Bio-inspired synthesis of mesoporous s licas using large molecular

weight poly-L-lysine at neutral pH

Joao Pires

Ana C. Fernandes

Romeu Avo
Received: 31 January 2014 / Accepted: 17 May 2014 / Published online: 3 June 2014
Springer Science+Business Media New York 2014
Abstract Among the various types of amorphous silica-
based materials, those that have mesopores (pore widths
between 2 and 50 nm) constitute an important group. Silica
mesoporous materials have been considered in elds such
as catalysis, adsorption, sensing, and electronics or, more
recently in drug delivery. The synthesis of silica meso-
porous materials usually involves corrosive reaction media
and high temperatures. Nevertheless, some living organ-
isms such as diatoms or sponges produce silica in non-
corrosive environments and at ambient temperatures.
Important progress has been achieved in the synthesis of
silica-based materials by biomimetic or bio-inspired
methodologies, but the number of studies that use bio-
molecules and address specically the preparation of
mesoporous materials is reduced. We report in this work, to
our knowledge for the rst time, a methodology to obtain
mesoporous silicas that involves, simultaneously, a bio-
molecule (poly-L-lysine) and neutral pH. The prepared
materials have pores with widths between 2 and 8 nm and
specic surface areas between 232 and 616 m
2
g
-1
.
Introduction
Silica-based porous materials have a wide range of appli-
cations covering electronics, catalysis, and adsorption pro-
cesses or, more recently, drug delivery [14]. IUPAC
classies the type of porosity according to the pore widths
in micropores (pore widths less than 2 nm), mesopores
(pore widths between 2 and 50 nm), and macropores as
those pores with widths exceeding 50 nm [5]. The research
concerning mesoporous silicas, as well as their effective
and potential applications, gained important momentum
with the discovery of MCM-type materials, such as MCM-
41 [6] or SBA type solids, as is the case of SBA-15 [7].
These two examples are particularly interesting because
MCM-41 material is synthesized at elevated pH and, con-
trarily, SBA-15 is synthesized under acid conditions [6, 7].
Additionally, both synthesis require relatively high tem-
peratures and, in some cases, high pressures (autoclave).
In contrast, living organisms such as diatoms or sponges
are able to form a variety of biogenic silica structures under
mild conditions such as pH near 7 and ambient temperature
and pressure [8]. The possibility of using these mild con-
ditions in the synthesis of porous silica materials would
have a great impact, not only due to environmental reasons,
but also due to operational reasons, reducing costs related
with corrosion, by avoiding extreme pHs, and decreasing
energy consummation. The biomimetic synthesis of silica
materials, that is, making silicas as the living organisms do,
remains a major challenge since the natural reaction
mechanism is still unknown [9]. However, important pro-
gress was made using methodologies that are, at least
partially, inspired by those used by living organisms. These
methodologies are often referred to bio-inspired synthesis
methods, and major key aspects were identied. For
instance, it is known that the terminal amine groups present
Electronic supplementary material The online version of this
article (doi:10.1007/s10853-014-8335-8) contains supplementary
material, which is available to authorized users.
J. Pires (&) A. C. Fernandes R. Avo
Department of Chemistry and Biochemistry, and CQB, Faculty
of Sciences, University of Lisbon, Ed. C8, Campo Grande,
1749-016 Lisbon, Portugal
e-mail: [email protected]
A. C. Fernandes
e-mail: [email protected]
R. Avo
e-mail: [email protected]
1 3
J Mater Sci (2014) 49:60876092
DOI 10.1007/s10853-014-8335-8
in the biomolecules are essential to interact with mono-
silicic acid to induce the silica precipitation. In this way, a
variety of different silica precursors have been used,
together with a number of templates with amine functions,
as for instance polypeptides [1020] which is also illus-
trated in related literature reviews [9, 21].
One of the main lines of the studies related with the bio-
inspired synthesis of silica materials using polypeptides or
proteins is the inuence of the synthesis conditions on the
particle-size and, particularly, on the morphology of the
silica particles, although various studies present electron
microscopy results, where it can be seen that macroporosity
(as dened by IUPAC) is created. Surprisingly, only a very
restricted number of works mention the formation of
porosity, particularly mesoporosity [17, 20]. In this work,
we focused on the development of mesoporosity by a
methodology that involves the use of a biomolecule, in this
case a large molecular weight poly-L-lysine (PLL), at
neutral pH and ambient temperature, using tetramethyl
orthosilicate (TMOS) as silica source. We choose PLL
(Supporting InformationScheme 1) because it is one of
the reported biomolecules that can precipitate silica, with
the amino groups of the side chain cationically charged at
neutral pH [10, 22], playing a major role. Additionally,
PLL is commercially available, avoiding the difcult
experimental steps related to the extraction of specic
proteins that could be used also [13, 14, 23]. TMOS was
used because it is reported to be one of the more reactive
orthosilicates, for instance when compared with tetraethyl
orthosilicate [15].
To our knowledge, this is the rst study showing that by
a bio-inspired methodology that uses large molecular
weight PLL at neutral pH and ambient temperature, it is
possible to obtain mesoporous silicas whose pore size
distribution is within the range of dimensions of other
relevant silicas, such as MCM-41 and SBA-15 type mate-
rials. The methodology used in the present work differs
from the traditional templating route that is used to prepare
for instance MCM-41 or SBA-15 type materials because it
does not use harsh pHs conditions nor the organic molecule
(PLL) needs high temperature calcination or organic sol-
vent extraction to be removed.
Materials and methods
Pure water (Milli-Q, Millapore) was used all over the
study. PLL, MW 150.000300.000, 0.1 % w/v solution
was from Sigma-Aldrich and TMOS (98 %) from Flucka.
Phosphate buffer was prepared with NaH
2
PO
4
and Na
2-
HPO
4
.7H
2
O, both [99 % from Sigma-Aldrich. Materials
synthesis was adapted from previous reports [15, 16]. In
brief, to a solution of TMOS in chloridric acid phosphate
buffer was added, followed to the addition of PLL solution.
The mixture was stirred for 5 min. A gel was formed and,
to remove water in excess, the gel was centrifugated
(Bioblock Scientic, model Sigma). The gel was then left
to dry at ambient temperature, ground in a mortar, and
washed with water (250300 cm
3
). The precise amounts of
each reactant and the corresponding sample labeling are
given in Table 1, where the materials were divided into
two series, A and B, to emphasize the differences between
the total volume of reactants between the two series.
Since to get reproducible results in the surface area and
porosity analysis (by nitrogen adsorption), it is important to
degas correctly the materials, normally under vacuum, and
to be sure that the analyzed properties were characteristic
of each material, we took the option of heating all the
samples in a tubular oven at 200 C. To verify if there was
an effect of this heat treatment on the surface area and
porosity, we also heated selected samples at 50 and 400 C.
Nitrogen (Air Liquid, 99.999 %) adsorption isotherms
were determined in a volumetric automatic apparatus
(Quantachrome, Nova 2200e) at -196 C. The sample,
(between 50 and 100 mg), was degassed for 2.5 h at a
pressure lower than 0.133 Pa and the temperature of
200 C for 2.5 h or 50 C for 15 h where indicated. FTIR
spectra were collected on a Nicolet 6700 at 2 cm
-1
reso-
lution in KBr wafers. Thermogravimetry with differential
scanning calorimetry experiments was carried out in an
apparatus from Setaram (TG-DSC 111), with 0.001 mg of
precision, under dry air ux. X-ray diffractograms were
obtained by the powder method in a Philips PW 1710
diffractometer, with automatic data acquisition (APD
Phillips (v.35B) software) and CuKa (k = 1.5406 A

)
radiation.
Results and discussion
Two series of samples, A and B, were prepared. Table 1
presents the main experimental differences in the prepa-
ration of the samples as well as their labeling. Comparison
of the FTIR spectra of the samples (an example in Sup-
porting Information Fig. S1) shows the expected three
characteristic peaks of silica: SiOSi asymmetric
stretching at 1090 cm
-1
, the symmetric stretching at
790 cm
-1
, and SiOH stretching near 960 cm
-1
[10, 24];
and a broad band centered at 3500 cm
-1
attributable to OH
groups, from the silica and from physisorbed water.
Comparison of the spectra of the samples before and after
the washing step (Supporting Information Fig. S1) shows
that the unwashed sample presents bands in the region
16201350 cm
-1
that are attributable to the PLL [10, 25].
This result also indicates that, in the experimental condi-
tions used, the washing process removed the PLL from the
6088 J Mater Sci (2014) 49:60876092
1 3
silica-PLL composite. TG-DSC curves for selected exam-
ples of materials from the A and B series (see Table 1 for
labeling) are given in Fig. 1. Mass loss, depending on the
samples, in a range of values between 10 and 16 % occurs
essentially below 200 C, in an endothermic process which
is due to the loss of physisorbed water [26]. The mass
decrease for temperatures above 200 C is small (about
3 %) and is attributable essentially to a progressive dehy-
droxylation of the samples [26] and not to the decompo-
sition of PLL, a result that is in line with the FTIR data
discussed above.
All samples resulted to be amorphous from XRD
analysis (one example is given in Supporting Information
Fig. S2). The nitrogen adsorptiondesorption isotherms at
-196 C (Fig. 2a, b) can be classied into type IV
according to the IUPAC classication. This type of iso-
therm, with its characteristic hysteresis loop is associated
with mesoporous materials [5, 27]. To conrm that the
materials are exclusively mesoporous, particularly to
verify if microporosity is not present to some extent, the
nitrogen adsorption data were analyzed by the a
s
-method,
as described in fundamental textbooks [27, 28]. The a
s
-
plots obtained (selected examples in Supporting Infor-
mation Fig. S3) show straight lines that extrapolate to the
origin, evidence that there is no micropore lling [27].
The specic surface area (A
BET
) and mesoporous volumes
of each sample are given in Table 1, where the samples
A0 and B0 are presented for comparison purposes, since
no PLL was used in these cases. To obtain the meso-
porous-size distributions the Broekhoff-de Boer method,
in a version simplied with the FrenkelHalseyHill
equation (BdBFHH) [29] was used. This method is
known to give more reliable mesopore-size distributions
than the more generally used BarretJoynerHalenda
(BJH) method [30], and the results often approach those
obtained by more elaborated methodologies such as
methods based on the density functional theory (DFT)
[31]. The mesoporous-size distributions are given in
Fig. 3a, b) where, for comparison purposes, the results for
a MCM-41 and a SBA-15 samples were included also.
Considering the materials from the A series it can be seen
from Table 1 and Fig. 3 that the total mesoporous volume
only varies within 6 % but to ascertain the effect of PLL
we must not consider only the total values of A
BET
or
total mesopore volume, but also the effect of PLL on the
type of mesopores that are created, that is, in the meso-
pore-size distribution of the samples. The mesoporous-
size distribution in the sample without PLL (A0) is much
more irregular than in the remaining samples of this series
since it presents peaks between 2 and 4 nm and also a
broad peak near 9 nm. The samples with PLL, particu-
larly A1 and A2 present more uniform pore sizes.
Therefore, in the A series, PLL had the effect of allowing
to produce materials with more uniform pore dimensions.
In the case of the samples from the B series, which were
prepared with high amounts of PLL and total volume of
reaction mixture, the variations on the mesopore volume
(Table 1) are now larger and some materials present
higher pore volumes than those obtained in the A series.
The pore-size distributions in series B are shifted to
higher values in relation to the A series. However, the
sample without PLL (B0) presents a mesopore-size dis-
tribution that shows that the material has the narrow
mesopores of this series. As can be seen in Fig. 3, all the
materials from A or B series present distributions of pore
sizes that are broader than those for MCM-41 or SBA-15,
that is, the materials from A and B series present mes-
opores that have less uniform sizes. Nevertheless, the
samples A2 and B3 approach the distributions of MCM-
41 and SBA-15, respectively, although with less values of
total mesopore volumes [6, 7], thus has overall, PLL has a
benecial effect since, depending on the experimental
conditions, it allows to obtain materials with various types
of mesopore size distributions.
Table 1 Volumes of reactants
(in cm
3
) used in the synthesis of
the various materials. Specic
surface area (A
BET
) and
mesoporous volumes of each
sample are also given
a
0.1 % w/v
b
10 mM
c
0.1 M
Material PLL
a
HCl
b
TMOS Phosphate
buffer
c
A
BET
(m
2
g
-1
)
V
meso
(cm
3
g
-1
)
A0 0 17.0 3.0 0.8 586 0.48
A1 0.1 17.0 3.0 0.8 454 0.49
A2 0.5 17.0 3.0 0.8 576 0.46
A3 1.0 17.0 3.0 0.8 616 0.49
B0 0 17.0 3.0 24.0 510 0.52
B1 6.0 17.0 3.0 24.0 493 0.54
B2 3.0 17.0 3.0 24.0 243 0.41
B3 1.5 17.0 3.0 24.0 232 0.38
B4 6.0 8.5 1.5 24.0 421 0.59
B5 6.0 8.5 1.5 12.0 393 0.46
J Mater Sci (2014) 49:60876092 6089
1 3
As mentioned in the Materials and Methods section,
before the analysis by nitrogen adsorption all samples were
heated under air in a tubular oven at 200 C, to prevent that
the standard vacuum degassing, needed just before the
analysis, would introduce any change in the materials. To
conrm the stability of the porosity with the temperature,
some samples were heated at 50, 200, and 400 C. The
results obtained are given in Fig. 4, in that case for the
material B1, and show that the three nitrogen adsorption
desorption isotherms are almost superimposed, that is, the
porosity of the material did not changed after the different
heat treatments. This result also enlightens one of the
advantages of the bio-inspired method used here since, for
instance, in the synthesis process of MCM-41 or SBA-15 a
step of removing the organic template is mandatory to free
the porosity. This step need harsh conditions since it can be
done by extraction with organic solvents, usually under a
reuxing temperature, or by calcination at temperatures not
less than 500 C.
Fig. 1 Thermogravimetric (TG) and differential scanning calorime-
try (DSC) curves for the indicated samples
Fig. 2 a, b Nitrogen adsorptiondesorption isotherms at -196 C for
the samples from the A and B series, respectively
6090 J Mater Sci (2014) 49:60876092
1 3
Conclusions
In mild reaction conditions of synthesis, by the simulta-
neous use of high molecular weight PLL and neutral pH, it
was possible to obtain mesoporous silica materials that
have pores with widths between 2 and 8 nm and A
BET
between 232 and 616 m
2
g
-1
. The prepared materials have
distributions of pore sizes that, although broader, in certain
conditions approach those found for well-known MCM-41
or SBA-15 type materials, but with less values of A
BET
and
total mesopore volumes.
Acknowledgments We acknowledge the nancial help from Fun-
dacao para a Ciencia e Technologia (FCT; Portugal) to CQB and
strategic Project PEst-OE/QUI/UI0612/2013 and the grant SFRH/BD/
72058/2010 to ACF.
References
1. Taguchi A, Schuth F (2005) Ordered mesoporous materials in
catalysis. Micropor Mesopor Mater 77:145
bFig. 3 a, b Mesopore-size distributions, by the Broekhoff-de Boer
method, in a version simplied with the FrenkelHalseyHill
equation (BdBFHH), for the samples from the A and B series,
respectively
Fig. 4 Nitrogen adsorptiondesorption isotherms at -196 C for the
sample B1 heated at the indicated temperatures
J Mater Sci (2014) 49:60876092 6091
1 3
2. Wang S (2009) Ordered mesoporous materials for drug delivery.
Micropor Mesopor Mater 117:19
3. Wu Z, Zhao D (2011) Ordered mesoporous materials as adsor-
bents. Chem Commun 47:33323338
4. Perego C, Bagatin R, Tagliabue M, Vignola R (2012) Zeolites
and related mesoporous materials for multi-talented environ-
mental solutions. Micropor Mesopor Mater 166:3749
5. Sing KSW, Haul RAW, Moscou L, Pierotti RA, Rouquerol J,
Siemieniewska T (1985) Reporting physisorption data for gas/
solid systems with special reference to the determination of
surface area and porosity. Pure Appl Chem 57:603619
6. Beck JS et al (1992) A new family of mesoporous molecular
sieves prepared with liquid crystal templates. J Am Chem Soc
114(27):10834
7. Zhao D, Huo Q, Feng J, Chmelka BF, Stucky GD (1998) Non-
ionic triblock and star diblock copolymer and oligomeric sur-
factant syntheses of highly ordered, hydrothermally stable,
mesoporous silica structures. J Am Chem Soc 120:60246036
8. Simpson TL, Volcani BE (1981) Silicon and siliceous structures
in biological systems. Springer-Verlag, New York
9. Deravi LF, Swartz JD, Wright DW (2009) The biomimetic syn-
thesis of metal oxide nanomaterials. In: Challa Kumar SSR (ed)
Nanomaterials for the life sciences: nanostructured oxides.
WILEY, Weinheim, p 507
10. Li N, Zhang X, Wang Q, Wang F, Shen P (2012) Biomimetic
synthesis of silica hollow spheres using poly (L-lysine) and
mechanism research. RSC Adv 2:32883297
11. Vrieling EG, Beelen TPM, van Santen RA, Gieskes WWC (2002)
Mesophases of (bio)polymer - silica particles inspire a model for
silica biomineralization in diatoms. Angew Chem In. Ed
41:15431546
12. Sun Q et al (2003) The formation of well-dened hollow silica
spheres with multilemellar shell structure. Adv Mater 15:10971100
13. Naik RR et al (2003) Controlled formation of biosilica structures
in vitro. Chem Commun. 238239
14. Poulsen N, Sumper M, Kroger N (2003) Biosilica formation in
diatoms: Characterization of native silafn-2 and its role in silica
morphogenesis. Proc Natl Acad Sci USA 100:1207512080
15. Patwardhan SV, Mukherjee N, Steinitz-Kannan M, Clarson SJ
(2003) Bioinspired synthesis of new silica structures. Chem
Commun.:11221123
16. Melanie MT et al (2005) Polypeptide-Templated Synthesis of
Hexagonal Silica Platelets. J Am Chem Soc 127:1257712582
17. Bauer CA, Robinson DB, Simmons BA (2007) Silica Particle
Formation in Conned Environments via Bioinspired Polyamine
Catalysis at Near-Neutral pH. Small 3:5862
18. Patwardhan SV, Clarson SJ (2003) Silicication and biosilici-
cation Part 7: Poly-L-arginine mediated bioinspired synthesis of
silica. J Inorg Organomet Polym 13:193203
19. Cha JN et al (1999) Silicatein laments and subunits from a
marine sponge direct the polymerization of silica and silicones
in vitro. Proc Natl Acad Sci USA 96:361365
20. Hawkins KM, Wang SS-S, Ford DM, Shantz DF (2004) Poly-L-
lysine templated silicas: using polypeptide secondary structure to
control oxide pore architectures. J Am Chem Soc 126:91129119
21. Sun QY, Vrieling EG, van Santen RA, Sommerdijk NAJM
(2004) Bioinspired synthesis of mesoporous silicas. Curr Opin
Solid Mat Sci 8:111120
22. Dos A, Schimming V, Tosoni S, Limbach HH (2008) Acid-base
interactions and secondary structures of poly-L-lysine probed by
15
N and
13
C solid state NMR and ab initio model calculations.
J Phys Chem B 112:1560415615
23. Kroger N, Deutzmann R, Sumper M (1999) Polycationic peptides
from diatom biosilica that direct silica nanosphere formation.
Science 286:11291132
24. Pires J, Pinto M, Estella J, Echeverria JC (2008) Characterization
of the hydrophobicity of mesoporous silicas and clays with silica
pillars by water adsorption and DRIFT. J Colloid Interface Sci
317(1):206213
25. Rozenberg M, Shoham G (2007) FTIR spectra of solid poly-L-
lysine in the stretching NH mode range. Biophys Chem
125:166171
26. Odlyha M, Scott RPW, Simpson CF (1993) The hydroxyl content
of silica gel. J Therm Anal 40:11971212
27. Rouquerol F, Rouquerol J, Sing K (1999) Adsorption by Powders
and Porous Solids. Academic Press, London
28. Gregg SJ, Sing KSW (1982) Adsorption Surface Area and
Porosity. Academic Press, London
29. Lukens WW, Winkel PS, Zhao D, Feng J, Stucky G (1999)
Evaluating pore sizes in mesoporous materials: A simplied
standard adsorption method and a simplied Broekhoff-de Boer
method. Langmuir 15:54035409
30. Barret EP, Joyner LG, Halenda PP (1951) The determination of
pore volume and area distributions in porous substances.
I. Computations from nitrogen isotherms. J Am Chem Soc
73:373380
31. Pires J, Carvalho A, Pinto M, Rocha J (2006) Characterization of
Y zeolites dealuminated by solid-state reaction with ammonium
hexauorosilicate. J Porous Mater 13(2):107114
6092 J Mater Sci (2014) 49:60876092
1 3