Microsoft Word - 48dallavecchia.docx

CHEMICAL ENGINEERING TRANSACTIONS

VOL. 49, 2016

A publication of

The Italian Association

of Chemical Engineering
Online at www.aidic.it/cet

Guest Editors: Enrico Bardone, Marco Bravi, Tajalli Keshavarz
Copyright © 2016, AIDIC Servizi S.r.l.,
ISBN 978-88-95608-40-2; ISSN 2283-9216

Lanthanum Biosorption by Different Saccharomyces
cerevisiae Strains

Fabrizio Di Caprio*a, Pietro Altimaria, Elena Zannib, Daniela Uccellettib, Luigi Toroa,
Francesca Pagnanellia

a
Dipartimento di Chimica, Università “Sapienza” di Roma, Piazzale Aldo Moro 5, 00185, Roma, Italia.

b
Dipartimento di Biologia e Biotecnologie “C.Darwin”, Università “Sapienza” di Roma, Piazzale Aldo Moro 5, 00185, Roma,

Italia.
fabrizio.dicaprio@uniroma1.it

Biosorption can be a promising technology in rare earth metal separation and recovery due to the low costs of
waste biomasses (used as biosorbents) and the high selectivity exploiting specific interaction between metals
and biological active sites. In this work, Saccharomyces cerevisiae biomass was used to recover lanthanum.
Biosorption properties of two S. cerevisiae strains, wild type and rim20∆ mutant, have been tested.
Potentiometric titrations were carried out for rim20∆ mutant strain and compared with wild type. Nature of the
main active sites and their concentration were determined by implementing mechanistic models. Carboxylic,
amino and phosphoric sites are the main groups present. Higher concentration of negatively charged sites
was found in rim20∆ (0.0024 mol/g) than in wild type (0.0022 mol/g). The rate of lanthanum biosorption
process is very fast requiring only 10-20 minutes to reach equilibrium condition for both strains. Then
biosorption equilibrium tests were done for both biomasses by testing two equilibrium pH (4.0 and 6.0).
Maximum uptake capacities (qmax) were: 70 mg/g and 40 mg/g at pH 4.0 for rim20∆ and wild type,
respectively, and 67 mg/g and 80 mg/g at pH 6.0 for wild type and rim20∆, respectively. These data evidenced
that: rim20∆ mutant had a higher maximum biosorption capacity with respect to wild type counterpart, and that
pH had a relevant effect on lanthanum removal.
S. cerevisiae yeast denoted good lanthanum biosorption properties and, between tested strains, rim20∆ was
found to be the most promising for such aim.

1. Introduction
Actually rare earth metals (REM) are fundamental for the production of several new technologies. Lanthanum,
cerium, neodymium and yttrium constitute the most part of the global market of REM. Nevertheless,
geopolitical issues (95% of world supply for REM is located in China) and fast depletion of primary sources
(Das and Das, 2013) risk to compromise and limit future applications. An alternative way is promoting REM
recovery from Waste Electrical and Electronic Equipment (WEEE). It is calculated that each EU citizen
produces about 17 kg of WEEE/year and these wastes contain rare earth concentrations higher than primary
sources (Iannicelli Zubiani, et al. 2015). Following this aim different innovative hydrometallurgical processes
were developed, based leaching, precipitation, ion exchange and solvent extraction (SX) operations to
separate and purify REM contained in WEEE. However REM recovery technologies (and especially SX)
present high costs and sometimes also low selectivity (Lucas et al., 2014).
Biosorption can be a promising alternative technology for REM separation and recovery due to the low costs
of waste biomasses (used as biosorbents) and the high selectivity exploiting specific interaction between ionic
species in solutions (REM) and biological active sites (Wang and Chen, 2009). Several biomasses can be
used for such purpose. In this study Saccharomyces cerevisiae yeast was chosen due to its low cost as waste
of different industries. In addition it can be easy cultivated and it is a model organism in biology, easy to study
and to improve. S.cerevisiae is characterized by a cell wall, which is mainly constituted of proteins and
polysaccharides. These macromolecules contain different functional groups, which can be involved in
biosorption. As reported by Di Caprio et al. (2014), the main functional groups responsible for copper removal

DOI: 10.3303/CET1649007

Please cite this article as: Di Caprio F., Altimari P., Zanni E., Uccelletti D., Toro L., Pagnanelli F., 2016, Lanthanum biosorption by different
saccharomyces cerevisiae strains, Chemical Engineering Transactions, 49, 37-42 DOI: 10.3303/CET1649007

by S. cerevisiae are carboxylic, amino and phosphoric groups according to an ion exchange mechanism.
Biosorption onto yeast can be affected by different operating parameters, extensively investigated in previous
works (Wang and Chen, 2006). Most influential parameters are pH, temperature and presence of competing
ions in solution. Although S. cerevisiae was extensively studied in many biosorption applications, no one
reported sorption data concerning lanthanum removal using such biomass.
In this work different S. cerevisiae strains were evaluated for their lanthanum biosorption properties. Wild type
strain was compared with rim20∆ mutant. Previous works reported that rim20∆ mutant has higher surface
concentration of negatively charged groups on the cell wall with respect to the wild type, when stained with
alcian blue (Conde et al., 2003). According to such properties this mutant is a good candidate for biosorption
of lanthanum, generally present as cationic species in aqueous conditions. rim20∆ mutant is lacking of the
rim20p protein, which is required for the response to external pH variation. It has been reported that this
mutant is characterized by a higher concentration of mannosylphosphates than wild type, giving the higher
negative superficial charge.
In this study these two strains were characterized to determine their negative charge concentration by
potentiometric titrations. Then biosorption properties were tested in lanthanum removal equilibrium tests at pH
4.0 and 6.0. Maximum uptake capacity (qmax) was then estimated for all tested conditions and strains.

2. Materials and Methods
2.1 Yeast production
Yeast cells were grown in YPD medium (1% peptone, 1% yeast extract, 2% glucose, DIFCO) for 72 h at 303
K. Then cells were centrifuged and rinsed twice with distilled water to remove residual traces of the culture
medium. The obtained pellet was dried in an oven at 338 K until a constant weight was attained.

2.2 Potentiometric titrations
Potentiometric titrations were carried out by following the method reported by Di Caprio et al. (2014).

2.3 Kinetic tests
For realizing kinetic tests 0.25 g of dried yeast biomass was suspended in 150 mL flask with 80 mL of
lanthanum solution (La3+ 100 mg/L, NaNO3 0.01 M, fixed pH). Tests were carried out for both strains rim20∆
and wild type at pH 4.0 ± 0.2 and 6.0 ± 0.2. At 10, 15, 20, 70 and 90 minutes samples were collected and
centrifuged at 8000 rpm. Supernatant was stored at 277 K.

2.4 Lanthanum analysis
For all biosorption tests, La3+ concentration in the collected samples was analysed by atomic absorption
spectroscopy (Analytic Jena Contra 300).

2.5 Adsorption equilibrium tests
Equilibrium isotherms were determined for rim20∆ and wild type strains both at pH 4.0 ± 0.2 and 6.0 ± 0.2. For
each isotherm different initial concentrations in the range of 100-400 mg/L of La3+ were prepared and used.
Dried yeast biomass was suspended in 150 mL flask with 80 mL of lanthanum solution. For solutions with a
concentration between 200 and 400 mg/L, 0.16 g of biomass was weighted. Instead for the tests with 100
mg/L of initial lanthanum concentration, 0.05 g of biomass was weighted. Lanthanum solutions were prepared
using La(NO3)3·6H2O salt. After 10, 30 and 60 minutes pH was measured and corrected at desired value with
NaOH 1 M or HCl 1 M. After 60 minutes, 5 mL samples were collected, centrifuged at 8,000 rpm, and
supernatant stored at 277 K.

3. Results and discussion
3.1 Potentiometric titrations
In potentiometric titrations, pH was determined as a function of titrant addition (mol of NaOH added to the
biomass suspension). Imposing electro-neutrality condition in the system, the net surface charge
concentration onto the biomass (Q, mol/g) was estimated for each titrant addition (i.e. for each point of the
titration curve). Net surface charge concentration was calculated by using the following equation Q = (1)
where V (mL) is volume of the suspension, m (g) the amount of biomass in the volume and Q the net charge
of the biomass.

In Figure 1 potentiometric titrations of both strains were reported: only a subsample of the experimental data
were reported in this graph in order to distinguish experimental data and modelling prediction. Both rim20∆

and wt biomasses are characterized by net positive surface charge at low pH values, and by a predominance
of negative charged sites at high pH values (Figure 1). Accordingly if ionic exchange is the main mechanism
involved in lanthanum biosorption, it will be influenced by pH variation, and in particular lanthanum-site
interactions will be favoured as pH increases. The point of zero charge (PZC) were 5.78 ± 0.06 and 6.3 ± 0.6
for rim20∆ mutant and wild type, respectively. Titration data showed that the mutant was characterized by a
higher concentration of negative charges. The previous results obtained by using alcian blue method (Conde
et al., 2003) are therefore confirmed also by potentiometric titrations.
Mechanistic models developed by Di Caprio et al. (2014) were used in order to describe experimental data,
characterize acid-base properties and quantify concentration of charged sites. In particular, the model 2s-A
and 3s-A were tested assuming specific schemes of protonation reactions for the active sites on the biomass
surface. Model 3s-A assumed the presence of three sites, first and second sites are negative charged in
deprotonated form and neutral in protonated form, while the third site is positive when protonated and neutral
when deprotonated. Model 2s-A is a reduction of 3s-A model, obtained removing the second site.
Negative charge surface concentration given by 3s-A model is reported in Eq. (2). = (2)
where the concentration of the three active sites is indicated by the terms S1, S2 and S3, while KH,1, KH,2 and
KH,3 are the equilibrium constants relative to the protonation reactions of the corresponding sites. They
indicate average KH of the active sites estimated. It is important to remark that logKH,j is equal to –logKa,j,
(pKa), where Ka,j is the acid dissociation constant of the corresponding site. The terms m1,m2 and m3, are
related to the heterogeneity of the sites, these terms are near to 1 when the heterogeneity is low and on the
contrary, when it is high, they are near to 0 (Pagnanelli, 2011).
For rim20∆ mutant the best results were obtained by using 3s-A model (i.e. model giving the lowest
confidence intervals of the estimated parameters), while best model for wild type is 2s-A model.
In Table 1 the estimated parameters obtained by non-linear regression of experimental data of potentiometric
titrations by using these two models were reported.

Figure 1: Potentiometric titrations for rim20∆ mutant and wild type strain of S. cerevisiae. With a black line the
best models which describe experimental data are reported.

The total concentration of negative charged sites was 0.0024 ± 0.0001 mol/g for rim20∆ mutant and 0.0022 ±
0.0001 mol/g for wild type strain. Three active sites for rim20∆ had logKH of 3.05, 6.60 and 10.12 for S1,S2 and
S3, respectively. Sites S1 and S3 probably correspond to phosphodiester and carboxylic groups (S1) and amino
groups (S3). The S2 site had a logKH probably corresponding to a phosphoric groups. This kind of groups is
relevant only for representing rim20∆ titration, even though with one order magnitude concentration lower than
the other two kinds of sites. These results confirmed previous works (Di Caprio et al. 2014) and are in

agreement with other findings (Conde et al., 2003) indicating mannosylphosphate groups as responsible of
the high negative charge in rim20∆.

Table 1: Parameters estimated by non-linear regression of potentiometric data using mechanistic models 2s-
A and 3s-A for wild type and rim20∆, respectively; intervals of confidence calculated with α=0.05.

Rim20∆
(model: 3s-A)

Wild type
(model: 2s-A)

[S1] (mol/g)
[S2] (mol/g)
[S3] (mol/g)
logKH,1 (pKa,1)
logKH,2 (pKa,2)
logKH,3 (pKa,3)
m1
m2
m3

0.0021 ± 0.0001
0.00025 ± 0.00004
0.00192 ± 0.00008
3.05 ± 0.06
6.60 ± 0.03
10.12 ± 0.06
0.34 ± 0.02
1.2 ± 0.1
0.55 ±0.02

0.0022 ± 0.0001
-
0.0019 ± 0.0001
3.6 ± 0.1
-
10.2 ± 0.1
0.35 ± 0.03
-
0.59 ± 0.03

3.2 Determination of lanthanum equilibrium isotherms
The effect of operating conditions on La biosorption onto S. cerevisiae strains was assessed by performing
equilibrium tests for different initial La concentration in solution and at different equilibrium pH. In particular the
effect of two pH values (4.0 ± 0.2 and 6.0 ± 0.2) was tested for both strains.
Equilibrium time was preliminary determined by kinetic tests. These preliminary tests showed that 20 minutes
are sufficient for having no more relevant variation of lanthanum residual concentration in solution for both
strains and pH tested (data not shown here).
Results of equilibrium tests were reported in Table 2. Experimental data show that there is not a dependence
of the amount of adsorbed lanthanum (qe) with respect to corresponding tested equilibrium concentrations in
solution (Ce). This indicates that both strains experienced saturation conditions at the two levels of pH in the
investigated range of concentrations. This means high efficiency in lanthanum removal at low initial lanthanum
concentrations. For each isotherm the measured qe were therefore considered as independent estimations of
the maximum adsorption capacity (qmax). The average value of the qe experimentally determined was used to
estimate qmax. Data were then statistically analyzed by using an analysis of variance (ANOVA) by considering
a 22 design with α=0.05. Through this analysis both pH and strains used have shown a statistically positive
significant effect on maximum sorption capacity. This is shown in Figure 2, in which the effects calculated for
each investigated factor are compared with respect to the least significant difference (LSD).

Table 2: Amount of equilibrium lanthanum adsorbed for the corresponding concentrations in solution are
reported for both strains at pH 4 and 6.

Strain

Ce (mg/L)

pH 4
qe (mg/g)

average (qmax)

Ce (mg/L)

pH 6
qe(mg/g)

average (qmax)

Wild type

112
171
260
320
68

54
56
26
47
39

40 ± 10 75
151
216
279
47

68
60
69
64
75

67 ± 6

Rim20∆ 60
133
202
266
30

73
68
58
72
98

70 ± 10 28
116
176
237
40

94
87
74
67
96

80 ± 10

Figure 2: Effects of the investigated factors on the maximum lanthanum adsorption capacity (qmax). With
straight lines value of the least significant difference (LSD) is indicated.

As proved by previous works (Di Caprio et al., 2014, Wang and Chen, 2006) pH has a relevant effect on
biosorption of metals onto yeast biomass. This is mainly due to competition between metals and protons for
the active sites on the biomass: when pH increases the concentration of negative charges on the biomass
surface increases too (Figure 1). This effect leads to an increment of negatively charged active sites available
for lanthanum biosorption when pH changes from 4.0 to 6.0, and accordingly to an increment of lanthanum
uptake capacity. The best performance of rim20∆ mutant with respect to wild type strain can be explained
considering data reported in Figure 1 and in Table 1. For a given pH rim20∆ has a higher concentration of
negatively charged sites. La biosorption capacity found in this work is higher than those reported in previous
studies for other biomasses. For example La sorption capacity of 22.94 mg/g was found for Pinus brutia leaf
powder (Kütahyali et al., 2010), and 28.65 mg/g for Platanus orientalis leaf powder (Sert et al., 2008). Similar
capacity was found for Sargassum fluitans ranging between 40 and 70 mg/g for the same pH of the present
work (Palmieri et al. 2002). In contrast dramatically higher capacity was found by Kazy et al., (2006) for
Pseudomonas sp. corresponding to about 1,900 mg/g.

Figure 3: Removal efficiency of the equilibrium tests carried out for rim20∆ and wild type at pH 4.0 and 6.0.

In Figure 3 the removal efficiencies evaluated for the equilibrium tests carried out in this work were reported.
Tests done with an initial lanthanum concentration of 100 mg/L were not included because using a different
biomass concentration. The removal efficiency decreased when initial lanthanum concentration was
increased. The efficiency was 50-90 % for an initial concentration of 200 mg/L and became 20-35% for 400
mg/L.

4. Conclusions
S. cerevisiae yeast was found to be a promising biomass to recover lanthanum in recycling processes for
RAEE. In particular most promising properties are the ability to remove high amount of lanthanum also at low
concentrations and the loading capacity (qmax) higher than different other tested biomasses. The higher
biosorption capacity of the rim20∆ mutant with respect to wild type indicated that relevant improvement can be
reached by using mutant strains following the way of maximizing negative surface charge as target in strain
selection.

References

Conde R., Guadalupe P., Cueva R. and Larriba G., 2003. Screening for new yeast mutants affected in
mannosylphosphorylation of cell wall mannoproteins, Yeast. 20, 1189-1211.

Das N. and Das D. 2013. Recovery of rare earth metals through biosorption: An overview, J. Rare Earth. 31,
933-943.

Di Caprio F., Altimari P., Uccelletti D., Pagnanelli F. 2014. Mechanistic modelling of copper biosorption by wild
type and engineered Saccharomyces cerevisiae biomasses, Chem. Eng. J. 244, 561-568.

Iannicelli Zubiani E. M., Cristiania C., Dotellia G., Stampino P. G., Pelosato R. 2015. Recovery of Rare Earths
and Precious Metals from Waste Electrical and Electronic Equipment by Acid Leaching and Immobilized
Chelating Agents, Chemical Engineering Transactions. 43, 2413-2418.

Kazy S. K., Das S. K., Sar P. 2006. Lanthanum biosorption by a Pseudomonas sp.: equilibrium studies and
chemical characterization, J Ind Microbiol Biotechnol. 33, 773-783.

Kütahyali C., Sert Ş., Çetinkaya B., İnan S. and Eral M. 2010. Factors Affecting Lanthanum and Cerium
Biosorption on Pinus brutia Leaf Powder, Sep. Sci. Technol. 45, 1456-1462.

Lucas J., Lucas P., Le Mercier T., Rollat A., 2014. Davenport W. G. Rare Earths: Science, Technology,
Production and Use, Elsevier.

Pagnanelli F. 2011. Equilibrium, Kinetic and dynamic modelling of biosorption processes, Microbial
Biosorption of Metals, Springer, 59-120.

Palmieri M. C., Volesky B., Garcia O. J. 2002. Biosorption of lanthanum using Sargassum fluitans in batch
system, Hydrometallurgy. 67, 31–36.

Şert S., Kütahyali C., 1, İnan S, Talip Z., Çetinkaya B., Eral M. 2008. Biosorption of lanthanum and cerium
from aqueous solutions by Platanus orientalis leaf powder, Hydrometallurgy. 90, 13–18.

Wang J., Chen C. 2006. Biosorption of heavy metals by Saccharomyces cerevisiae: A review, Biotechnol.
Adv. 24, 427–451.

Wang J., Chen C. 2009. Biosorbents for heavy metals removal and their future, Biotechnol. Adv. 27, 195–226.