See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/276454794

Determination of the external mass transfer

coefficient and influence of mixing intensity in
moving bed biofilm reactors for...

Article in Water Research May 2015

DOI: 10.1016/j.watres.2015.05.010

CITATION READS

1 212

6 authors, including:

Bruno Nogueira Mark Van Loosdrecht

Federal University of Rio de Janeiro Delft University of Technology
1 PUBLICATION 1 CITATION 1,005 PUBLICATIONS 38,952 CITATIONS

SEE PROFILE SEE PROFILE

Argimiro R. Secchi Marcia Dezotti

Federal University of Rio de Janeiro Federal University of Rio de Janeiro
178 PUBLICATIONS 661 CITATIONS 108 PUBLICATIONS 2,312 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Biofilm Reactor Modeling View project

nothing special View project

All content following this page was uploaded by Marcia Dezotti on 09 June 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8

Available online at www.sciencedirect.com

ScienceDirect

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

Determination of the external mass transfer

coefficient and influence of mixing intensity in
moving bed biofilm reactors for wastewater
treatment

Bruno L. Nogueira a,b, Julio P
erez b, Mark C.M. van Loosdrecht b,

rcia Dezotti a, Evaristo C. Biscaia Jr.a
Argimiro R. Secchi a,*, Ma
a
Programa de Engenharia Qumica, COPPE e Universidade Federal do Rio de Janeiro e Centro de Tecnologia,
~o, CEP: 21941-972 Rio de Janeiro, RJ, Brazil
Bl. G-115, Ilha do Funda
b
Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands

article info abstract

Article history: In moving bed biofilm reactors (MBBR), the removal of pollutants from wastewater is due to
Received 5 February 2015 the substrate consumption by bacteria attached on suspended carriers. As a biofilm pro-
Received in revised form cess, the substrates are transported from the bulk phase to the biofilm passing through a
25 April 2015 mass transfer resistance layer. This study proposes a methodology to determine the
Accepted 5 May 2015 external mass transfer coefficient and identify the influence of the mixing intensity on the
Available online 10 May 2015 conversion process in-situ in MBBR systems. The method allows the determination of the
external mass transfer coefficient in the reactor, which is a major advantage when
Keywords: compared to the previous methods that require mimicking hydrodynamics of the reactor
MBBR in a flow chamber or in a separate vessel. The proposed methodology was evaluated in an
External mass transfer aerobic lab-scale system operating with COD removal and nitrification. The impact of the
Mixing intensity mixing intensity on the conversion rates for ammonium and COD was tested individually.
When comparing the effect of mixing intensity on the removal rates of COD and ammo-
nium, a higher apparent external mass transfer resistance was found for ammonium. For
the used aeration intensities, the external mass transfer coefficient for ammonium
oxidation was ranging from 0.68 to 13.50 m d
1 and for COD removal 2.9 to 22.4 m d
1. The
lower coefficient range for ammonium oxidation is likely related to the location of nitrifiers
deeper in the biofilm. The measurement of external mass transfer rates in MBBR will help
in better design and evaluation of MBBR system-based technologies.
2015 Elsevier Ltd. All rights reserved.

They are applied also for the upgrade and retrofit of existing
1. Introduction treatment plants (Ferrai et al., 2010). In these systems, the
growth of microorganisms occurs on carriers which freely
Moving bed biofilm reactors (MBBR) have been used for the move inside the reactor. This movement can be achieved by
biological treatment of industrial and municipal effluents. aeration or mechanical stirring in aerobic or anaerobic/anoxic

* Corresponding author. Tel.: 55 21 3938 8307; fax: 55 21 3938 8300.

E-mail address: arge@peq.coppe.ufrj.br (A.R. Secchi).
http://dx.doi.org/10.1016/j.watres.2015.05.010
0043-1354/ 2015 Elsevier Ltd. All rights reserved.
 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8 91

The external mass transfer resistance is usually described

Nomenclature as a stagnant film between the bulk phase and the biofilm
surface where all external mass transfer processes are
Abbreviations included (Beyenal and Tanyolac, 1998). One important factor
COD Chemical Oxygen Demand which affects the external mass transfer to the biofilm is the
DO Dissolved Oxygen mixing intensity within the reactor (Kugaprasatham et al.,
EMT External Mass Transfer 1992; Chen et al., 2006). High mixing intensities increase the
MBBR Moving Bed Biofilm Reactor external mass transfer coefficient causing an increase in the
vvm Volume of Gas per Volume of Liquid per Minute mass transfer rate and an improvement on the pollutant
removal performance (Wanner et al., 2006).
Symbols
Several works (Zhang and Bishop, 1994; Stoodley et al.,
A Total superficial area (L2)
1997; Rasmussen and Lewandowski, 1998; Wa sche et al.,
D Coefficient of diffusion (L2 T
1)
2002; Boessmann et al., 2004) utilized microelectrode mea-
J Flux (M L
2 T
1)
surements for the determination of the external mass transfer
k External mass transfer coefficient (L T
1)
coefficient in different biofilm processes. The use of this
L Thickness of the biofilm (L)
methodology provides the oxygen profile along and nearby the
q Substrate specific conversion rate (M M
1 T
1)
biofilm, which makes it possible to determine the mass
r Volumetric removal rate (M L
3 T
1)
transfer coefficients (Rasmussen and Lewandowski, 1998).
R Volumetric removal rate without external mass
Microelectrode measurements on MBBR carriers require fixing
transfer resistance (M L
3 T
1)
the mobile carriers in a flow cell, which influences the
S Mass concentration (M L
3)
external mass transfer boundary and make the measure-
V Volume of the liquid phase (L3)
ments non representative.
X Biomass density (M L
3)
Despite the crescent use of MBBR, there are few studies
z Spatial variable in the biofilm (L)
concerning external mass transfer, and all of them in a nitri-
g Stoichiometric factor
fying system. Hem et al. (1994) found a near first order nitri-
d Boundary layer thickness (L)
fication kinetic for oxygen when it was the limiting substrate.
s2 Variance of the volumetric removal rate
Since the oxygen concentrations studied were above values
(M2 L
6 T
2)
for substrate half saturation constant, this could be explained
Superscripts by a strong influence of the external mass transfer in these
biof Biofilm phase systems. Gapes and Keller (2009) studied the influence of two
BL Boundary layer different carrier types (Kaldnes K1, Natrix C10/10) under two
bulk Bulk different growth conditions using a titrimetric and off-gas
exp Experimental analysis (TOGA) sensor. Differences in external mass trans-
interf Interface fer coefficient values were observed for biofilms grown under
water Water different ammonium loading rates. At higher loading rates the
more heterogeneous biofilm surface resulted in higher mass
Subscripts
transfer coefficients. Masic et al. (2010) using BiofilmChip P
Het Heterotrophic bacteria
carrier measured the oxygen profile by microelectrodes. These
Max Maximum
measurements showed the strong drop in oxygen concen-
Nit Nitrifier Bacteria
tration in the boundary layer.
In this study, a method to determine the external mass
transfer coefficient in-situ in MBBR systems is presented. The
method employs the same reactor configuration in which the
processes, respectively (Rusten et al., 2006). The carriers are biofilms are formed at normal operation, i.e., the measure-
retained inside the reactor by a sieve arrangement at the ments are performed under the same mixing conditions.
reactor outlet and, therefore, the microorganisms are kept Moreover, this methodology is also capable to evaluate the
inside the reactor favouring the retention of slow growing effect of mixing intensity on the external mass transfer
bacteria as nitrifiers (Wang et al., 2005; Rusten et al., 2006; resistance and consequently on the conversion of the sub-
Bassin et al., 2011). strates by MBBR systems. The studied system was operated
In biofilm processes, the substrates are transported from performing simultaneous nitrification and COD removal in
the bulk phase to the biofilm, where they diffuse through and order to allow comparing the effect of external mass transfer
are consumed by bacteria. The compounds diffusion in and resistance on both processes.
out of the biofilm plays an important role (Rusten et al., 2006),
being the performance of the reactor controlled by both con-
sumption rate and substrate transport (Rasmussen and 2. Material and methods
Lewandowski, 1998). Before reaching the biofilm, the pollut-
ants pass through a mass transfer resistance layer. Within the 2.1. Reactor
biofilm, the substrates are transported by diffusion due the
concentration gradient generated by the consumption of the The experiments were performed at a lab-scale moving bed
pollutants. biofilm reactor operated in continuous mode to remove COD
92 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8
and ammonium nitrogen. The carriers K1 of Anoxkaldnes
were used to perform the growth of the biofilm. These carriers
are made of polyethylene (density of 0.95 g cm
3) and have a
useful surface area of 500 m2 m
3 (or 0.00049 m2 carrier
1)
(Gapes and Keller, 2009). The volume of the reactor was 2 L and
was operating under a hydraulic retention time of 2 h with
about 815 carriers (around 40% of filling ratio).
Before the beginning of the experiments for determining
the influence of external mass transfer, the reactor was
already being operated for some months using the synthetic
medium detailed in Table 1. A short hydraulic retention time
was used to minimize the contribution of the suspended
biomass inside the reactor for the performance of the reactor.
The pH was kept around 7.5 using phosphorus buffer solution,
the temperature inside the reactor was kept at 20  C using a
heat exchanger, and the dissolved oxygen (DO) was kept
around 5 mg L
1 using a air-flow rate controller maintained in



1 L min
1. The off-line analyses of N
NH4 , N
NO2 , N
NO3
and COD were made three times per week spectrophotomet-
rically using a commercial cuvette test kit (Hach Lange, Ger-
many) to verify the efficiency of the reactor.
2.2. External mass transfer experiments
The experiments to determine the external mass transfer in-
fluence on the performance of the reactor were performed
directly inside the reactor, and the major advantage is that the
tests are made in the same hydrodynamic conditions and
reactor geometry that in the normal operation of the reactor.
In these experiments the mixing intensity inside the reactor
was increased until no effect at the removal rate of the
pollutant was observed, as can be visualized by the scheme in
Fig. 1. Since no mechanical stirrer was used in this aerobic
system, the mixing intensity was evaluated by the gas flow
rate (liter per minute) per volume of liquid in the reactor (liter).
Gas flow rate divided by the reactor section (116.5 cm2) could
be also used to express mixing intensity as gas superficial
velocity. The gas flow rate was composed by air and nitrogen
gas flow rates, combined to a specified total value for obtain-
ing a desired and constant DO concentration for all the ex-
periments inside the reactor.
In order to evaluate the influence of external mass transfer
on COD removal disregarding the effects on the ammonium
oxidation, and vice-versa, independent experiments were
performed in batch mode. For this, before starting each one of
the experiments, the supernatant was removed and new
medium was carefully added inside the reactor. The
Table 1 e Composition of the influent.
Substance Value Unit
Acetic acid (C2H4O2) 80.80 mg
NH4Cl 66.60 mg
NaHCO3 64.80 mg
NaH2PO4.7H2O 305.37 mg
KH2PO4 9.05 mg
MgSO4.7H2O 24.30 mg
CaCl2.2H2O 12.15 mg
NaOH 22.28 mg
Trace metal solution (Vishniac and Santer, 1957) 0.08 mL
Fig. 1 e Schematic variation of the substrate removal rate
as a function of mixing intensity within the reactor.
composition of the media utilized for the tests were similar to
that used for the normal operation of the reactor (Table 2). In
the tests for COD removal, only the necessary amount of
ammonium for assimilatory activity was added at the me-
dium. In the tests for ammonium oxidation, no organic matter
was added and NaHCO3 was added in excess. The pH and
temperature were kept at the same level than those used at
the normal operation of the reactor, i.e., 7.5 and 20  C,
respectively.
Each experiment lasted one hour, with sampling time of
15 min. Off-line measurements were made for ammonium
nitrogen and COD. The ammonium concentration in the su-
pernatant was determined spectrophotometrically using a
commercial cuvette test kit (Hach Lange, Germany). The ace-
tate concentration in the supernatant was measured using
high-performance liquid chromatography and then converted
in COD concentration. During the experiments, the DO was
always the limiting substrate for the removal of the pollutants
and its concentration was measured online and kept at a
specified value.
The experimental removal rate (rexp) for each experiment
was obtained from the slope of the linear decrease obtained
from the values of ammonium nitrogen or acetate concen-
trations as function of time. Oxygen was the limiting substrate
and the fluxes were, therefore, defined as a function of the
oxygen concentration. The calculation of the experimental
exp
flux of oxygen to the biofilm (JO2 ) was computed by Eq. (1),
where gO2 =substrate is the stoichiometric factor between the ox-
ygen and substrate uptake, Vbulk is the volume of liquid inside
Table 2 e Initial Concentrations of acetic acid and
ammonium chloride on the external mass transfer
L
1 experiments.
L
1
Substance Value Unit
L
1
L
1 COD experiments
L
1 Acetic acid (C2H4O2) 220 mg L
1
L
1 NH4Cl 47 mg L
1
L
1 Ammonium experiments
L
1 Acetic acid (C2H4O2) 0 mg L
1
L
1 NH4Cl 95.5 mg L
1
 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8 93

the reactor and A is the total surface area of the carriers within parameter for each substrate (gO2 =substrate ), which is the stoi-
the reactor. chiometric relation between oxygen and substrate conver-
sions, as well the diffusivity of oxygen in water to calculate the
exp Vbulk
JO2 rexp gO2 =substrate (1) boundary layer thickness (Table 3). For the ammonium
A
removal, it was not observed nitrite formation in the experi-
The flux of oxygen over the boundary layer (JBL O2 ) can be ments; therefore, it was assumed full oxidation of ammonium
calculated from Eq. (2), as a function of the concentration to nitrate. However, in the case of nitrite accumulation this
difference between the liquid phase (Sbulk O2 ) and the surface of must be taken into account for the stoichiometric parameter
interf
the biofilm (SO2 ) and the external mass transfer coefficient value.
(kO2 ). High values of the coefficient indicate low mass transfer
resistance and vice-versa.

 3. Results
interf
O2 kO2 SO2
SO2
JBL bulk
(2)
Biof 3.1. Long-term reactor operation
The flux of substrate over the surface of the biofilm (JO2 )
can also be estimated when a zero order approach for con-
This study used a lab-scale MBBR process operated for the
version in the biofilm is used, according to Eq. (3) (Harremoe s,
biof simultaneous oxidation of COD and ammonium. Two different
1978), where DO2 is the oxygen diffusivity in the biofilm,qO2 ;max
groups of bacteria are responsible for the removal of the sub-
is the maximum substrate specific conversion rate and X is
strates. While COD is removed by heterotrophic bacteria,
the biomass density.
ammonium is removed by both assimilatory activity and
biof 
 qq mostly by the nitrification process, the last one performed only
Biof biof dSO2  biof interf
JO2 DO2  2DO2 qO2 ;max X SO2 (3) by autotrophic bacteria. The nitrification process rate is char-
dz 
zL
acterized by the sum of the production rates of N
NO
2 and
The removal rate without influence of external mass N
NO
3 . Under standard operation all acetate supplied and an
transfer (R) is obtained from the weighted average by the average of 64% of ammonium was removed by the MBBR. The
experimental variance, s2j , (Eq. (4)) of the experiments (j) with average surface specific conversion rate for ammonium and
high mixing intensity. It is assumed that under this condition nitrification found were 0.68 g NeNH4 m
2 d
1 and 0.52 g N-
there is no influence of the external mass transfer, since the (NO


2 NO3 ) m

2
1
d . Therefore, about 76% of the ammonium
conversion become independent of the mixing intensity. was removed by nitrification.
P .
exp
rj s2j 3.2. External mass transfer experiments
j
R P . (4)
1 s2j 3.2.1. Ammonium oxidation
j
The batch experiments for determining the external mass
Moreover, in this region of the experiments, since there is a transfer coefficient were performed at different mixing in-
negligible small boundary layer thickness and no accumula- tensities. By blending air and nitrogen gas, it was possible to
tion of substrate on the biofilm surface, the fluxes from Eqs. (1) change the mixing intensity by changing aeration rates while
exp biof
and (3) were assumed to be equal, JO2 JO2 , and the concen- keeping the DO concentration at the same value in all the
tration of oxygen at the surface of the biofilm was assumed performed experiments. A DO concentration of 5 mg L
1 was
interf
the same as the bulk concentration, SO2 Sbulk O2 . This results chosen, since this was the same concentration used in the
in Eq. (5). normal operation of the reactor. The results of three experi-
qq ments for ammonium removal are illustrated in Fig. 2. From
Vbulk biof
RgO2 =substrate 2DO2 qO2;max X Sbulk O2 (5) these experiments it is possible to verify the rise of ammo-
A
nium removal rate with the increase in the gas flow rate, i.e.,
In the region where the influence of external mass transfer
in the mixing intensity. The removal rates were estimated by
is relevant, assuming no accumulation of substrate on the
linear regression of the experimental measurements.
biofilm surface, all three fluxes from Eqs. (1)e(3) are equal,
exp biof The ammonium removal rate increased with increasing
JO2 JBLO2 JO2 . This criterion coupled with Eq. (5) allows
gas flow rate until it reaches a maximum value (Fig. 3). The
computing directly the external mass transfer coefficient for
aeration rate of 1 vvm (or 2 L min
1) separated the behaviour
each experiment by Eq. (6).
of the ammonium removal rate in two different regions: below
Vbulk rexp R2 gO2 =substrate this value there was an increase with the mixing intensity;
kO2 (6) and, above it, the ammonium removal rate stabilized in a
A R2
rexp 2 Sbulk
O2
constant level. Comparing the values of the removal rate
The boundary layer thickness (d) estimate, used to evaluate within the range studied in these experiments, the difference
the influence of the mixing intensity, is defined by Eq. (7). between the minimum and the maximum removal rate was
about 40%.
Dwater
O2
d (7)
kO2
3.2.2. COD removal
Using the presented methodology for estimating the The COD removal rates were estimated in the same way as
external mass transfer coefficient requires only one biological described for the ammonium removal rates, although the
94 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8

Table 3 e Biological and diffusivity parameters.

Parameter Value Unit Reference
gO2 =COD;Het 0.35 gO2 (gCOD)
1 Mannina et al. (2011)
gO2 =N
NH4 ;Nit 4.35 gO2 (gN-NH4)
1 Mannina et al. (2011)
Dwater
O2 1.74  10
4 m2 d
1 Gujer and Boller (1986)

experiments, an increase of the COD removal rate with the

increment of mixing intensity was observed until a maximum
conversion was achieved at 1 vvm and higher. The higher
removal rate observed at 1.75 vvm also seemed to be caused
due some detachment of biofilm from the carriers. By
comparing the experimental removal rates values, a differ-
ence of about 27% was found between the minimum and the
maximum (without influence of external mass transfer)
removal rates.
The dissolved oxygen concentration had an influence on
the conversion rates with marginal external mass transfer
limitation (Table 4). This indicates, under these conditions,
that the internal mass transfer was still limiting the conver-
sion. The low oxygen uptake rate under pure nitrifying

Fig. 2 e Difference of ammonium concentrations during

three experiments with different vvm and linearization
(solid line e 0.375 L L1 min1; dashed line e
0.75 L L1 min1; dotted line e 1.25 L L1 min1).

experiments are not shown in the present document in order

to avoid repetition. Initially, the experiments were performed
at DO of 5 mg L
1 (Fig. 4a). However, to achieve this DO value,
very high mixing intensities were needed (above 1 vvm). In
this range of aeration rate no impact on the conversion was
noticed, indicating that external mass transfer was not
limiting the conversion. The high aeration rates seemed to
lead to some biofilm detachment as well.
To evaluate the influence of the external mass transfer for
COD removal, tests were performed at a lower DO of 3 mg L
1
(Fig. 4b). Similarly to the ammonium removal rate

Fig. 3 e Ammonium removal rates in the MBBR reactor as Fig. 4 e COD removal rates in the MBBR as function of vvm
function of vvm at a DO 5 mg L1. for (a) DO 5 mg L1, (b) DO 3 mg L1.
 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8 95

conditions compared to COD conversion conditions indicates

that, for the nitrifying conditions, conversion was limited by
the amount of nitrifying biomass.

3.3. Mixing intensity influence

From the experimental data presented in this study, the

external mass transfer coefficients and the respective
boundary layer thicknesses for the different mixing in-
tensities were estimated by the method described in Section 2.
The results show different values of the coefficients for
ammonium and COD removal. For COD removal, a higher
external mass transfer coefficient was found, suggesting a
smaller external boundary layer thickness. For nitrification,
the external mass transfer coefficient was lower indicating a
higher external mass transfer resistance.
The influence of the mixing intensity on the estimated
boundary layer thickness is presented in Fig. 5. The curves
obtained reveal a near linear trend of boundary layer thick-
ness decreasing with the aeration rate and, also, higher in- Fig. 5 e Boundary layer thickness as function of vvm for
fluence of the mixing intensity for the removal rate of N
NH 4 oxidation (DO 5 mg L
1
) and COD removal
1
ammonium (note the different slopes of the curves in Fig. 5). (DO 3 mg L ).
Another important detail is that although the effect of mixing
intensity on the decrease of the external mass transfer is
different, the same aeration rate is observed for the point
where the external mass transfer resistance becomes negli- The specific surface removal rate of ammonium obtained
gible (around 0.91 vvm). in this study was 0.68 g m
2 d
1. This rate can be increased by
using a lower organic loading rate. Hem et al. (1994) reported,
for a DO of 5 mg L
1 and 15  C, a decrease of nitrification to
insignificant values when the organic loads were higher than
4. Discussion 5 g BOD m
2 d
1. The nitrification rate obtained in this present
study was higher than reported by Hem et al. (1994), 0.52 g N-
4.1. Long-term reactor operation (NO


2
1
d for an organic loading of 4.8 g m
2 d
1,
2 NO3 ) m
this was likely due to the higher temperature, 20  C in our
The moving bed biofilm reactor was operated with complete study.
removal of COD (acetate) and an average ammonium removal
of 64%. During the long-term operation, the majority of the
4.2. External mass transfer experiments
ammonium removal was due to nitrification. In a process
where nitrification and removal of organic matter take place
For the removal of ammonium (DO: 5 mgO2 L
1) and COD (DO:
simultaneously, fast-growing heterotrophic bacteria, which
3 mgO2 L
1), an increase of the removal rate with the mixing
consume organic carbon, outcompete slow-growing autotro-
intensity was observed (Figs. 3 and 4b). Since the other vari-
phic bacteria, causing a decrease of the nitrification (Wanner
ables were kept constant, this increase is attributed to a
and Gujer, 1986). Due to difference in the growth rate, a
decreased boundary layer thickness or increased external
stratified biofilm arises, with heterotrophic bacteria occupying
mass transfer coefficient. This is in agreement with the
the superficial layers of the biofilm. Therefore, this stratified
expectation that mixing intensity is a relevant aspect that
structure causes a problem for nitrifiers since they will be
affects the external mass transfer (Kugaprasatham et al., 1992;
more affected by oxygen diffusion limitation (Zhang et al.,
Chen et al., 2006). MBBR reactors are in general more prone to
1994; Van Loosdrecht et al., 1995).
external mass transfer resistance, despite potentially high
mixing intensity, since most of the biofilm resides on the
protected surface inside of the carriers (Rusten et al., 2006).
Table 4 e Removal rates in the MBBR under conditions Moreover, the biofilm carriers move with the liquid flow.
where the external mass transfer rate is negligible. These factors make the turbulence of the bulk liquid strongly
Experiment Volumetric rate (R) Unit dampened near the biofilm surface inside the carriers.
NeNH4 (DO: 5 mg L
1) 6.7 0.2 mg NeNH4 L
1 h
1 The strong influence of external mass transfer in this
29.1 0.8 mg O2 L
1 h
1 system is revealed by the difference of the removal rates
COD (DO: 5 mg L
1) 158.1 0.8 mg COD L
1 h
1 observed within the ranges of aeration intensities evaluated
55.3 0.3 mg O2 L
1 h
1 (40% for ammonium and 27% for COD). In the experiments,
COD (DO: 3 mg L
1) 129.9 2.0 mg COD L
1 h
1 the removal rate of COD and ammonium were influenced by
45.5 0.7 mg O2 L
1 h
1
the mixing intensity up to 1 vvm (normal operation was at 0.5
96 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8

vvm). Above this mixing intensity, no further increase in

conversion rate was observed. The stabilization of the con-
version rate at high aeration rates indicates that external
mass transfer is not limiting the conversion at these
conditions.
By comparing the COD removal rates without influence of
external mass transfer a slight increase when increasing DO
from 3 to 5 mg L
1 was observed. This can be explained by a
higher penetration depth of oxygen inside the biofilm. Similar
observation were found in the nitrifying MBBR system studied
by Gapes and Keller (2009) where with the increase of DO
concentration inside the reactor higher nitrification rates
were obtained.
A comparison between the experimental measurements
obtained in this study and model results using the external
mass transfer coefficient determined are evaluated in a parity
plot (Fig. 6). Only the experiments where the external mass
transfer was varied are included in this figure. The initial
concentrations used in the simulations were the same as the
respective experiments. Most experimental data were well
described by the model. Only at the aeration rate of 0.625 vvm
the simulation underestimated the ammonium oxidation
rate. This was likely due to a deviation in the measured
ammonium oxidation rate (Fig. 5), and the experimental value
also had a relative large standard deviation. The good agree-
ment between experimental and calculated values indicates
the adequacy of the proposed methodology to determine the
external mass transfer coefficients.

4.3. Mixing intensity influence

The influence of the mixing intensity in MBBR was evaluated

for ammonium and COD removal. A stronger influence was
observed for ammonium removal when compared with COD
removal. The values of the external mass transfer coefficients
in the investigated range were 0.7e13.5 and 2.9e22.4 m d
1 for
ammonium and COD, respectively. Even though the values
have the same order of magnitude, the external mass transfer
coefficients were found always higher for COD experiments.
In other words, the values of boundary layer thickness were
always higher for the ammonium experiments being in the
range of 12.9e256.6 mm whereas for the COD experiments the
range was 7.8e61.1 mm.
This can be explained by the distribution of the bacteria in
the biofilm. As explained previously, the removal of ammo-
nium and COD are made by different groups of bacteria, which
can be located at different depths of the biofilm, with the ni-
trifiers growing deeper inside the biofilm. Since in this meth-
odology a homogeneous biofilm model was utilized, the
stratification aspects of the biofilm was not taken into account
in the model and this reflected directly in an apparent external
mass transfer coefficient determination for ammonium. In
fact this coefficient is affected by the existence of an external
layer of heterotrophs, which is producing an extra resistance
to mass transfer.
Studying a nitrifying MBBR system, Hem et al. (1994) found
a first order function of oxygen for the nitrification rate, which
indicates the influence of external mass transfer in that sys-
tem. External mass transfer coefficients of 1.2 and 1.3 m d
1
were found for acclimated biofilm under oxygen limitation
Fig. 6 e Comparison between experimental and simulated
(a) ammonium and (b) COD conversions in the tests for
determination of external mass transfer coefficient.
and ammonium limitation, respectively. Gapes and Keller
(2009), in a nitrifying MBBR system found 3.34 and
1.10 m d
1 for a high and low ammonium loading, respec-
tively. The higher value was explained by a more heteroge-
neity at the biofilm surface, which affects positively the
external mass transfer. Masic et al. (2010), in a nitrifying MBBR
system, found a boundary layer thickness in a range of
40e85 mm (3.2e1.5 m d
1). All of these results in MBBR are in
agreement with the range of the values determined for the
present methodology.
Siegrist and Gujer (1987), studying a trickling filter, found
an external mass transfer coefficient of 2 m d
1, a value
similar as an MBBR reactor. Nicolella et al. (1998), for more
intensely mixed biofilm airlift reactors, estimated this coeffi-
cient at an average of 10 m d
1 at a range of 0.3e3 vvm. Liquid
flow rates in the MBBR reactor are higher than in trickling
filters and a bit lower or similar to airlift reactors. The external
mass transfer is, however, comparable to trickling filters since
the biofilms are mainly present on the protected surface area
 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8
inside of the carriers. On the other hand, when the mixing
intensity is increased in the MBBR, the external mass transfer
coefficients can achieve similar values to those determined in
airlift reactors. A comparison between the external mass
transfer coefficients found in this work and in an airlift reactor
(Nicolella et al., 1998) is shown in Fig. 7. This shows how
relevant the external mass transfer rate is in the case of MBBR
systems, and how low mixing intensity can severely penalize
the removal rates achieved. Consequently, the knowledge of
this influence is fundamental for the design and optimal
operation of these systems.
5. Conclusions
Based on reactor hydrodynamics, a novel in-situ methodology
for the determination of external mass transfer coefficients in
a MBBR was developed. The proposed method is simple, with
a spreadsheet-based calculation for the external mass trans-
fer coefficient and only one single biological parameter (oxy-
gen consumed per unit of substrate converted) is required for
the substrate of interest. The simplicity of the experimental
methodology makes it possible to use it in full-scale systems,
with the possibility to use (the often present) mechanical
stirrers to maintain a similar turbulence level in the bulk
liquid. A potential challenge would be the need to know the
removal rate with no influence of external mass transfer,
which may require extra experiments until reaching this
mixing intensity condition.
Due to biofilm stratification, in a heterotrophic/autotrophic
biofilm the external mass transfer coefficient for the nitrifi-
cation process will be larger than for the COD conversion.
The external mass transfer coefficients varied significantly
as function of the mixing intensity in the reactor. This shows
the strong influence of mixing intensity on the conversion
rates in MBBR. This can be used in optimizing the design of
new reactors and/or determination of the optimal operation of
existing systems.
Fig. 7 e Comparison between the external mass transfer
coefficients found in this study and those determined in
airlift reactors (Nicolella et al., 1998) as a function of vvm.
97
Acknowledgements
The authors thank CNPq (302893/2013-0, 140595/2012-0,
245253/2012-2) and CAPES for providing scholarships and for
supporting this work. The authors also thank AnoxKaldnes for
providing carrier samples for this research.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.watres.2015.05.010.
references
Bassin, J.P., Dezotti, M., Sant'Anna, G.L., 2011. Nitrification of
industrial and domestic saline wastewaters in moving bed
biofilm reactor and sequencing batch reactor. J. Hazard. Mater.
185 (1), 242e248.
Beyenal, H., Tanyolac, A., 1998. The effects of biofilm
characteristics on the external mass transfer coefficient in a
differential fluidized bed biofilm reactor. Biochem. Eng. J. 1 (1),
53e61.
Boessmann, M., Neu, T.R., Horn, H., Hempel, D.C., 2004. Growth,
structure and oxygen penetration in particle supported
autotrophic biofilms. Water Sci. Technol. 49 (11e12), 371e377.
Chen, S., Ling, J., Blancheton, J.P., 2006. Nitrification kinetics of
biofilm as affected by water quality factors. Aquac. Eng. 34 (3),
179e197.
Ferrai, M., Guglielmi, G., Andreottola, G., 2010. Modelling
respirometric tests for the assessment of kinetic and
stoichiometric parameters on MBBR biofilm for municipal
wastewater treatment. Environ. Model. Softw. 25 (5), 626e632.
Gapes, D.J., Keller, J., 2009. Impact of oxygen mass transfer on
nitrification reactions in suspended carrier reactor biofilms.
Process Biochem. 44 (1), 43e53.
Gujer, W., Boller, M., 1986. Design of a nitrifying tertiary trickling
filter based on theoretical concepts. Water Res. 20 (11),
1353e1362.
Harremoe s, P., 1978. Biofilm kinetics. In: Michell, R. (Ed.), Water
Pollution Microbiology. Wiley-Interscience, New York, NY,
USA.
Hem, L.J., Rusten, B., degaard, H., 1994. Nitrification in moving
bed biofilm reactor. Water Res. 28 (6), 1425e1433.
Kugaprasatham, S., Nagaoka, H., Ohgaki, S., 1992. Effect of
turbulence on nitrifying biofilms at non-limiting substrate
conditions. Water Res. 26 (12), 1629e1638.
Mannina, G., Di Trapani, D., Viviani, G., degaard, H., 2011.
Modelling and dynamic simulation of hybrid moving bed
biofilm reactors: model concepts and application to a pilot
plant. Biochem. Eng. J. 56 (1e2), 23e36.
Masic, A., Bengtsson, J., Christensson, M., 2010. Measuring and
modeling the oxygen profile in a nitrifying moving bed biofilm
Reactor. Math. Biosci. 227 (1), 1e11.
Nicolella, C., Van Loosdrecht, M.C.M., Heijnen, J.J., 1998. Mass
transfer and reaction in a biofilm airlift suspension reactor.
Chem. Eng. Sci. 53 (15), 2743e2753.
Rasmussen, K., Lewandowski, Z., 1998. Microelectrode
measurements of local mass transport rates in heterogeneous
biofilms. Biotechnol. Bioeng. 59 (3), 302e309.
Rusten, B., Eikebrokk, B., Ulgenes, Y., Lygren, E., 2006. Design and
operations of the Kaldnes moving bed biofilm reactors. Aquac.
Eng. 34 (3), 322e331.
98 w a t e r r e s e a r c h 8 0 ( 2 0 1 5 ) 9 0 e9 8
Siegrist, H., Gujer, W., 1987. Demonstration of mass transfer and
pH effects in a nitrifying biofilm. Water Res. 21 (12),
1481e1487.
Stoodley, P., Yang, S., Lappin-Scott, H., Lewandowski, Z., 1997.
Relationship between mass transfer coefficient and liquid flow
velocity in heterogenous biofilms using microelectrodes and
confocal microscopy. Biotechnol. Bioeng. 56 (6), 681e688.
Van Loosdrecht, M.C.M., Tijhuis, L., Wijdieks, A.M.S., Heijnen, J.J.,
1995. Population distribution in aerobic biofilms on small
suspended particles. Water Sci. Technol. 31 (1), 163e171.
Vishniac, W., Santer, M., 1957. The thiobacilli. Bacteriol. Rev. 21
(3), 195e213.
Wanner, O., Gujer, W., 1986. A multispecies biofilm model.
Biotechnol. Bioeng. 28 (3), 314e328.
Wanner, O., Eberl, H., Morgenroth, E., Noguera, D., Picioreanu, C.,
Rittmann, B.E., van Loosdrecht, M.C.M., 2006. Scientific and
View publication stats
Technical Report No. 18: Mathematical Modeling of Biofilms.
IWA Publishing, London.
Wa sche, S., Horn, H., Hempel, D.C., 2002. Influence of growth
conditions on biofilm development and mass transfer at the
bulk/biofilm interface. Water Res. 36 (19), 4775e4784.
Wang, R.C., Wen, X.H., Qian, Y., 2005. Influence of carrier
concentration on the performance and microbial
characteristics of a suspended carrier biofilm reactor. Process
Biochem. 40 (9), 2992e3001.
Zhang, T.C., Fu, Y.C., Bishop, P.L., 1994. Competition in biofilms.
Water Sci. Technol. 29 (10e11), 263e270.
Zhang, T.C., Bishop, P.L., 1994. Experimental determination of the
dissolved oxygen boundary layer and mass transfer resistance
near the fluid-biofilm interface. Water Sci. Technol. 30 (11),
47e58.