfer study for any open or closed
in Input flux PBR system, up to our knowledge, has not been reported yet. In
l Liquid solar photobioreactors, evaporation is often observed during
out Output flux microalgae production (Torzillo et al., 1986; Moheimani et al.,
r Residence 2011; Hindersin et al., 2013), also in Algofilm © photobioreactor
v Vapor (Goetz et al., 2011), which can be driven to develop conden-
w Condensate sate droplets or a film at the inner surface of the top cover.
w/g Glass-water The developed condensate film can absorb infrared radiations
reducing then the overheating in the Algofilm © PBR. Subse-
Acronyms quently, water (condensate) is a promising filter which can
 Inclination angle transmit the visible light, required for microalgae growth, and
PBR Photobioreactor absorb the infrared the major cause of overheating in outdoor
PEC Passive evaporative cooling PBRs (Nwoba et al., 2019). For instance, Krauter (2004) used a
thin water film over a photovoltaic panel and achieved a huge
reduction of 22 ◦ C in the panel temperature.
mixing (Pruvost et al., 2016), and gas hold up (Sabri et al., The absorption of infrared radiation is a volumetric pro-
2019). Temperature and light are important factors to control cess (Kirk, 1988), therefore, it is imperative to quantify the
the biomass and lipid production in algae (Ugwu et al., 2008; amount of condensate which will be developed in outdoor
Pruvost et al., 2012; Xu et al., 2009; Solimeno et al., 2017; Shuba conditions. The estimation of condensate film is extremely
and Kifle, 2018). Algal growth rates increases with tempera- complicated in outdoor condition, due to the involvement
ture up to an optimum range Muller-Feuga et al. (1998). The of complex phenomena such as temperature variations with
optimum light (Cheng et al., 2020) and temperature (Singh simultaneous heat and mass transfer with phase change
and Singh, 2015) were studied for the maximum growth of (evaporation and condensation), in addition to solar radiation,
different species of algae. Ras et al. (2013) reported the opti- ambient air velocity, relative humidity and the day/night cycle
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 413

(Apel and Weuster-Botz, 2015; Apel et al., 2017; Huesemann

et al., 2017). Therefore, an optimized study cannot be
carried out to establish a robust basis for modelling pur-
poses.
Understanding and quantifying the phase change phe-
nomena at controlled indoor conditions will constitute the
basis for the estimation of condensate in outdoor conditions.
For that purpose, this paper aims to build up a theoretical and
experimental study which can be further used for outdoor
conditions. The experimental study is conducted on indoor
lab-scale Algofilm © prototype with the constant operating
parameters to investigate the influence of flowing liquid tem-
perature, air injection, and the inclination angle of the cover
on the hourly condensation mass flux. A theoretical study was
also carried out by considering the global simultaneous heat
and mass transfer model with phase change.

2. Experimental setup and model

considerations

2.1. Experimental setup

The lab-scale Algofilm © photobioreactor of the GEPEA lab-

oratory, as shown in Fig. 1a, has a 0.33 m2 surface area and
it is mainly made up of stainless steel and aluminum with a
transparent cover of Plexiglas. This prototype has an ability
to be inclined up to 65◦ , which made possible the collection
of condensed mass that could be compared with the numer-
ical predictions. The Algofilm © PBR was divided into four
subsystem including bottom plate (i.e stainless steel), flow-
ing liquid on bottom plate, condensate as well as the top
cover plate (i.e. plexiglas), and humid air in between flowing
liquid and condensate film as shown in (see Fig. 1a). All exper-
iments were conducted in fully controlled indoor conditions
such as regulated room temperature, the flowing liquid tem-
perature was maintained constant through the external water
heater, and was continuously re-circulated, at a constant vol-
ume flow rate, through an integrated pump. Liquid film was
injected uniformly through a flow stabilizer at the inlet point
(see Fig. 1c), which is quite effective to make a smoother and
uniform flow of liquid on the bottom of Algofilm © PBR. The
size and distribution of holes were already optimized on the
same experimental setup in previous studies by Le Borgne
(2014), Pruvost et al. (2017).

2.2. Experimental data acquisition

Indoor experimental setup is mainly based on the precise data

acquisition regarding mass quantification and temperature
distribution in every subsystem of Algofilm © PBR. A dedi-
cated LabVIEW program was established to record the data
(i.e. temperature, airflow rate etc) of each subsystem (glass-
condensate film, humid air medium (inside air), flowing liquid Fig. 1 – (a) Schematic description of indoor AlgoFilm ©. (b)
and the bottom) as shown in Fig. 1b. The thermocouples for Experimental setup for AlgoFilm © indoor prototype. (c)
each sub-system are uniformly distributed on the centre in Flow for uniform film distribution and (d) Distribution of
every zone (i.e., top, middle, and lower zones) see Fig. 1c the thermocouples on sub-system “Flowing liquid”.
(showing the thermocouples distribution for the flowing liquid
sub-system). The horizontal surface of exchange is 30 × 100
cm, i.e. 0.3 m2 .
414 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

3.1. Heat balance equations

It should be observed that the momentum Equations are not

included since it is assumed a constant mass flow rate and
the coupling with the heat balance is achieved through the
residence time.

(3) Flowing Liquid:

The liquid film thickness ıl can be considered as constant
over the length L of the PBR (smooth interface). The local
Fig. 2 – Graphical description of indoor model. energy Equation for a fully developped flow with an aver-
age velocity Ul , is expressed as follows:

∂Tl ∂T ∂2 T
The interest of the indoor setup was to measure accurately l cpl + l cpl Ul  l = kl 2l 0 < x < L, 0 < y < ıl
∂t ∂x ∂y
the condensate mass flux. After reaching steady state the con-
(3)
densate mass was collected every hour, and every experiment
was at least triplicated on same operating conditions (Incli-
nation angle, liquid flow rate and temperature) and the mean
where kl is the thermal conductivity of the liquid, Tl (x, y,
data is considered for the averaged experimental condensa-
t) is the temperature field of the liquid film. The transport
tion rate for each experiment.
mechanism with respect to the direction x is assumed to
be purely advective, and purely diffusive in the y direction.
3. Theoretical modeling Eq. (3) is subjected to the following initial and boundary
conditions:
The mathematical model was developed for the prediction ⎧
of heat and mass transfers involved in the passive thermal ⎪ Tl (x, y, t = 0) = Tl,0
⎪
⎪
regulation technique of the Algofilm © photobioreactor. ⎪
⎪
⎪
⎪ Tl (x = 0, y, t) = Tl,i
All heat and mass fluxes are briefly illustrated in Fig. 2. ⎪
⎪ 
⎪
⎨ ∂Tl 
The following assumptions hold: kl  = qconv,l−b
∂y y=0 (4)
⎪
⎪ 
⎪
⎪ ∂T 
⎪
⎪ kl ∂yl  = −qconv,l−ha − qevap
(i) The film flow is assumed to be at a constant velocity. The ⎪
⎪
⎪
⎪ y=ıl
residence time of liquid is very short with respect to char- ⎩
acteristic evaporation time, therefore the variation in the
flowing film thickness is assumed to be negligible.
(ii) The temperature of the film is assumed to be constant S = L × W is the surface of exchange. Tl,i is the inlet temper-
throughout the process. ature. qconv,l−b and qconv,l−ha are density of convective heat
(iii) The overall system of the photobioreactor is subdivided fluxes (W/m2 ) between bottom-liquid, and liquid-humid
into subsystems (bottom, liquid, humid air, condensate, air, respectively, and qevap is the evaporation heat flux den-
glass), where each subsystem is characterized by its aver- sity.
aged temperature Tl , Tb , Tha , Tw/g . It was assumed that the The integration of Eq. (3) over the liquid film thickness
condensate and the glass are thermally thin body and leads to:
characterized by the temperature Tw/g  ıl  ıl
∂ ∂Tl
(iv) The heat and mass conservation in each subsystem is l cpl Tl dy + l cpl Ul  dy
∂t ∂x
assumed to be modeled by global balances, based on the y=0 y=0

variation in enthalpy of each subsystem. Therefore, the  

∂Tl  ∂T 
spatial average values were considered that is: = kl  − ∂yl 
∂y y=ı
(5)
y=0
l

∂Hi  ∂Hi   
+ Ui  = Qin − Qout (1)
∂t ∂x The averaged temperature over the thickness of the liquid
film reads as follows:
where Hi  = Hi,ref + mi cpi Ti denotes the enthalpy of the sub-  ıl
system i = b, l, ha, g/w. mi , cpi were respectively the mass 1
Tl  = Tl (x, y, t) dy (6)
and specific heat capacity of the subsystem i. Qin , Qout ıl 0
refers to input and output heat fluxes, respectively.Ui  is
the averaged velocity over thickness of each participating The substitution of the boundary equations (4) and the
subsystem. averaged temperature over the thickness (6) into Eq. (5)
gives:
 ıi  ıi
1 1 ∂Tl  ∂Tl 
Ui  = Ui (y, t)dy and Hi  = Hi (x, y, t)dy l cpl ıl + l cpl ıl Ul  = −qconv,l−b − qconv,l−ha − qevap
ıi 0
ıi 0 ∂t ∂x
(2) (7)
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 415

The integration of Eq. (5) over the length L gives the fol- transports them to the condensate film. The global heat
lowing equation: balance Eq. (1) in the case of humid air read as:

 L dHha  dHha 
d + Uha  = Qevap
l ıl cpl Tl dx + l ıl cpl Ul  ( Tl |x=L − Tl |x=0 ) dt dx
dt x=0
 + Qconv,l−ha − Qconv,ha−w − Qcond (15)
L
= (−qconv,l−b − qconv,l−ha − qevap ) dx (8)
x=0
where,
• <Uha > is the average velocity of the injection air (see
The spatial average temperature of the flowing liquid, and
Fig. 1a).
the averaged heat fluxes over the surface of exchange are
• Hha  = Hha,ref + mha cpha Tha is the enthalpy of the humid
defined as follows:
air.
 L  ıl • Qevap is the evaporation heat flux.
1
Tl = Tl dxdy (9) • Qconv,l−ha is the convective heat flux between liquid and
L ıl x=0 y=0
humid air.
 L • Qconv,ha−w is the convective heat flux between humid air
Qi = W qi dx (10) and condensate film.
x=0 • Qcond is the condensation heat flux.
One should note that some experiments were conducted
where i = conv, l − b;conv, l − ha;evap. The substitution of the
without air injection which leads to reduction of Eq. (15)
averaged temperature and fluxes ((9, 10)) in Eq. (8) give the
to:
global energy equation of the flowing liquid.
dHha 
= Qevap + Qconv,l−ha − Qconv,ha−w − Qcond (16)
dT dt
ml cpl l + ṁl cpl ( Tl |x=L − Tl |x=0 )
dt
(5) Glass-condensate: The condensate film development is
= (−Qconv,l−b − Qconv,l−ha − Qevap ) (11)
mainly a function of vapour transported by humid air and
condensed at the undersurface of the glass cover. The
heat balance for condensate film is a function of heat and
where ṁl = l ıl WUl  is the mass flowrate of the liquid and
vapour transported from humid air to the glass and heat
ṁl = l ıl WL.
evacuated from the glass to the ambient air. In addition,
The characteristic time for heat exchange between the
an assumption of the thermally thin body was considered
liquid and participating sub-systems i.e. (humid air and
for the glass and condensate due to the small thickness
bottom) is defined as follows:
of condensate film, accordingly, the global heat balance
ml cpl equation for this subsystem is written as follows:
tc = (12)
S(hfc + hcm + hevap )
dHw/g 
= Qcond + Qconv,ha−w − Qconv,w−amb (17)
The residence time can be defined as function of the aver- dt
age velocity Ul  and the length of PBR as follows:
where,
L • Hw/g  = Hw/g,ref + (mw cpw + mg cpg )Tw/g is the enthalpy
tr = (13) of the glass-condensate.
Ul 
• Qconv,ha−w is the convective heat flux between
When the residence time is smaller than the heat transfer condensate-glass and ambient air.
characteristic time tr < tc , the heat flow is approximately (6) Bottom: In the case of indoor Algofilm © bottom is exposed
equal to the heat exchanged between the liquid and to the ambient atmosphere without having any insulation,
the participating subsystems ṁl cpl ( Tl |x=L − Tl |x=0 ) ≈ therefore, it receives heat from flowing liquid in the form
−Qconv,l−b − Qconv,l−ha − Qevap . In such case the accumula- of forced convection and transfer heat to ambient atmo-
tion of the averaged sensible heat over the residence time sphere in the form of natural convection. The global heat
is negligible, which allow to write dT l /dt ≈ 0. Such approxi- balance equation (1) for the bottom will be rearranged as:
mation was experimentally verified (see Fig. 7b), since the
liquid film residence time was short and the theoretical dHb 
= Qconv,l−b − Qconv,b−amb (18)
endorsement of tr < tc , the global modeling approach can dt
be comfortably applied in the lab-scale Algofilm © . There-
where
fore the Equation (11) can be re-written as:
• Hb  = Hb,ref + mb cpb Tb is the enthalpy of the bottom.
• Qconv,l−b is the forced convective heat flux between liquid
dT l
=0 (14) and bottom.
dt
• Qconv,b−amb is the convective heat flux between bottom
(4) Humid air: The overall heat and mass transfer of the and ambient air.
system are highly dependent on humid air as it is the
main transporting medium between the flowing liquid and 3.2. Mass balance equation
condensate film. The development of condensate film is
highly dependent on this subsystem as it receives con- The mass transfer is mainly happening due to evaporation,
vective heat flux and mass from the flowing liquid and and condensation. It is assumed saturated humid air and all
416 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

the vapour produced at the flowing liquid interface is con- where kl , L is the thermal conductivity of liquid and the length
densed at the undersurface of the glass cover. The mass of the Algofilm © PBR, and Nu is the Nusselt number expressed
balance equation expressing the variation of mw reads as fol- as function of Re and Pr as follows Welty et al. (2014):
lows:
1 1
  Nu = 0.664Re 2 Pr 3 (27)
ṁha  − ṁha  = ṁevap − ṁw (19)
out in
The amount of heat evacuated through external surfaces
where ṁevap is the evaporation rate, ṁha is the mass flow rate of Algofilm © PBR read as follows:
of the injected air. Without air injection Eq. (19) becomes:
Qconv,b−amb = S hb−amb (Tb − Tamb ) (28)
ṁw = ṁevap (20)
Qconv,w−amb = S hw/g−amb (Tw/g − Tamb ) (29)
3.3. Heat and mass transfer fluxes
where hb−amb , hw/g−amb are the heat transfer coefficients for
The heat and mass transfer model was developed by consid- the heat flow in upward and downward direction.
ering all significant transport fluxes including convective and
latent heat transfers. The heat transfer in Algofilm © photo- Nub kair
hb−amb = (30)
bioreactor can be divided into two main categories internal Lc
and external heat transfers as shown in Fig. 1a. Internal heat
transfer includes latent heat transfer (i.e. evaporation and con- Nuw/g kair
hw/g−amb = (31)
densation), and free/forced convection heat transfer between Lc
flowing liquid to humid air and bottom surface, and humid air
to glass cover of Algofilm © . However, the external heat trans- where kair is the thermal conductivity of ambient air,Lc is
fer is considered for the heat evacuated from the top cover the characteristic length of the heat exchanging surface and
and bottom surface to the ambient atmosphere by natural Nub , Nuw/g are the Nusselt numbers for natural convection for
convection. downward and upward heat flow respectively, which are the
given as a function of Rayleigh number Ra.
3.3.1. Convective heat transfer fluxes
1
The convective heat fluxes are described by Newton’s law of Nub = 0.27Ra 4 (32)
cooling. The internal heat transfer coefficients are dependent
1
on temperature and partial pressure of the vapour, due to the Nuw/g = 0.54Ra 4 (33)
presence of simultaneous heat and mass transfer. However,
the external heat transfer coefficients do not involve any mass 3.3.2. Latent heat transfer fluxes
transfer. The internal heat transfer fluxes are given by: The latent heat fluxes of evaporation and condensation are
given by the analogical expression of Newton’s law of cooling
Qconv,l−ha = Shcm (Tl − Tha ) (21) but in terms of partial vapour pressures instead of tem-
peratures of the participating subsystems. These fluxes are
Qconv,ha−w = Shcha (Tha − Tw ) (22) expressed as:

where S is the surface of the Algofilm © PBR and hcm , hcha are Qevap = Levap ṁevap (34)
the heat transfer coefficients (from flowing liquid to humid
air, humid air to condensate) and are theoretically derived by Qcond = Levap ṁcond (35)
a heat and mass transfer analogy by Dunkle (1961), applied
on solar photobioreactor Artu (2016) to quantify the free con- where Levap is the latent heat of evaporation, and ṁevap , ṁcond
vective heat transfer in closed geometries. These coefficients are the evaporation and condensation mass fluxes. In this case
are: the humid air was assumed to be saturated with vapour, allow-
1
ing then to write ṁevap = ṁcond and Qevap = Qcond Yu and Wang
pv,l − pv,ha 3
(2012).
hcm = 0.884 (Tl − Tw ) + Tl (23)
2016 − pv,l
3.3.3. Mass transfer fluxes
1
The evaporation mass flux ṁevap is computed from:
pv,ha − pv,w 3
hcha = 0.884 (Tl − Tw ) + Tl (24)
2016 − pv,l
ṁevap = S hevap (pv,l − pv,ha ) (36)

The heat transferred between the flowing liquid and the

where pv,i is the partial vapour pressure of subsystem i =
bottom is given by:
l, w/g, and the mass transfer coefficient hevap is obtained from
the heat transfer coefficient hcm (Eq. (23)) with a correlation
Qconv,l−b = S hfc (Tl − Tb ) (25)
factor of 9.15 × 10−7 , and is theoretically derived by a heat and
mass transfer analogy by Dunkle (1961), applied on solar pho-
where hfc is the forced convective heat transfer coefficient and
tobioreactor by Artu (2016) to quantify the evaporative heat
expressed as:
transfer as follows:
Nukl
hfc = (26) hevap = 9.15 × 10−7 hcm (37)
L
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 417

3.4. Numerical solution for heat and mass transfer

model equations

The substitution of all the above defined heat and mass fluxes
and enthalpy expressions in equations ((14, 16),(17, 18),(20))
lead to system of five ODEs to be solved simultaneously as
given in the following for the case of zero injected air flow:

⎧
⎪
⎪
dT l
⎪ dt = 0
⎪
⎪
⎪
⎪
⎪
⎪
⎪
dT
mair cpair ha = Qevap + Qconv,l−ha − Qconv,ha−w − Qcond
⎪
⎪ dt
⎪
⎨
dT w/g
(m cp + m cp ) =Q +Q − Qconv,w−amb
⎪
⎪
w w g g
dt cond conv,ha−w
⎪
⎪
⎪
⎪ dT
⎪ mb cpb b = Qconv,l−b − Qconv,b−amb
⎪
⎪
⎪ dt
⎪
⎪
⎪
⎩ dmw = S h (p − p )
evap v,l v,ha
dt
(38)

Fig. 3 – Effect of inclination angle on condensation rate.

with the following initial conditions:

⎧
⎪ Tl (t = 0) = Tl,0
⎪
⎪
⎪
⎪
⎪
⎪ T (t = 0) = Tha,0
⎪ ha
⎪
⎪
⎨ Tw/g (t = 0) = Tw/g,0
(39)
⎪
⎪ Tb (t = 0) = Tb,0
⎪
⎪
⎪
⎪
⎪
⎪ mw (t = 0) = 0
⎪
⎪
⎩

The solution of heat and mass transfer model was obtained

by integration with respect to time of the system of ordinary
differential equations (38) taking into account the initial con-
ditions (39). The integration of these ODEs was carried out by
the MATLAB predefined function.

4. Results and discussions

The experimental and theoretical results are presented with

different environmental and controlled operating conditions. Fig. 4 – Effect of flowing liquid temperature on
The experimental results were also compared with the theo- condensation rate for  = 35◦ .
retical model in terms of condensed mass at the undersurface which drives condensate to flow into the condensate collector
of the glass cover, and the temperatures of each subsystem. (Goshayeshi and Safaei, 2019).
The influence of the flowing liquid temperature on the con-
4.1. Experimental results densation rate is depicted in Fig. 4. The results showed that the
condensation rate increased with the increasing flowing liq-
The experiments were conducted by varying the inclination uid temperature, where a large increase was observed from
angle, flowing liquid temperature, and the air injection to 9.5 g/h to 49.5 g/h on 35 and 50 ◦ C respectively. This effect
evaluate the effect on hourly condensation rate, and the tem- can be explained by the increase of transport of the vapour
peratures of each subsystem. The effect of inclination angle is through buoyancy force which highly depends on the tem-
a very important factor (Goshayeshi and Safaei, 2019) not only perature gradient between flowing liquid and the glass cover.
to see its impact but also to quantify the maximum experi- Moreover, the increase of the temperature of the flowing liquid
mental condensation rate. Fig. 3 shows the results for hourly also increased the partial pressure of the vapour at the inter-
condensation rate on a wide range of inclination angles 4–65 face, which results in the increase in evaporation as well as the
◦ , on fixed parameters such as liquid temperature (35, 40 and condensation rate. This can help to correlate the indoor study
50 ◦ C) and flow rate of 0.08 l/s in the Algofilm © PBR. with the outdoor atmosphere while dealing with the relevant
The experimental results showed that the collection rate is temperature range.
increased with increasing  up to a critical value  = 35◦ ), where Fig. 5 shows the photos taken from the top of the Algofilm
the collection rate reached the asymptotic value and revealed © PBR for inclination angle of 6◦ at the temperature of 50 ◦ C of
the experimental quantification of condensation rate. This the flowing liquid. These photos show the existence of poly-
fact was caused due to the increased gravitational force disperse and connected droplets for all the cases considered.
418 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

Fig. 5 – Development of irregular condensate film on Tl = 50 ◦ C and  = 6◦ .

This observation made it possible to consider an irregular film

of condensate at the undersurface of the glass cover.
The existence of irregular film made possible the estima-
tion of an averaged condensate film on different inclination
angles  as shown in Fig. 5, calculated from the data of
condensation rate experiments on different . The averaged
condensate film thickness was calculated as follows:

ṁw td1
ıcond = (40)
l S

where ıcond in the averaged film thickness, ṁw is the max-

imum condensation rate (maximum condensation rate was
collected by varying the inclination angles up to 65◦ , however
after 35◦ the experimental condensation rate was constant,
which allowed to assume that the condensation rate achieved
at the critical inclination angle 35◦ is equivalent to the maxi-
mum experimental condensation rate and independent of ),
td1 is the time taken by the first droplet to slide into the con-
densate collector, and S is the glass surface area. Furthermore,
to estimate the averaged condensate film thickness the time Fig. 6 – Estimation of averaged condensate film thickness
between the initial time t0 and the time when the first droplet over the range of inclination angles .
to start slide td1 was recorded, and the condensation rate was
taken into account for the observed duration to compute the
amount of condensate stocked at the under the surface of the mation can be useful to estimate the filtering of infrared
top cover. The computed volume of condensate per surface radiations.
area of glass cover gave the approximate averaged condensate
film thickness. 4.2. Experimental validation of the theoretical
Fig. 6 represents the averaged condensate film thickness assumptions
estimated on the range of  =4–35◦ . The maximum film thick-
ness 2.2 mm was achieved on lowest  of 4◦ , however, the film The major assumptions made in the theoretical modeling
thickness remained minimum 0.1 mm on higher  of 35◦ . The include (i) natural convection in the humid air and (ii) uni-
averaged condensate film thickness is highly dependent on form temperature within the subsystems. Accordingly, these
inclination angle . If inclination angle  is higher the conden- assumptions were endorsed by the experimental results dis-
sate will continue to slide down into the condensate collector cussed hereafter.
and less amount of condensate will remain on the undersur- It is well known that the influence of air injection in PBR
face of glass cover which leads the reduction in condensate is important for the microalgae as they require CO2 injection
film thickness. Nevertheless, when  is lower as 4◦ the conden- for pH regulation and growth. From the point of view of heat
sate could not slide into the collector and remained at the glass and mass transfer, the air flow rate can influence the transport
cover due to the least gravitational effect which results a sig- mechanism in the humid air, which can impact the conden-
nificant condensate thickness as presented in Fig. 6, however sation rate. To understand the impact of air injection on the
sometimes droplets coalescence was observed due to short transport behavior, experiments with different flow rates (i.e.
sliding of droplets, but that coalescence could not drive the 2.5–7.5 l/min) were conducted, while considering the airflow
condensate droplets into the collector. Such results showed rate range of outdoor Algofilm © PBR i.e. 2–4 l/min. The vapour
the importance of the indoor experiments, which made pos- can be transported by air outside by forced convection reduc-
sible, to estimate this condensate film thickness which was ing the amount of condensate on the undersurface glass cover.
very difficult to obtain in outdoor conditions, where this infor- Fig. 7a represents the effect on condensation rate for different
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 419

Fig. 7 – Experimental validation of theoretical assumptions: (a) Effect of air injection on condensation rate; (b) Temperature
gradient of each subsystem for Tl = 50 ◦ C (without air injection).

airflow rates form 2.5 to 7.5 l/min, a case without air injection behaviour in the first hour before reaching steady-state. The
was included as base of comparison. simulated results were in a good agreement with the experi-
Fig. 7a showed that natural convection dominated in mental results for all the investigated temperatures. Whereas
Algofilm © PBR, therefore no any significant impact on the con- the maximum deviation at the steady-state was observed
densation rate was observed. Furthermore, it was also shown at most 1 ◦ C. However, considering the complexities of the
that the Richardson Number Ri for the air flow rates con- simultaneous heat and mass transfer with phase change phe-
sidered (2.5, 5, 7.5 l/min), had respectively the values (8440, nomena and the global model this deviation is acceptable.
2110, 940), showing that the natural convection was the main Furthermore, from the point of view of microalgae growth and
heat transfer process. These results endorsed the assumption temperature modeling studies this deviation will not have a
considered in the theoretical modeling that the dominated significant effect (Pruvost et al., 2019).
transport mechanism in humid air is natural convection. Fig. 10a represents the evolution of experimental and
Regarding the assumption of uniform temperature within numerical condensed mass for different temperature of flow-
each subsystem was also verified by measuring the temper- ing liquid i.e (Tl = 35, 40, 50 ◦ C). The evolution of mw over the
ature at entrance/upper zone and exit/ lower zone of the time exhibited linear behaviour after achieving steady-state
Algofilm © (see Fig. 1a), the comparison for each subsystem for all the temperatures of flowing liquid considered. The com-
was depicted in Fig. 7b. parison of results is in acceptable range with a maximum
The results on Fig. 7b show that the gradient between the deviation of 2 g/h for 50 ◦ C, but for lower temperatures which
upper and lower zone is negligible. These results endorsed represent the actual range for microalgae production about
the global modeling approach for predicating heat and mass 20–30 ◦ C, the modeling results are in very good agreement with
transfer in Algofilm © PBR. The assumption of uniform tem- a maximum deviation 0.8 g/h.
perature within each subsystem was also discussed and The linear behaviour of condensed mass over time allowed
validated experimentally by (Goetz et al., 2011). However in us to calculate the averaged condensation rate with respect
the large-scale outdoor PBR the temperature gradient may not to the liquid temperature. Fig. 10b represents the comparison
be negligible, in such situation the local modeling approach is of experimental and simulation results of averaged condensa-
recommended. In the case of lab-scale Algofilm PBR the length tion rate for all the investigated temperatures. The mass flux
and the height of flowing liquid are small enough to acquire is seeming to be a highly dependent on the flowing liquid tem-
the temperature gradient within the flowing liquid film as it perature as it increased from 10 g/h to 49.5 g/h by varying the
is controlled through and external water heater, and the resi- temperature from 35 to 50 ◦ C.
dence time is very short. Fig. 9 shows the evolution of heat fluxes i.e. evaporation
and condensation over the range of studied flowing liquid
4.3. Theoretical results and comparison with temperature (35, 40, and 50 ◦ C) with respect to time. The
experiments humid air between flowing liquid (evaporating surface) and
top cover (condensing surface) was assumed to be instanta-
The validation of the simulated results consists of the compar- neously saturated with vapour in this case, therefore all the
ison with the experimental results for temperature, averaged vapour produced at the liquid/air interface will be condensed
mass transfer flux and the condensed mass. The comparison at under surface of the top cover, since the PBR is closed.
was made for the temperature of humid air, condensate/glass, This assumption was justified for a similar case through lat-
and the bottom of Algofilm © PBR. tice Boltzmann simulations, accordingly the latent heat of
The solution of Eqs. (38) for indoor conditions allowed to evaporation and condensation are same, since mevap = mcond
predict the temperatures of each subsystem and condensate (Yu and Wang, 2012). The maximum value for evaporation
amount at the undersurface of the glass cover. The simula- and condensation heat fluxes after reaching fully developed
tions were carried out on three liquid temperatures 35, 40, and condition is about 21.9, 39.19, 89.65 w/m2 for (35, 40, 50 ◦ C),
50 ◦ C as reported in Fig. 8. For all the cases considered, the evo- respectively. The driven force for evaporation and condensa-
lution of the temperature of each subsystem showed transient tion is the temperature difference between the flowing liquid
420 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

Fig. 8 – Comparaison between simulated and experimental temperature of each subsystem for Tl = 35, 40, 50 ◦ C.

(which is controlled through water heater) and the ambient The evolution of averaged condensation rate over the range
temperature (regulated indoor temperature). Therefore, at the of liquid temperature is presented in Fig. 11 which shows the
start of simulation the magnitude of these fluxes is higher same trend of as Fig. 10b, where the increase in liquid tem-
because of the large temperature gradient as the initial con- perature results an increase in averaged condensation rate.
ditions for liquid temperature are (35, 40, 50 ◦ C), however the This figure also shows a critical temperature for which no
indoor ambient temperature is almost constant at 23 ◦ C which condensation was observed, that corresponds perfectly to the
also corresponds to the humid air temperature experimen- ambient temperature. As expected, this result seems to be
tally as well as numerically at initial conditions. The transient logical while there is no temperature gradient which is driv-
time for simulation or experiments is maximumly about one ing force for evaporation, therefore the system has achieved
hour for studied range of temperature as shown in Fig. 8, thermal equilibrium. Such ascertainment shows the impact
which caused higher values for evaporation and condensation of the ambient temperature on evaporation which was not so
fluxes for first hour but declining continuously. This decline expected even when the liquid temperature is automatically
in the heat fluxes is due to decrease in the temperature dif- regulated. This result was further studied to see the impact of
ference (between flowing liquid and humid air) which was ambient temperature on the condensation rate and some sim-
maximum at initial conditions, however after reaching fully ulations were carried out by considering the ambient indoor
developed conditions (after one hour) these fluxes became temperature variation. The results of these simulations exhib-
constant as the temperature gradient between humid air and ited that the ambient temperature can also be an influencing
flowing liquid reached at a constant value (numerically as well parameter for condensation rate, as it was significantly influ-
as experimentally). enced by these small temperature deviations i.e. (23 ± 2 ◦ C).
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 421

Fig. 9 – Evolution of evaporation and condensation latent

heat fluxes for Tl = 35, 40, 50 ◦ C.

The influence of flowing liquid and ambient temperatures can

be stretched to develop a correlation to estimate the conden-
sation rate for closed PBRs in realistic conditions. Liu et al.
(2019) showed that the mass flux of an evaporating liquid film
can be proportional to the mass flux derived from the analyt-
ical solution of Stefan flow transport Equation in binary gas
mixture, which is expressed as follows:

Sair Dv 1 − Yv,amb
ṁ = ˇ × ln (41)
ıa ir 1 − Yv,l

where:

• Yv,l , Yv,∞ are the mass fraction of vapor in liquid and gas
respectively, ˇ is the factor of proportionality, and ıair is the
thickness air gap.
• Dv is the binary diffusion coefficient, which can be
expressed in this case as a function of the flowing liq-
uid temperature by the relation of Fuller–Schettler–Giddings
(Green and Perry, 2019):

10−7 Tl1.75 1
Ma + 1
Mv
Dv =  1/3 1/3
 (42)
Pamb a + v

where Mi , i = a, v is the molecular mass for air and vapor

and i is the diffusion volumes for air and vapor molecules.
• air is the density of vapor evaluated at the liquid temper- Fig. 10 – Comparison of experimental and simulated results
ature, which can be expressed by the perfect gas law as for Tl = 35, 40, 50 ◦ C: (a) Condensed mass (b) Mass flux.
follows:

Pamb Mair
air = (43)
R̂Tamb

The influencing parameters of the evaporation flux (Liu

et al., 2019) are the temperatures of the evaporating liquid flux given by Eq. (41) showed the same trend with simulation
film and the gas. However in current study the driving force of results, while considering a constant of proportionality ˇ = 98.
evaporation is the temperature gradient between the flowing However, the constant of proportionality ˇ appeared as a func-
liquid film and ambient air, accordingly the mass flux in cur- tion of ambient air temperature when some simulations were
rent study can be expressed in same way as Eq. (41), therefore a carried out by considering the wide ange of ambient tempera-
simulation was carried out by considering ambient air temper- ture with respect to seasonal variations. Therefore, to extend
ature as 23 ◦ C and results are plotted in Fig. 12, where the mass the applicability the developed correlation ˇ was fitted as func-
422 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

can be applied to predict the condensation rate over the wide

range of ambient temperature, which will be of great interest
in outdoor conditions.

4.4. Validation of the indoor model in a simulated

sunny day

Fig. 13a and b are showing the evolution in simulated sunny

day experiment and simulation for the temperatures of differ-
ent sub-systems and the condensed mass, respectively.
In order to test the response of the developed heat
and mass transfer model on dynamic conditions, a mimic
experiment was conducted to validate the model on a repre-
sentative condition of a typical outdoor day, by varying liquid
temperature using an external water heater. Flowing liquid
temperature was increased from 35 ◦ C up to 50 ◦ C and then
decreased inversely with a time interval of one hour by con-
sidering the thermal inertia of solar radiation in a typical
sunny day. This approach of simulated sunny day with arti-
ficial solar light was also considered by Goetz et al. (2011) in
Fig. 11 – Evolution of condensation rate as a function of controlled indoor conditions for the validation of heat transfer
flowing liquid and ambient temperature. model to predict the temperatures of the different subsys-
tems.
The simulated results for temperature predictions of each
subsystem (without any fitting or adjustment) were compared
with the experimental ones and showed a good agreement
(see Fig. 13a). In addition to temperature validation, the con-
densed mass prediction was also in good agreement (Fig. 13b)
over the period of mimic outdoor day. The experimental
mass collected over the period of whole experiment was 95
g, however the deviation of 15 g was observed at the end
of experiment but that corresponds to the mass remained
at the glass cover before the sliding of the first droplet as
discussed in Fig. 6. These results showed the capability of
developed model to be applied on real outdoor solar condi-
tions.
Goetz et al. (2011) carried out some outdoor experiments
and the mean temperature of flowing liquid reached up to 55
◦ C in a typical sunny day in the month of July, but in the current

study it is experimentally and numerically revealed that evap-

oration rate at this temperature can easily reached around 65
g/h (see Fig. 11), which is not a negligible value. The condensed
water mass can serve the purpose of passive evaporating cool-
ing and filtration of infrared spectrum to avoid overheating of
Fig. 12 – Evolution average condensation rate as a function
microalgae culture with a passive approach. Indeed, Nwoba
of ambient temperature (comparison of developed
et al. (2019) have shown the influence of the infrared filtration
correlation with simulations).
on microalgae culture temperature, where passive evaporat-
ing cooling was used as a filter for infrared radiation to avoid
the overheating of microalgae and temperature reduction of 8
tion of ambient air temperature, which leads to the following
◦ C was observed when compared with and without PEC. This
expression:
fact endorsed the need of incorporation of simultaneous heat
T 2 Tamb and mass transfer model for thermal regulation of Algofilm ©
amb
ˇ = 2.011 − 5.168 + 3.335 (44) PBR.
TK TK

TK = 273.15 K is the absolute temperature. 5. Conclusion

This correlation takes into account the temperature of the
flowing liquid and the ambient air. These temperatures, in The present work was aimed to develop a simultaneous heat
outdoor conditions, are always varying with respect to the and mass transfer model to predict the condensate film thick-
weather conditions, which may allow to predict the mass ness which will absorb the Infrared spectrum (main source of
of condensed water on the undersurface of the glass cover overheating). The interest of the indoor study was to validate
of Algofilm © PBR, which is very important to quantify the the model with controlled operating parameters (not possi-
infrared filtration. Fig. 12 showed the comparison of results ble in outdoor conditions) and experimentally investigating
of simulation and Eq. (41) for average condensation rate with all the influencing parameters for evaporation and conden-
respect to Tl . The comparison showed that the correlation sation. Accordingly, a detailed experimental and theoretical
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 423

Fig. 13 – Evolution of condensed mass and temperatures of each sub-system in simulated sunny day.

study on lab-scale has been conducted. The experimental Acknowledgment

study focused on the evolution of temperature and con-
densed mass as a function of inclination angle, air injection, The Higher Education Commission, Pakistan is acknowledged
flowing liquid and ambient temperature. The results showed for the financial support to the PhD student under their project
that the temperature of flowing liquid is the most impor- HEC-UESTP, Pakistan.
tant factor on condensation rate, whereas the air injection
has not shown any significant effect for the investigated References
range.
These results revealed that air injection in outdoor condi- Al Ketife, A.M., Judd, S., Znad, H., 2016. A mathematical model for
tion will not affect the transport of heat and mass by natural carbon fixation and nutrient removal by an algal
convection. The numerical results were carried out for the photobioreactor. Chem. Eng. Sci. 153, 354–362,
prediction of temperature and condensed for the wide range http://dx.doi.org/10.1016/j.ces.2016.07.042.
of flowing liquid temperatures i.e. 30–50 ◦ C. The experimen- Apel, A., Pfaffinger, C., Basedahl, N., Mittwollen, N., Göbel, J.,
Sauter, J., Brück, T., Weuster-Botz, D., 2017. Open thin-layer
tal and numerical results were found in a good agreement
cascade reactors for saline microalgae production evaluated
for the whole range of flowing liquid temperature in steady
in a physically simulated mediterranean summer climate.
state conditions. Besides, the model has the ability to predict Algal Res. 25, 381–390,
the temperature and condensed mass on artificial dynamic http://dx.doi.org/10.1016/j.algal.2017.06.004.
condition. The good agreement between numerical and exper- Apel, A.C., Weuster-Botz, D., 2015. Engineering solutions for open
imental results made possible to develop a correlation to microalgae mass cultivation and realistic indoor simulation of
estimate the condensation rate as a function the flowing liquid outdoor environments. Bioprocess Biosyst. Eng. 38, 995–1008,
http://dx.doi.org/10.1007/s00449-015-1363-1.
and ambient temperature; the latter appeared to be important
Artu, A., 2016. Etude et optimisation de la culture de microalgues
in the evaporation process. en photobioréacteurs solaires, Ph. D. thesis. PhD Thesis.
The water is the potential filter for the infrared spectrum University of Nantes.
(main source of overheating the microalgae). Active cooling Axelsson, L., Franzén, M., Ostwald, M., Berndes, G., Lakshmi, G.,
systems such as passive evaporating cooling system is not Ravindranath, N.H., 2012. Perspective: Jatropha cultivation in
viable due to the water constraints, therefore the developed southern India: Assessing farmers’ experiences. Biofuels
passive thermal regulation approach will serve both purposes: Bioprod Biorefining 6, 246–256,
http://dx.doi.org/10.1002/bbb.1324.
(i) do not need any fresh water to filter infrared spectrum as
Baer, S., Heining, M., Schwerna, P., Buchholz, R., Hübner, H., 2016.
reported by Krauter (2004), Nwoba et al. (2019), Ghosal et al. Optimization of spectral light quality for growth and product
(2020); (ii) condensate film development will be a function formation in different microalgae using a continuous
of solar irradiation intensity (condensation would be maxi- photobioreactor. Algal Res. 14, 109–115,
mum at the peak hours of the day which will filter infrared http://dx.doi.org/10.1016/j.algal.2016.01.011.
radiations since the present indoor study showed that the Bechet, Q., 2014. Modelling the impact of temperature on
microalgae productivity during outdoor cultivation, Ph. D.
evaporation rate is a function of flowing liquid and ambi-
thesis. Massey University.
ent temperatures). The development of current simultaneous
Béchet, Q., Shilton, A., Fringer, O.B., Munoz, R., Guieysse, B., 2010.
heat and mass transfer model with a good agreement on a Mechanistic modeling of broth temperature in outdoor
simulated sunny will be a prime step to apply this model photobioreactors. Environ. Sci. Technol. 44, 2197–2203,
on real solar conditions, where the estimation of condensed http://dx.doi.org/10.1021/es903214u.
mass is essential to quantify the amount of filtered infrared Béchet, Q., Shilton, A., Guieysse, B., 2014. Full-scale validation of
radiations. a model of algal productivity. Environ. Sci. Technol. 48,
13826–13833, http://dx.doi.org/10.1021/es503204e.
Benemann, J., 2013. Microalgae for biofuels and animal feeds.
Declaration of Competing Interest
Energies 6, 5869–5886, http://dx.doi.org/10.3390/en6115869.

The authors report no declarations of interest.

424 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425

Cheng, J., Lai, X., Ye, Q., Guo, W., Zhou, J., 2020. Numerical Mata, T.M., Martins, A.A., Caetano, N.S., 2010. Microalgae for
simulation on optimizing flow field and flashing-light effect in biodiesel production and other applications: a review. Renew.
jet-aerated tangential swirling-flow plate photobioreactor to Sustain. Energy Rev. 14, 217–232,
improve microalgal growth. Chem. Eng. Sci. 215, 115371, http://dx.doi.org/10.1016/j.rser.2009.07.020.
http://dx.doi.org/10.1016/j.ces.2019.115371. Moheimani, N.R., Isdepsky, A., Lisec, J., Raes, E., Borowitzka, M.A.,
Deprá, M.C., Mérida, L.G., de Menezes, C.R., Zepka, L.Q., 2011. Coccolithophorid algae culture in closed
Jacob-Lopes, E., 2019. A new hybrid photobioreactor design for photobioreactors. Biotechnol. Bioeng. 108, 2078–2087,
microalgae culture. Chem. Eng. Res. Design 144, 1–10, http://dx.doi.org/10.1002/bit.23161.
http://dx.doi.org/10.1016/j.cherd.2019.01.023. Nwoba, E.G., Ayre, J.M., Moheimani, N.R., Ubi, B.E., Ogbonna, J.C.,
Derakhshan, Z., Ehrampoush, M.H., Mahvi, A.H., Dehghani, M., 2016. Growth comparison of microalgae in tubular
Faramarzian, M., Eslami, H., 2019. A comparative study of photobioreactor and open pond for treating anaerobic
hybrid membrane photobioreactor and membrane digestion piggery effluent. Algal Res. 17, 268–276,
photobioreactor for simultaneous biological removal of http://dx.doi.org/10.1016/j.algal.2016.05.022.
atrazine and CNP from wastewater: A performance analysis Nwoba, E.G., Parlevliet, D.A., Laird, D.W., Vadiveloo, A., Alameh,
and modeling. Chem. Eng. J. 355, 428–438, K., Moheimani, N.R., 2019. Can solar control infrared blocking
http://dx.doi.org/10.1016/j.cej.2018.08.155. films be used to replace evaporative cooling for growth of
Draaisma, R.B., Wijffels, R.H., Slegers, P.M.E., Brentner, L.B., Roy, Nannochloropsis sp. in plate photobioreactors? Algal Res. 39,
A., Barbosa, M.J., 2013. Food commodities from microalgae. 101441, http://dx.doi.org/10.1016/j.algal.2019.101441.
Curr. Opin. Biotechnol. 24, 169–177, Pruvost, J., Cornet, J., Goetz, V., Legrand, J., 2012. Theoretical
http://dx.doi.org/10.1016/j.copbio.2012.09.012. investigation of biomass productivities achievable in solar
Dunkle, R., 1961. Solar water distillation: the roof type still and a rectangular photobioreactors for the cyanobacterium
multiple effect diffusion still. In: in: Proc. International Heat arthrospira platensis. Biotechnol. Progress 28, 699–714,
Transfer Conference. University of Colorado, USA, pp. 895. http://dx.doi.org/10.1002/btpr.1540.
Ghosal, M., Sethi, A., Behera, D., 2020. Performance of solar Pruvost, J., Goetz, V., Artu, A., Das, P., Jabri, H.A., 2019. Thermal
photovoltaic module through combined air and water cooling modeling and optimization of microalgal biomass production
in warm and humid climatic condition of India. MATTER: Int. J. in the harsh desert conditions of State of Qatar. Algal Res. 38,
Sci. Technol., 6, http://dx.doi.org/10.20319/mijst.2020.61.1525. 101381, http://dx.doi.org/10.1016/j.algal.2018.12.006.
Goetz, V., Borgne, F.L., Pruvost, J., Plantard, G., Legrand, J., 2011. A Pruvost, J., Le Borgne, F., Artu, A., Cornet, J.F., Legrand, J., 2016.
generic temperature model for solar photobioreactors. Chem. Industrial photobioreactors and scale-up concepts. Adv.
Eng. J. 175, 443–449, http://dx.doi.org/10.1016/j.cej.2011.09.052. Chem. Eng., Elsevier 48, 257–310,
Goshayeshi, H.R., Safaei, M.R., 2019. Effect of absorber plate http://dx.doi.org/10.1016/bs.ache.2015.11.002.
surface shape and glass cover inclination angle on the Pruvost, J., Le Borgne, F., Artu, A., Legrand, J., 2017. Development
performance of a passive solar still. Int. J. Numer. Methods of a thin-film solar photobioreactor with high biomass
Heat Fluid Flow 30, 3183–3198, volumetric productivity (AlgoFilm©) based on process
http://dx.doi.org/10.1108/HFF-01-2019-0018. intensification principles. Algal Res. 21, 120–137,
Green, D.W., Perry, R.H., 2019. Perry’s Chemical Engineers’ http://dx.doi.org/10.1016/j.algal.2016.10.012.
Handbook/edición Don W. Green y Robert H. Perry. C 660.28 Pruvost, J., Pottier, L., Legrand, J., 2006. Numerical investigation of
P47 2008. hydrodynamic and mixing conditions in a torus
Hindersin, S., Leupold, M., Kerner, M., Hanelt, D., 2013. Irradiance photobioreactor. Chem. Eng. Sci. 61, 4476–4489,
optimization of outdoor microalgal cultures using solar http://dx.doi.org/10.1016/j.ces.2006.02.027.
tracked photobioreactors. Bioprocess Biosyst. Eng. 36, Ras, M., Steyer, J.P., Bernard, O., 2013. Temperature effect on
345–355, http://dx.doi.org/10.1007/s00449-012-0790-5. microalgae: a crucial factor for outdoor production. Rev.
Huesemann, M., Dale, T., Chavis, A., Crowe, B., Twary, S., Barry, A., Environ. Sci. Biotechnol. 12, 153–164,
Valentine, D., Yoshida, R., Wigmosta, M., Cullinan, V., 2017. http://dx.doi.org/10.1007/s11157-013-9310-6.
Simulation of outdoor pond cultures using indoor led-lighted Sabri, L.S., Sultan, A.J., Al-Dahhan, M.H., 2019. Investigating the
and temperature-controlled raceway ponds and phenometrics cross-sectional gas holdup distribution in a split internal-loop
photobioreactors. Algal Res. 21, 178–190, photobioreactor during microalgae culturing using a
http://dx.doi.org/10.1016/j.algal.2016.11.016. sophisticated computed tomography (ct) technique. Chem.
de Jesus, S.S., Maciel Filho, R., 2017. Potential of algal biofuel Eng. Res. Design 149, 13–33,
production in a hybrid photobioreactor. Chem. Eng. Sci. 171, http://dx.doi.org/10.1016/j.cherd.2019.06.017.
282–292, http://dx.doi.org/10.1016/j.ces.2017.05.041. Shuba, E.S., Kifle, D., 2018. Microalgae to biofuels: ‘Promising’
Kirk, J.T., 1988. Solar heating of water bodies as influenced by alternative and renewable energy, review. Renew. Sustain.
their inherent optical properties. J. Geophys. Res.: Energy Rev. 81, 743–755,
Atmospheres 93, 10897–10908, http://dx.doi.org/10.1016/j.rser.2017.08.042.
http://dx.doi.org/10.1029/JD093iD09p10897. Singh, S., Singh, P., 2015. Effect of temperature and light on the
Krauter, S., 2004. Increased electrical yield via water flow over the growth of algae species: a review. Renewable and sustainable
front of photovoltaic panels. Solar Energy Mater. Solar Cells energy reviews 50, 431–444,
82, 131–137, http://dx.doi.org/10.1016/j.solmat.2004.01.011. http://dx.doi.org/10.1016/j.rser.2015.05.024.
Le Borgne, F., 2014. Développement d’un photobioréacteur solaire Solimeno, A., Gabriel, F., García, J., 2017. Mechanistic model for
intensifié en vue de la production à grande échelle de design, analysis, operation and control of microalgae cultures:
biomasse microalgale, Ph. D. thesis. PhD Thesis. University of calibration and application to tubular photobioreactors. Algal
Nantes. Res. 21, 236–246, http://dx.doi.org/10.1016/j.algal.2016.11.023.
Litchman, E., Klausmeier, C.A., 2001. Competition of Spolaore, P., Joannis-Cassan, C., Duran, E., Isambert, A., 2006.
phytoplankton under fluctuating light. Am. Naturalist 157, Commercial applications of microalgae. Journal of Bioscience
170–187, http://dx.doi.org/10.1086/318628. and Bioengineering 101, 87–96,
Liu, H., Jia, M., Cai, C., Zhang, Y., et al., 2019. An analytical model http://dx.doi.org/10.1263/jbb.101.87.
of the heating and evaporation of bi-component wall film. Int. Torzillo, G., Pushparaj, B., Bocci, F., Balloni, W., Materassi, R.,
Commun. Heat Mass Transfer 105, 28–36, Florenzano, G., 1986. Production of spirulina biomass in
http://dx.doi.org/10.1016/j.icheatmasstransfer.2019.03.014. closed photobioreactors. Biomass 11, 61–74,
Luo, Y., Logan, A., Henderson, R.K., Le-Clech, P., 2021. Evaluating http://dx.doi.org/10.1016/0144-4565(86)90021-1.
the resilience of photobioreactors in response to hazardous Ugwu, C.U., Aoyagi, H., Uchiyama, H., 2008. Photobioreactors for
chemicals. Chem. Eng. J. 405, 126666, mass cultivation of algae. Bioresour. Technol. 99, 4021–4028,
http://dx.doi.org/10.1016/j.cej.2020.126666. http://dx.doi.org/10.1016/j.biortech.2007.01.046.
 Chemical Engineering Research and Design 1 6 8 ( 2 0 2 1 ) 411–425 425

Welty, J., Rorrer, G.L., Foster, D.G., 2014. Fundamentals of Yang, Z., Nie, C., Hou, Q., Zhang, L., Zhang, S., Yu, Z., Pei, H., 2019.
Momentum, Heat, and Mass Transfer. John Wiley & Sons. Coupling a photosynthetic microbial fuel cell (PMFC) with
Wondraczek, L., Tyystjärvi, E., Méndez-Ramos, J., Müller, F.A., photobioreactors (PBRs) for pollutant removal and bioenergy
Zhang, Q., 2015. Shifting the Sun: solar spectral conversion recovery from anaerobically digested effluent. Chem. Eng. J.
and extrinsic sensitization in natural and artificial 359, 402–408, http://dx.doi.org/10.1016/j.cej.2018.11.136.
photosynthesis. Adv. Sci. 2, 1–13, Yu, J., Wang, H., 2012. A molecular dynamics investigation on
http://dx.doi.org/10.1002/advs.201500218. evaporation of thin liquid films. Int. J. Heat Mass Transfer 55,
Xie, B., Gong, W., Tang, X., Bai, L., Guo, Y., Wang, J., Zhao, J., Fan, 1218–1225,
Y., Li, G., Liang, H., 2019. Blending high concentration of http://dx.doi.org/10.1016/j.ijheatmasstransfer.2011.09.035.
anaerobic digestion effluent and rainwater for cost-effective Zhao, L., Zeng, G., Gu, Y., Tang, Z., Wang, G., Tang, T., Shan, Y.,
Chlorella vulgaris cultivation in the photobioreactor. Chem. Sun, Y., 2019. Nature inspired fractal tree-like photobioreactor
Eng. J. 360, 861–865, http://dx.doi.org/10.1016/j.cej.2018.12.009. via 3D printing for CO2 capture by microaglae. Chem. Eng. Sci.
Xu, L., Weathers, P.J., Xiong, X.R., Liu, C.Z., 2009. Microalgal 193, 6–14, http://dx.doi.org/10.1016/j.ces.2018.08.057.
bioreactors: Challenges and opportunities. Eng. Life Sci. 9,
178–189, http://dx.doi.org/10.1002/elsc.200800111.