Microporous and Mesoporous Materials 363 (2024) 112840

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials

journal homepage: [Link]/locate/micromeso

In-depth study of factors affecting the formation of MCM-41-type

mesoporous silica nanoparticles
Vicente Candela-Noguera a, e, **, María Alfonso a, Pedro Amorós b, Elena Aznar a, c, d, e, f, María
Dolores Marcos a, c, d, e, *, Ramón Martínez-Máñez a, c, d, e, f, ***
a
Instituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat Politècnica de València, Valencia, Spain
b
Instituto Universitario de Ciencia de los Materiales (ICMUV), Universitat de València, Catedrático José Beltrán 2, 46980, Paterna, Valencia, Spain
c
Departamento de Química, Universitat Politècnica de València, Camino de Vera s/n, 46022, Valencia, Spain
d
CIBER de Bioingeniería, Biomateriales y Nanomedicina, Instituto de Salud Carlos III, Spain
e
Unidad Mixta UPV-CIPF de Investigación en Mecanismos de Enfermedades y Nanomedicina, Universitat Politècnica de València, Centro de Investigación Príncipe Felipe,
Valencia, Spain
f
Unidad Mixta de Investigación en Nanomedicina y Sensores, Universitat Politècnica de València, IIS La Fe, Valencia, Spain

A R T I C L E I N F O A B S T R A C T

Keywords: Mesoporous silica nanoparticles (MSN), especially MCM-41 type, have been broadly studied and used in different
Mesoporous silica nanoparticles applications such as catalysis, nanomedicine and environmental science. One of the most advantageous prop­
MCM-41 erties of these materials is their easy and tunable synthesis. Nevertheless, in spite of the efforts made by re­
Synthesis
searchers in the last decades, there is still no complete understanding of the mechanism and factors affecting the
Mechanism
Parameters
formation of MCM-41 type MSN, and some parameters are disregarded in the reported synthesis protocols.
Considering that the fine-tuning of the parameters affecting nanoparticle synthesis is the basis for obtaining well-
defined nanodevices with advanced complex functionalities, an in-depth study of the factors affecting the for­
mation of MCM-41 type MSN is carried out in this work. We analyzed the first stages of MSN formation, namely
the silica-surfactant templating and nanoparticle nucleation and growth, and found that hydrolysis and
condensation rates play a central role in nanoparticle formation. Furthermore, we found that some disregarded
parameters in the MCM-41 type synthesis, such as the stirring strength or the surfactant addition rate, modulate
nucleation and growth, influencing the hydrolysis and condensation rates. Finally, we studied the aging and the
evolution of the MSN by analyzing the consolidation of the silica framework over time. Therefore, we further
elucidate some points about the mechanism and factors affecting the formation of mesoporous silica, which
brings us closer to obtaining reproducible, homogeneous and tailored-made MCM-41 type MSN.

1. Introduction demonstrated [6] as their synthesis is easily tunable, producing nano­

particles with different structures, morphologies, sizes, or surface
In the last decade, mesoporous silica nanoparticles (MSN) have been properties [7,8]. These characteristics make MCM-41 type MSN the most
widely applied to several fields, including sensing [1], catalysis [2] and extensively studied silica particles in biomedicine [9–11] as sensors
biomedicine [3–5] due to their unique properties. Among mesoporous [12–15], and in other advanced applications [16–19], compared to
silicas, one of the most commonly studied is the MCM-41 type because of other mesostructured silica materials, such as worm-like [20],
the ordered framework, narrow pore size distribution, cylindrical dendrimer-like [21], or containing disordered pores.
arrangement of pores, and high specific surface area and volume. In In this context, researchers have pursued the control of particle
addition, the versatility of MCM-41, in the form of nanoparticles, is characteristics, as this allows them to be tailored to the requirements of

* Corresponding author. Departamento de Química, Universitat Politècnica de València, Camino de Vera s/n, 46022, Valencia, Spain.
** Corresponding author. Instituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat Politècnica de
València, Valencia, Spain.
*** Corresponding author. Instituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat Politècnica de
València, Valencia, Spain.
E-mail addresses: vicanno@[Link] (V. Candela-Noguera), mmarcos@[Link] (M.D. Marcos), rmaez@[Link] (R. Martínez-Máñez).

[Link]
Received 16 May 2023; Received in revised form 14 September 2023; Accepted 30 September 2023
Available online 3 October 2023
1387-1811/© 2023 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license ([Link]
nc-nd/4.0/).
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

the application [8,22]. Therefore, understanding the chemistry involved been observed [8,47,64–68], but few explanations has been proposed
in the synthesis of MCM-41 type MSN becomes an important goal [23]. for them; or the silica precursor addition rate.
Nevertheless, despite the efforts made by some groups, there is still no In the case of MCM-41 type MSN, the synthesis procedure is rela­
complete understanding of the parameters affecting the formation of tively standardized and based on the conditions employed by Cai et al.,
MSN, and some parameters are disregarded in the synthesis protocols. in 2001 [37], which remains the most widely used protocol owing to its
According to the literature, the formation of MSN could be divided facile implementation [46]. Yet, this protocol leaves some poorly
into four steps represented in Fig. 1: silica-surfactant micelles formation defined parameters that have been freely interpreted over the last years
(templating), nucleation, growth, and aging [24]. The silica templating and whose modifications are not always well studied, such as we
is the first step in the MSN formation, for which surfactant molecules are mentioned above, the stirring strength or the TEOS addition rate among
used as scaffold. In the case of MCM-41 type MSN, supramicellar cyl­ others.
inders hexagonally ordered are formed. Different models are proposed Taking into account that an understanding of the formation mech­
to explain the formation of mesoporous silica, such as the liquid crystal anism and parameters affecting the synthesis of nanoparticles is the
templating model [25], the cooperative assembly model [26], the basis for obtaining well-defined nanodevices that can exhibit advanced
soft-templating model [27], silica polymerization model [28] or complex functionalities, we report herein a study of the MCM-41 type
swelling-shrinking model [29]. Nevertheless, the most accepted is the MSNs synthesis to further elucidate some points about the mechanism
cooperative assembly model, which describes that the interaction of the and factors affecting their formation. For this, we have taken and
silica moieties with the surfactant molecules is crucial to form the cy­ characterized aliquots during the first stages of MSN formation. In
lindrical micelles, since the surfactant at the concentration used in the addition, we analyzed the formation of MSN as a function of the volume
synthesis does not reach the critical micellar concentration to form cy­ of TEOS added, the stirring conditions and the TEOS addition rate.
lindrical micelles by itself [30–32]. Subsequently, the aggregation of Finally, we have also studied the aging processes during the formation of
silica-surfactant micelles to form initial and small cores provokes the the MSNs. We hope that this study will help researchers to gain a more
nucleation [33]. These aggregations are possible thanks to shielding the detailed picture of the various factors that can affect the synthesis of
electrostatic repulsion between the hydrolyzed silica species when they more reproducible and tailored MCM-41-type MSNs.
bind to the surfactant micelles [34]. These small cores act as nucleation
sites, to which the rest of the micelles bind and where a preferential 2. Experimental section
silica polymerization occurs [35,36]. Then, growth means increasing the
core size by the aggregation or coalescence with other small cores or Materials. Tetraethylorthosilicate (TEOS), 1-hexadecyltrimethy­
simple silica-coated micelles, which align along crystal planes [37]. The lammonium bromide (CTAB) and sodium hydroxide (NaOH) were pur­
growth rate is also associated with the hydrolysis and condensation rate chased from Sigma-Aldrich. Hydrochloric acid 37% (HCl) was provided
of the silica precursor and can be modified by the presence of by Scharlab.
co-solvents, amines, or inorganic compounds [34]. Nanoparticles’ General techniques. Transmission electron microscopy (TEM), N2
growth is halted when the charge density reaches a specific value that adsorption-desorption isotherms, powder X-ray diffraction (PXRD), dy­
prevents the deposition of new silicate species [7] or when the materials namic light scattering (DLS), electrophoretic light scattering (ELS) and
are consumed [31]. Finally, aging means forming and reorganizing silica solid-state nuclear magnetic resonance (ss-NMR) were used to charac­
bonds, strengthening and consolidating the nanoparticle skeleton. This terize the prepared materials. The instruments used were the following:
process is slower than the previous ones. In fact, some studies report that JEOL JEM-1010 microscope for TEM images acquisitions; Bruker D8
complete silicate condensation takes at least one [7] or one and a half Advance diffractometer (Cu Kα radiation) for PXRD measurements;
hours [38]. This is why most synthesis protocols leave the mixture of Micromeritics ASAP 2010 automated analyzer for the recording of N2
silica precursor and surfactant in the reaction vessel for at least 2 h. adsorption-desorption isotherms, samples were degassed at 120 ◦ C in a
Regarding the synthesis of MSN some parameters have been broadly vacuum overnight. The specific surface areas were calculated from the
and deeply analyzed, such as the pH [7,8,23,28,29,31,39–46] and the adsorption data within the low-pressure range using the BET (Bru­
temperature [7,8,47–52], and they clearly influence MSN formation, nauer–Emmett–Teller) model. Pore size was determined following the
modulating the hydrolysis and condensation rates of silica species and BJH (Barrett–Joyner–Halenda) method. Particle size (DLS) and ƺ po­
affecting to the size of the nanoparticles and their mesostructure. Other tential (ELS) in solution were measured by ZetaSizer Nano ZS (Malvern
parameters, such as the silica precursor used [53–56], the silica/­ Instruments Ltd.) equipped with a laser of 633 nm and collecting the
surfactant ratio [22,57], and the concentration of alcohols added as signal at 173◦ . Ss-NMR were performed in Bruker Avance III 400 WB
cosolvents [33,50,58–63] are also commonly studied in the MSN syn­ spectrophotometer.
thesis and their effects on the formation of nanoparticles are clear. Synthesis of nanoparticles. Different conditions were used to
However, there are other parameters whose effect on MSN formation synthesize MCM-41-type nanoparticles, specified in each assay (Results
remain unclear or are unaccounted for. This is the case of the reaction and discussion section). Nevertheless, the assays and synthesis per­
time [7,38,48,51], which has been studied, but its effect remains formed during the work were based on a standard synthesis. The stan­
controversial [7,22,41,45,50]; the stirring strength, whose effects have dard synthesis was established as follows: 1 g of CTAB (2.74 mmol) was

Fig. 1. Schematic representation of the steps involved in the formation of MSN: (a) silica-surfactant micelles formation, (b) nucleation, (c) growth, and (d) aging.

2
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

mixed with 480 mL of deionized water in a 1L (10.5 cm of outer they were briefly spun in a centrifuge (a spin pulse to 8600 rcf) to
diameter) cylindrical beaker. The mixture was heated at 50 ◦ C to remove the presence of large aggregates or sedimentation, which could
dissolve the surfactant and stirred with a cylindrical and plain magnetic disturb the DLS measurements. The average of three measurements was
stir bar (60 mm long and 15 mm diameter) at 500 rpm. Then, 280 mg of calculated and represented. ELS samples were prepared by suspending,
NaOH (7 mmol) dissolved in 3.5 mL of deionized water were added to then generously sonicating the nanoparticles in distilled water (10 mg
reach a pH of 12.16, and the temperature solution was adjusted to 80 ◦ C. NP mL− 1) and then spinning briefly (a centrifuge pulse to 8600 rcf) to
A watch glass was placed to cover the beaker to minimize solvent remove the presence of large aggregates or sedimentation, which could
evaporation and prevent the volume from decreasing during the reac­ disturb the electrophoretic movement measurements of the particles.
tion. When the temperature reached 80 ◦ C, 5 mL of TEOS (25.7 mmol)
were added dropwise (5 mL min− 1) using a syringe and needle. Soon 3. Results and discussion
after, a white precipitate appeared. The mixture was stirred for 2 h. After
this time, the suspension was cooled at room temperature (measured pH 3.1. Templating, nucleation and growth studies
11.75). The solid product was then vacuum filtered with a Buchner over
a filter paper (Whatman™ Grade 3 MM Chr), and deionized water was To analyze the initial stages of the formation of the MCM41-type
added until the suspension reached neutral pH. Then, the solid was dried MSN, some aliquots were taken while TEOS was being added, as
under vacuum and ground to obtain a white powder. After the synthesis, “snapshots”. Beside, a quenching of the reaction was carried out after
nanoparticles were calcined at 550 ◦ C using an oxidant atmosphere for 5 the aliquots were taken to avoid further reaction in the tubes. The ali­
h to remove the surfactant. Calcination was only performed in the last quots were named as “quenched aliquots”, QA. Aliquots were taken just
aging study, since the previous samples were taken as aliquots to be after adding 1, 2, 3, 4 and 5 mL of TEOS in a standard synthesis (see
characterized with TEM and DLS. Experimental Section) procedure (then, taken at 12, 24, 36, 48 and 60 s),
Image analysis. Transmission electron microscopy was performed and the samples obtained were named QA1, QA2, QA3, QA4 and QA5,
to acquire a comprehensive representation of each sample to check the respectively. Besides, an aliquot called A0 was taken before starting the
size and shape of synthesized nanoparticles. At least 100 particles for TEOS addition to analyze the pre-existence of CTAB micelles. The
each sample were counted to obtain a representative size distribution of quenching was performed by diluting 5 times the aliquots with deion­
the particles. In the case of non-spherical nanoparticles, the size was ized water and placing them in an ice bath.
measured in the longer direction. DLS measurements and representative TEM images can be seen in
DLS measurements. DLS measurements on aliquots samples were Fig. 2 (and Fig. S1 and Fig. S2, Supplementary Information). In the case
performed directly on the corresponding aliquots samples or by diluting of the A0 sample, DLS reveals a peak of ca. 5.5 nm, attributed to CTAB
them with distilled water. The rest of the DLS samples were resuspended micelles, which are confirmed in TEM images, despite the low electron
and generously sonicated in distilled water (1 mg NP mL− 1). After that, density of organic matter. TEM images show worm-like aggregates of

Fig. 2. Size distribution according to Intensity PSD analysis and TEM images of QA1, QA2, QA3, QA4 and QA5.

3
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

small dots of about 5–6 nm (Fig. S1, Supplementary Information). These growth processes are very quick, with fully formed nanoparticles being
images are similar to previously reported TEM observations of CTAB observed even before 5 min of reaction. Besides, in the first aliquots
solutions [69–71]. For QA1, two DLS peaks are observed at 5 nm and (QA1 and QA2), we can find CTAB micelles and silica-surfactant com­
370 nm. It is suggested that the former corresponds to CTAB micelles (or plexes of ca. 6 nm (in DLS curves), which are consumed as the reaction
single silica-surfactant complexes) and the latter is suggested to corre­ progresses, since they are not measurable in the following aliquots.
spond to large entities such as CTAB crystallizations or aggregates of Additionally, applying higher magnification in the QA TEM images
amorphous silica, which are also observed in the TEM images. For QA2, (Fig. S2, Supplementary Information), the typical 2D hexagonal
DLS shows a population at ca. 6 nm (presumably corresponding to CTAB arrangement of pores were shown as both hexagonal pores or parallel
micelles), and peaks at 32 nm and 155 nm. In the TEM images, small cylindrical channels (in QA3, QA4 and QA5). Even in the case of QA2,
nanoparticles still in formation can be seen, sized at ca. 28–30 nm. In when the particles are only small seeds, a quasi-hexagonal arrangement
addition, some larger particles are observable in additional TEM images of pores was observed. These observations demonstrate that the hex­
of QA2 (Fig. S2, Supplementary Information), whose sizes vary from 30 agonal arrangement and templating of silica already occur at the initial
to 300 nm, which correspond to the larger population in DLS. DLS curve stages of the process, in contrast to the observations made by Sadasivan
of QA3 displays two evident populations at 38 nm and 205 nm. TEM et al. [27], who described the formation of disordered silica-surfactant
images show a first population corresponding to small mesoporous micelles when small nanoparticles (seeds) are formed. Therefore, as
nanoparticles in formation, and a second one formed by particles in the TEOS is added, the nanoparticles grow from seeds via the deposition of
range from 150 to 400 nm. QA4 and QA5 display single populations at new surfactant-silica micelles, which increase the number of cylindrical
114 nm and 132 nm in DLS analysis, respectively. TEM images showed micelles arranged hexagonally and elongate them.
for QA4 and QA5 populations of nanoparticles already formed of ca. 88
± 8 nm and 98 ± 12 nm, accordingly.
3.2. The role of TEOS amount in nucleation and growth of MSN
The results show that the nanoparticles are forming and growing
progressively, from the 6 nm CTAB micelles, then seeds of 30 and 40 nm
The study of quenched aliquots (QA) above allowed us to follow the
approximately, to finally fully formed nanoparticles of ca. 90–100 nm.
progress of the formation of the particles throughout the synthesis
Some bigger size minority populations are formed and can be observed
procedure, however, two simultaneous parameters were modified: the
in DLS and TEM in some samples, such as QA2 (DLS population at 155
amount of TEOS added, and the reaction time. In order to establish the
nm) and QA3 (DLS population at 205 nm) (see sample QA2 and QA3 in
individual role of the TEOS availably in the formation of the nano­
Fig. S2, Supplementary Information). It has to be remarked that In­
particles, we decided to study the samples obtained when different
tensity PSD analysis overestimates the larger populations in DLS curves.
amounts of TEOS are added to the reaction but, in this case, allowing the
All these observations suggest that templating, nucleation and
mixture to proceed until the end of the standard reaction procedure (2

Fig. 3. Size distribution according to Intensity PSD analysis and TEM images of SA1, SA2, SA3, SA4, SA5, and SA10. * DLS measurements of SA10 cannot performed
due to the great presence of agglomerates in the sample.

4
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

h). Therefore, we synthesized nanoparticles following the standard achieved, which implies i) the condensation of silica monomers to form
conditions, but adding 1, 2, 3, 4, 5 (standard volume) and 10 mL of oligomers, ii) the silica oligomers cooperate with surfactant to form
TEOS. After stirring for 2 h, aliquots were taken and characterized. The cylindrical micelles, iii) several spherical and cylindrical micelles
aliquots were named SA1, SA2, SA3, SA4, SA5, and SA10 (“stirred al­ nucleate to form seeds, and iv) the consumption of silicate monomers
iquots”, SA), respectively. until falling below the self-nucleation threshold. This phase corresponds
The aliquots were analyzed by DLS and TEM (Fig. 3 and Fig. S3, with QA2 and partially QA3. Then, starts the third phase, the silica
Supporting Information). DLS and TEM images of SA1 are similar to condensation is predominant and it preferentially deposit into already
those of QA1 (vide ante). This fits with previous results in which 1 mL of existing nuclei, which promotes the growth of formed seeds into nano­
TEOS is not enough to produce nanoparticles but only small silica- particles (corresponding to QA4 and QA5). The growth is stopped when
surfactant complexes. Regarding SA2, SA3, SA4 and SA5, clear, sin­ the concentration of silicate monomers decreases until the critical su­
gle, and narrow populations can be observed on DLS curve, centered at persaturation level, below which they remain dissolved in equilibrium
458, 135, 132, and 134 nm, respectively. SA10 cannot be measured as monomers.
properly in DLS due to the great amounts of agglomerates in the sample. On the other hand, it can be claimed that the lower amount of TEOS
TEM pictures show that SA2 material consists of sub-micron mesoporous in MSN synthesis (in SA2 and SA3) limits and slows down the reaction
particles of ca. 401 ± 99 nm. In the case of SA3, SA4 and SA5, TEM progress, that is, the number of TEOS molecules hydrolyzed and silicate
images show standard MCM41-type MSN, being somewhat larger in the monomers polymerized per time. Considering both phenomena (LaMer-
case of SA3. What regards to SA10, TEM pictures show MCM41-type based growth and TEOS amount limitation), we can hypothesize that,
MSN with a layer of amorphous silica in the nanoparticles’ surface, when smaller amounts of TEOS are added (e.g. 2 or 3 mL), the propor­
whose thickness is 6.8 ± 2.2 nm (Fig. S4, Supplementary Information). tion of silicate monomers above this self-nucleation threshold is lower.
The sizes of the nanoparticles from TEM are 110 ± 22 nm in SA3, 94 ± This means that the number of nucleation seeds formed are lower, and
12 nm in SA4, 95 ± 14 nm in SA5, and 119 ± 18 nm in SA10. this gives, in turn, larger particles, since the silicate monomers available
The most striking fact of this assay is the significant increase in the after the nucleation are driven to grow the existent seeds but not to form
size of the particles in sample SA2. We can also observe an increase, but new ones. In contrast, according to our observations, when higher
to a lesser extent, in SA3 compared to the other samples (SA4 and SA5). concentrations of TEOS are added (SA10), the nanoparticles formed are
While 1 mL of TEOS (SA1) is not enough to nucleate and form MSN, not smaller, but larger. Thus, we hypothesize that the hydrolysis and
when 2 mL of TEOS were added (SA2), the nucleation and growth condensation progress practically at maximum rates, that is, the ratio of
processes are boosted, and large nanoparticles are formed. Other au­ silica monomers formation per time reaches its maximum value.
thors have also observed this phenomenon [72], but no explanations Therefore, the proportion of silicate monomers above the self-nucleation
were given. It is reported that tuning pH, temperature or solvent pro­ threshold does not increase, leading to a higher growth, instead a higher
voke noticeable changes in nanoparticles’ size, giving the influence that nucleation. Furthermore, the surface layer of amorphous silica formed in
these factors have on the hydrolysis and condensation rates and the SA10 suggests the consumption of surfactant micelles that act as
silica solubility. However, in our case the studied samples (SA) are template.
synthesized in the same conditions in terms of pH, temperature and To reinforce these results, we decided to determine the time it takes
solvent. Therefore, we hypothesize that the larger size of nanoparticles for the reaction mixture to turn turbid white once different amounts of
in SA2 and, to a lesser extent in SA3, can be explained considering both TEOS are added. The emergence of a white suspension have been re­
the LaMer-based growth mechanism [73,74] and the influence of the ported by several authors in the range of ca. 1–10 min [7,34,60,72,
TEOS amount on the hydrolysis and condensation rates. According to 75–79]. The cloudy appearance is described to correspond with the
LaMer-based growth mechanism and our observations, the formation of seeds’ growth and formation of nanoparticles that reach a certain size
nanoparticles can be divided into three phases (Fig. 4). The first phase threshold [80,81], which some authors set at 25 nm [82]. Then, this
corresponds to the lag time [74], in which hydrolysis of silica precursor time can be correlated with the end of nucleation step and the beginning
is predominant, and therefore the concentration of silica monomers per of growth step. The volumes tested in this case were 1, 1.5, 1.75, 2, 3, 4,
time increases in parallel to the hydrolysis rate. The condensation is 5, 7.5, and 10 mL of TEOS. Considering previous observations, we
negligible, so no nucleation or growth takes place. This is because, on decided to instantly add the precursor (instead of 5 mL min− 1) to
the one hand, the self-nucleation threshold is not reached and, on the eliminate the TEOS addition rate as a variable and limiting factor. The
other hand, there are not nucleation sites over which silica monomers results show that the time required to obtain a white suspension in­
can deposit and polymerize. In this phase, silica monomers form creases as the amount of TEOS decreases (Fig. 5). This increasing is
spherical surfactant-silica micelles (sized at ca. 5–6 nm) and correspond especially steep when the amount of TEOS is lower than 2 mL, although
with QA1. The second phase starts when the self-nucleation threshold is the time interval in which the transition from translucent to cloudy
appearance occurs is broad and less clear. In contrast, when at least 2 mL
of TEOS are added this interval is narrower. On the other hand, it should
also be noted that 1 mL of TEOS was not enough to give a cloudy sus­
pension. Finally, it is also observable that, at the studied conditions, the
time it takes the reaction to turn cloudy reaches it minimum value from
7.5 mL (t = 31 s).
These results reinforce the hypothesis of LaMer-based growth and
the influence of the TEOS amount in the hydrolysis and condensation
progression. Therefore, the behavior of the reaction when different
amounts of TEOS are added can be explained, considering the TEM
images and the time it takes the reaction to turn cloudy (Fig. 6). As Fig. 6
indicates, when lower amounts of TEOS are added, the slope in which
the silicate monomers appear is shallower, the time required to reach the
growth step are higher, and then, the cloudy appearance emerges later.
Fig. 4. A) LaMer-based growth model proposed for MCM41-type MSN. Dotted In contrast, above 5 mL the time required is very similar (t = 36 s in the
lines divide the three phases. Dashed lines indicate the self-nucleation threshold case of adding 5 mL of TEOS and t = 31 s in the case of 7.5 and 10 mL),
and the critical supersaturation level. Gray area means the phase in which the which suggest that the slope barely increases. Therefore, it can be
suspensions appears cloudy. claimed that the hydrolysis and condensation rates reach their

5
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

samples were prepared by modifying the stirring rate (350 rpm, 500
rpm, 900 rpm and 1200 rpm). The samples named MSN-350 rpm, MSN-
500 rpm (corresponding to the standard synthesis), MSN-900 rpm and
MSN-1200 rpm were characterized by TEM and DLS. According to TEM
pictures (Fig. 7), the diameter of the nanoparticles decreases with
increasing stirring rates until reaching a minimum value of around 80
nm when using 900 rpm (MSN-350 rpm averaged at 112 nm, MSN-500
rpm at 89 nm, both MSN-900 rpm and MSN-1200 rpm at 80 nm). In
addition, it is also observable that the morphology of the nanoparticles
becomes more irregular when the stirring speed increases. Thus, MSN-
350 rpm and MSN-500 rpm tend to have a smooth surface and a
spherical shape, but MSN-900 rpm and especially MSN-1200 rpm are
poorly formed and their surface is rough. TEM images also indicate that
higher stirring rate leads to a higher number of fusions between nano­
particles. The tendency observed in DLS curves (Fig. S5, see Supple­
mentary Information) is in line with that found in TEM images, except
for MSN-1200 rpm in which a slightly increase in the diameter is
observed (i.e. MSN-350 rpm centered at 175 nm, MSN-500 rpm at 162
nm, MSN-900 rpm at 147 nm and MSN-1200 rpm at 168 nm).
Fig. 5. Representation of the time required for the reaction mixture to become
In addition, the time taken from the reaction to become cloudy was
cloudy, measured as a function of the amount of TEOS added. Gray area means
the time interval in which the transition from translucent to cloudy appear­
also measured, this time by tuning the stirring speed. On this occasion,
ance occurs. the amount of TEOS added was 3 mL (at 5 mL min− 1), considering that,
in standard conditions, the white turbidity was reached when 4 mL of
TEOS were added and this could mask the effect of tuning the stirring
rate. For these assays, the stirring was adjusted to 350, 500, 900, and
1200 rpm. It can be observed that the time required to appear the white
turbidity decreases as the stirring rate increases when 3 mL of TEOS are
added (from 95 s at 350 rpm to 26 s at 1200 rpm) (Fig. 8).
Considering the previous results, it can be stated that the effect of
stirring speed on the size and morphology of MSN is through the influ­
ence on the rate of hydrolysis and condensation rate, since the time
taken for the reaction to become cloudy is also affected. We can suggest
that at higher stirring speeds, the oil drops of TEOS formed after being
added to the aqueous solution are split due to the high radial flow and
the shearing forces derived. This provokes the surface of the interface
between TEOS oil drops and aqueous solution to increase, as it has been
previously described [29], and thus, more TEOS can be hydrolyzed,
since hydrolysis occurs in this interface [68]. The promotion of hydro­
lysis rate increases the slope in which silicate monomers appear per
time, and then, their number above the nucleation threshold. As we
suggested before, this explains the formation of higher amounts of seeds
and, in turn, smaller nanoparticles. In contrast, when lower stirring rates
Fig. 6. LaMer-based growth mechanism proposed for explaining the results
obtained when 2 mL (dark green), 3 mL (light green), 4 mL (yellow), 5 mL are applied, the interface surface between oil and aqueous phase is
(orange) and 10 mL (red) of TEOS are added instantly. Arrows indicate the time lower, which leads to a decrease in the hydrolysis rate, the formation of
in which the formed seeds start to grow. less seeds and larger nanoparticles. A similar correlation between stir­
ring rate and nanoparticles size was reported by Yokoi et al. for
maximum values at least from 7.5 mL of TEOS addition. non-porous silica nanospheres [68]. Moreover, the increase in fusions
between nanoparticles (observed by TEM and also suggested by DLS
measurements) and the deformation of their shape (observed by TEM)
3.3. Influence of stirring strength on MSN formation with increasing stirring speed (especially in MSN-900 rpm and
MSN-1200 rpm) could be the consequence of turbulence and chaotic
As we mentioned and it is widely described, the MSN formation is behavior. Under these conditions the collisions between cores are more
mainly conducted by hydrolysis and condensation rates. These reactions common, the fluid dynamics becomes inhomogeneous and phenomena
are principally modulated by the pH, the temperature, and the solvent. such as swirling and cavitation appear. In this line, it has been described
These parameters are fairly studied. However, there are other dis­ that cavitation can disturb even the mesoporous geometry and order
regarded parameters whose influence on the hydrolysis and condensa­ [64]. In addition, high stirring speeds can lead to less controllable hy­
tion rates are less studied. One of these parameters is the stirring rate. drolysis and condensation steps, increasing fusions and uneven nano­
The stirring rate has been reported to modulate the size of nanoparticles. particle growth.
At low stirring speed, large-sized nanoparticles are formed [47,64]. Since the radial flow depends not only on the stirring speed but also
Meanwhile, the increase in stirring speed leads to the formation of on the shape of the magnetic stirrer, we have also studied in detail the
smaller nanoparticles [47,65–68]. In addition, strong stirring rates later effect. A sample called MSN-egg was synthesized using an egg-
provoke morphological regularity deterioration due to shearing forces shaped stir bar at 500 rpm (see set up in Fig. S6, Supplementary Infor­
originated [8]. However, these studies are sometimes incomplete and no mation). The nanoparticles obtained were compared with those ob­
explanation has been given to establish a correlation between the stir­ tained with the MSN-500 rpm conditions synthesized using a
ring strength and the nanoparticles size in the case of MSN. cylindrical-shaped stir bar. We observed that the stirring force achieved
To improve our knowledge about this phenomenon, different with the egg-shaped magnetic stirrer is lower than with the cylindrical-

6
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

Fig. 7. Representative TEM images of MSN-350 rpm, MSN-500 rpm, MSN-900 rpm, and MSN-1200 rpm. In addition, histograms with particle size distribution of
the samples are also shown.

influence in the formation of the nanoparticles. At this point, it is

essential to note that stir bar characteristics or stirring details are not
usually reported in the literature despite their proven importance. This
issue must be especially cared for industry production, in which the set
up and scale up are not trivial matters.

3.4. TEOS addition rate effect

Preparation recipes to obtain MCM-41 nanoparticles normally

describe the TEOS addition rate as “drop by drop”, “dropwise” or
similar. Moreover, there are some works that reported a specific rate,
such as 0.1 [50], 0.25 [83], 0.38 [40,84], 0.5 [45], 1 [46,47,64,85], or
1.45 mL min− 1 [48]. However, most of these works did not explain the
criteria to establish this choice. Some authors, such as Chiang et al.,
considered the addition rate of TEOS to have a neglected effect on the
morphology and structure of MSN [45]. On the contrary, Moon­
graksathum et al. defended that the rate of TEOS addition, along with
Fig. 8. Representation of the time required for the reaction mixture to become the stirring speed and concentration of surfactant, influences the char­
cloudy, measured as a function of the stirring rate. acteristics of MSN [86]. Oliveira et al. [87] reported a synthesis in which
TEOS was added in two steps and tested the second addition in
shaped one. Therefore, as expected, the samples stirred with an egg- controlled dropwise or not controlled. The authors observed differences
shaped magnetic stirrer (MSN-egg) are larger than the ones stirred between both conditions, such as the formation of a bimodal distribution
with the cylindrical-shaped stirrer at the same speed (MSN-500 rpm) in particle size, variations in relative surface area and differences con­
and rather similar to those obtained at a lower speed (MSN-350 rpm) cerning the aggregation of nanoparticles. In another study, Lechevallier
when using a cylindrical stirrer. The average diameter of the nano­ et al. [83] obtained a quite polydispersed sample when TEOS was added
particles measured for MSN-egg using TEM images is 112 ± 19 nm, and at 0.25 mL min− 1. He et al. [84] added TEOS at 0.38 mL min− 1, and the
their hydrodynamic diameter by DLS is centered at 183 nm (Fig. S6, nanoparticles were highly fused and the relative surface area was low
Supplementary Information). (ca. 860 m2 g− 1) compared to standard MCM-41 type MSN.
In summary, it can be pointed out that the stirring strength affects the Aiming to improve the understanding of the implications of the TEOS
MSN formation in different manners. On the one hand, it can vary the addition rate, we have synthesized MSN adding 5 mL of TEOS at 0.25
size of the nanoparticles, due to the lower or higher homogenization of mL min− 1, 1 mL min− 1, 5 mL min− 1 and instantly. The samples were
the reaction mixture, which modulates the amount of TEOS available to named MSN-0.25, MSN-1, MSN-5, and MSN-instant, respectively.
be hydrolyzed and, in turn, the nucleation and growth process. On the The different samples obtained were characterized by DLS and TEM.
other hand, it can drive to different regularity of their morphology and TEM images of the synthesized samples are shown in Fig. 9. MSN-0.25
surface, and promote the fusion and coalescence between nanoparticles. displays a very broad size distribution for the MSN. Specifically, three
This can be produced due to the turbulence, chaotic behavior, cavita­ differentiable populations of MSN are observed: i) spherical nano­
tion, and swirling phenomena, when the reaction mixture is stirred particles in the 70–120 nm range; ii) rod-like nanoparticles with the long
vigorously. Thus, the control of the stirring during the MCM41-type axis ranging from 140 nm to 250 nm and the short axis from 90 nm to
MSN synthesis is crucial for controlling the characteristics of the nano­ 130 nm; and iii) bean-like nanoparticles, which were a kind of bent rods,
particles [74]. Furthermore, it is important to control not only the stir­ whose diameters are from 250 nm to more than 400 nm. MSN-1 shows
ring speed, but also the magnetic stir bar’s size and shape, as they also two different populations: i) spherical nanoparticles from 70 to 120 nm,
analogues to the spherical ones in MSN-0.25, and ii) rod-like

7
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

Fig. 9. TEM images of MSN-0.25, MSN-1, MSN-5, and MSN-instant. In addition, histograms with particle size distribution (from TEM images) and MSN morphology
indications are showed.

nanoparticles, whose long axis ranges from 140 to 230 nm (a bit smaller turbidity (from 36 s for instantaneous addition to ca. 600 s when adding
than rods found on MSN-0.25), and short axis in the 90–120 nm range. TEOS at 0.25 mL min− 1).
In the case of MSN-5 and MSN-instant, only one population of spherical The results indicate that the rate of TEOS addition also influences the
nanoparticles was found, in the range from 60 nm to 130 nm (centered rate of hydrolysis and condensation, as well as the nucleation and
at 94 ± 15 nm), and 50 nm–120 nm (centered at 85 ± 16 nm), respec­ growth of nanoparticles, by means of restricting the availability of TEOS
tively. Besides, nanoparticles from MSN-5 have a more regular and at a given time. It is observed that the lower the amount of TEOS
smoother surface than the ones in MSM-instant, for which some available, the slower the formation of nanoparticles, following the same
roughness can be seen. All the populations present in TEM images visible explanation that found at different volume of TEOS added or stirring
mesoporosity. Considering fusions between nanoparticles, MSN-0.25 speeds (vide ante). This is especially observable when low addition rates
and MSN-1 materials show few visible linkages, whereas MSM-instant are used, such as in MSN-0.25, in which the three populations observed
clearly shows bridges between nanoparticles and coalescing nano­ correspond to three different growth phases. These results can also be
particles. DLS curves of the samples are displayed in Fig. S7, Supporting explained according to the LaMer-based growth mechanism, consid­
Information. ering the progression of three differentiated nucleation cycles (Fig. 11).
The time it takes the reaction to become cloudy was also measured. 5 This can be described as follows. Firstly, spherical nanoparticles are
mL of TEOS were added at 0.25 mL min− 1, 0.5 mL min− 1, 1 mL min− 1, 5 formed from an initial seed formation, in which reaction turn cloudy (at
mL min− 1 and instantly. The results are reported in Fig. 10. It is observed ca. 630 s). Secondly, when spheres are formed, and the silica continues
that the slower the addition rates, the longer it takes to observe white depositing over them, the nanoparticles grow mainly in the axis parallel
to the channels, forming rod-like nanoparticles. These can be defined as
extended spheres, in which the smaller axis is similar to the diameter of
the spheres, and the long axis coincides with the direction of the CTAB
templated channels. In parallel, remaining surfactant-silica complexes
aggregate as TEOS is added and a second nucleation step starts, forming
additional spherical nanoparticles. Thirdly, as the reaction goes by, rod-
like nanoparticles from the previous step can still accumulate more silica
and grow. When they reach approximately 250 nm on the long axis,
their structure begins to bend and bean-shaped nanoparticles appear.
The spherical nanoparticles formed at second nucleation seed formation
can also suffer additional growth and acquire rod morphology. In this
latter stage, a third additional formation of seeds occurs, because new
silica precursor molecules are added to the suspension until the 1200 s.
Thus, new spherical nanoparticles are formed.
Regarding MSN-1, the process can be seen as similar to that observed
in MSN-0.25, but the third step does not occur. In this case, the addition
rate was faster and therefore the availability of TEOS was higher from
the beginning. At initial stages, a greater amount of nucleation seeds is
formed; hence, the added silica precursor is not enough to form bean-
shaped nanoparticles. In other words, since most of the TEOS has
Fig. 10. Representation of the time required for the reaction mixture to become
already been used to grow the higher number of nuclei, a new nucleation
cloudy, measured as a function of the TEOS addition rate. Gray area means the cycle does not occur, as the nucleation threshold is not reached. MSN-5
time interval in which the transition from translucent to cloudy appear­ and MSN-instant only produce the spherical population as the addition
ance occurs. rate of silica precursor is fast enough to form a sufficient number of

8
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

Fig. 11. LaMer-based growth mechanism proposed for explaining the sequential nucleation in the case of MSN-0.25. Arrow indicates the estimated time when it
takes the reaction to turn cloudy.

nucleation seeds to use all the silica available. Since no additional TEOS MSN-0.25 (Fig. S8, Supplementary Information).
is added once the nanoparticles are formed, the hydrolyzed silica species Therefore, both stirring rate and TEOS addition rate influence the
are distributed homogeneously among the nuclei to form monodisperse amount of TEOS available to be hydrolyzed and then the number of
nanoparticles. nuclei formed. The formation of a few nuclei in the initial stages leads to
Bean-shaped nanoparticles formation was also reported by Beltrán- the formation of large nanoparticles. However, the progressive emer­
Osuna et al. [64] when they applied a very low stirring speed and a TEOS gence of new silica precursor monomers from hydrolyzed TEOS, both
addition rate of 1 mL min− 1. To further check the influence of low due to the slow addition rate of TEOS or the low stirring speed, can lead
stirring speed in the formation of bean-shaped nanoparticles, we syn­ to new nucleation cycles and a multimodal population of nanoparticles.
thesized and characterized a sample at standard conditions (TEOS added
at 5 mL min− 1), but stirred at 250 rpm (MSN-250 rpm). TEM images of
MSN-250 rpm (Fig. S8, Supplementary Information) also show the 3.5. Aging and reaction time
presence of three populations found in MSN-0.25: spherical,
rod-shaped, and bean-shaped nanoparticles. DLS curve is also similar to The first steps in the synthesis of the nanoparticles (surfactant-silica
micelles formation, nucleation and growth) occur in less than 10 min, as

Fig. 12. Size distribution according to Intensity PSD analysis, ζ potential value and TEM images of MSN-5min, MSN-20min, MSN-1hour, MSN-2hours and MSN-
4hours. In addition, histograms with particle size distribution (from TEM images) and ζ potential values.

9
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

we have demonstrated in the previous sections and other authors also to − 28.4 mV for MSN-4hours.
described [29,88]. This means that the additional time left for the re­ From these data, it could be affirmed that 5 min of aging are enough
action corresponds to an aging process. To study this more in detail, we to obtain samples similar to common MSN in size and morphology. In
have characterized several samples after the complete addition of TEOS fact, the evolution of nanoparticles from 5 min to 2 h (the standard re­
(5 mL) following a standard synthesis, at different reaction times (5 min, action time) barely supposes an increase in the size of the nanoparticles.
20 min, 1 h, 2 h and 4 h), which were named MSN-5min, MSN-20min, These results disagree with Catalano et al. [7] studies, which suggest an
MSN-1hour, MSN-2hours and MSN-4hours, respectively. TEM images, increase from 108 nm after 7 min to 159 nm after 2 h. We observe in
DLS and ELS, Si29-ssNMR, PXRD and nitrogen adsorption-desorption TEM and DLS results that the aging step after the growth of the nano­
studies were performed to characterize the nanoparticles. This study is particles is necessary to reorganize and level the size between nano­
similar to the one performed by Catalano et al. [7], who characterized particles. This can be done by a mass transfer to larger nanoparticles
samples at 7 min, 30 min, 1 h, 1.5 h and 2 h. Nevertheless, we obtained which grow at the expense of the dissolution of smaller nanoparticles; a
different conclusions compared with these authors, as mentioned phenomenon known as Ostwald ripening [89].
throughout this section. Additionally, the measurement of ζ potential allows for monitoring
TEM pictures (Fig. 12) of calcined samples, show completely formed the progression of the reaction. Since the negative charge of calcined
nanoparticles. However, the samples obtained after shorter reaction MSN comes from silanolate groups, the decrease of ζ potential means the
times (MSN-5min and MSN-20min) show nanoparticles with an irreg­ progressive condensation of these groups. Therefore, the degree of
ular surface. Nanoparticles range from ca. 90–100 nm in TEM images as condensation of the nanoparticles is expected to be higher in those
histograms of distribution size show (Fig. 12). DLS measurements of samples in which the reaction time is longer and vice versa. These ob­
calcined samples were also performed (Fig. S9, Supplementary Infor­ servations agree with some authors [7,38] that reported that some hours
mation). MSN-5min shows two populations at 185 and 702 nm, MSN- are required to complete the condensation process. Still, they are also in
20min one broad population at 148 nm, and MSN-1hour, MSN-2hours conflict with the results of Catalano et al. [7], which did not find a
and MSN-4hours narrow distributions at 159 nm, 158 nm and 174 nm, correlation between the reaction evolution and the ζ potential value.
respectively. TEM images and DLS measurements show that the size This can be explained by the differences in the synthesis process as
distribution becomes narrower from 5 min to 2 h but increasing their Catalano et al. washed the nanoparticles with ethanol and extracted the
breadth at 4 h. ELS measurements of calcined samples show a lower surfactant by refluxing with ethanol-HCl.
negative ζ Potential over time (Fig. 12), from − 35.7 mV for MSN-5min We also studied as-made and calcined nanoparticles by powder X-ray

Fig. 13. A) PRXD patterns of as-made and calcined samples. Dashed lines indicate the position of peak (100) in the cases of MSN-4hours as-made and calcined. B) N2
adsorption-desorption isotherms of calcined samples (MSN-5min, MSN-20min, MSN-1hour, MSN-2hours, and MSN-4hours). Inset: magnification of capillary
condensation region.

10
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

diffraction techniques (PXRD), and the results are shown in Fig. 13A and Measurements of isothermal adsorption-desorption of nitrogen were
summarized in Table 1. Diffractograms of the as-made materials are also performed to characterize the pore size, pore volume, and surface
similar in all the samples (from MSN-5min to MSN-4hours) in terms of area (Fig. 13B and Table 2). The isotherms obtained are similar in all the
both the position and the broadness of their peaks. Therefore, this fact samples studied. They all exhibit a type IV isotherm typical of meso­
suggests that the supra-micellar hexagonal structure is formed already porous materials. The main difference observed is that the sharp jump at
after 5 min with the same characteristics as the standard synthesis intermediate partial pressures, corresponding to the capillary conden­
((100), (110), (200) and (210) reflections are present). The rapid for­ sation region, shifts to higher pressures as the synthesis time increases
mation of the hexagonal structure is already reported by other authors (from 0.28 p/po in MSN-5min to 0.32 p/po in MSN-4hours). This ten­
[88]. The main peak, corresponding to the (100) reflection in these dency can be seen in Table 2, in which the pore size increases progres­
samples, falls in the narrow range of 2.09–2.14◦ , corresponding to a unit sively with the reaction time (from 2.65 nm in MSN-5min to 2.88 nm in
cell ca. 4.76–4.88 nm. Furthermore, the full width at half maximum MSN-4hours). In the case of specific surface area, a decrease was
(FWHM) calculated for this peak ranges from 0.22 to 0.24. In the case of observed oppositely to the reaction time (from 1157 m2 g− 1 in MSN-
the calcined nanoparticles, the peaks corresponding to the reflections 5min to 1082 m2 g− 1 in MSN-4hours). Regarding the pore volume, no
(100), (110), and (200) are still observable, but (210) does not, due to clear tendency is observed, and it seems to be constant in the different
the partial loss of long-range ordering typical after calcination. In this samples (in the range from 0.97 to 1.01 cm3 g− 1). In addition, consid­
context, there is greater variability in the position and width of the peaks ering the PXRD and N2 isotherms results, we have calculated the wall
compared with the as-made samples. Moreover, the peak (100) is found thickness for the different materials by subtracting the pore size value
progressively at a lower diffraction angle with the progression of the from the unit cell value, but no clear tendency was observed as a func­
reaction (from 2.45 in MSN-5min to 2.36 in MSN-4hours). On the other tion of the reaction time.
hand, the broadness of the peak (100) (FWHM) is greater in the samples The results observed in N2 adsorption-desorption isotherms fit with
with a lower reaction time (0.32 and 0.30 for MSN-5min and the expected tendency in accordance with PXRD data. Then, the
MSN-20min, respectively) and smaller when the reaction time increases consolidation of the structure over reaction time reduces the shrinkage
(0.28, 0.26 and 0.28 for MSN-1hour, MSN-2hours and MSN-4hours, of the mesostructure. As a consequence of that, we find that the pore size
accordingly). Comparing as-made and calcined samples, the unit cell is larger as the synthesis time increases. Additionally, the decrease in
contraction appears to be lower as the reaction synthesis progresses and specific surface area in the samples with longer aging can also be an
vice versa. Thus, in MSN-5min and MSN-20min, the unit cell shrinks indication of the consolidation of their structure, which suffers fewer
0.65 and 0.68 nm, respectively, while in MSN-1hour, MSN-2hours and defects and nooks in the surface, in contrast to those samples stirred only
MSN-4hours the unit cell contracts 0.52, 0.49 and 0.49, correspond­ for 5 or 20 min. On the other hand, the minor variations in the wall
ingly. Besides, the ratio between the FWHM of the (100) peak after and thickness calculated for the different materials suggests the reaction
before calcination behaves in the same vein, being higher in MSN-5min time is not affecting this parameter. The changes in the unit cell
and MSN-20min samples (1.45 and 1.30) and smaller in MSN-1hour, measured in the PXRD analysis correspond with the decrease in pore
MSN-2hours and MSN-4hours ones (1.17, 1.18 and 1.27). size, which is observed in the N2 adsorption measurements.
As we observed in previous results, the PXRD diffractograms show To reliably verify the degree of condensation, the samples MSN-
that the porous hexagonal arrangement is formed from the first minutes 5min, MSN-1hour, MSN-2hours and MSN-4hours were analyzed by
of reaction. This is maintained along the reaction time in both as-made Si29-ss-NMR (Fig. 14). The NMR spectra of the samples (Fig. 14C) show 3
and calcined samples. Once again, these results dissent from the study of peaks at ca. 90, 100 and 110 ppm, corresponding to (HO)2Si-(OSi)2 (Q2),
Catalano et al. [7] as they observed unresolved peaks in as-made early
samples due to a disruption in the structure, which ethanol washing and
surfactant extraction processes may provoke. Table 2
Moreover, our samples show the typical shift of the diffractogram Mesoporosity of samples as a function of their reaction time.
peaks to higher 2θ values (reduction in d-spacing) and the loss of long- sample BET surface BJH pore volume BJH pore wall thickness
range ordering in calcined samples, which is related to the unit cell (m2 g− 1) (cm3 g− 1) size (nm) (nm)
contraction or shrinkage provoked by the calcination process. Our re­
MSN- 1157 0.97 2.65 1.51
sults suggest that cell contraction and structure disruption are inversely 5min
correlated with reaction time and aging. In this sense, when weak MSN- 1131 0.98 2.73 1.40
structures such as MSN-5min and MSN-20min are calcined, they tend 20min
MSN- 1134 1.01 2.71 1.53
to suffer more significant contraction and disorder of the mesostructure
1hour
[57,65], which is reflected in larger peaks shifts (unit cell contraction) MSN- 1101 0.98 2.84 1.55
and broadening of the peak (100) (FWHM ratio). By contrast, the sam­ 2hours
ples MSN-1hour, MSN-2hours and MSN-4hours are more consolidated MSN- 1082 0.98 2.88 1.44
and suffer less shrinkage and structure disruption. 4hours

Table 1
PXRD measurements of samples as a function of their synthesis time.
As made Calcined Compared data

Sample Peak [100] Unit cell FWHM peak Peak [100] Unit cell FWHM peak Peak [100] shift Unit cell contraction FWHM
(2θ) (nm) [100] (2θ) (nm) [100] (2θ) (nm) ratio
MSN-5min 2.12 4.81 0.22 2.45 4.16 0.32 0.33 0.65 1.45
MSN- 2.12 4.81 0.23 2.47 4.13 0.30 0.38 0.68 1.30
20min
MSN- 2.14 4.76 0.24 2.41 4.24 0.28 0.27 0.52 1.17
1hour
MSN- 2.09 4.88 0.22 2.32 4.39 0.26 0.23 0.49 1.18
2hours
MSN- 2.12 4.81 0.22 2.36 4.32 0.28 0.24 0.49 1.27
4hours

11
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

Fig. 14. Si29-ss-NMR characterization of samples. A) Degree of condensation (DC) equation, B) Table with relative content of silica species, and C) Si29-ss-NMR
spectra of as-made samples (MSN-5min, MSN-1hour, MSN-2hours, and MSN-4hours).

(HO)Si-(OSi)3 (Q3) and Si(OSi)4 (Q4) silicon species. The area under the 4. Conclusions
curve (AUC) was calculated for each peak by deconvolution of the
spectra, fitting them in a Gaussian distribution. At first glance, the We have systematically studied the factors affecting the formation of
spectra show a progressive decrease in the Q2 signal and an increase in MCM41-type MSN. On the one hand, we have studied and characterized
Q3 and Q4. Regarding the AUC calculated from spectra deconvolution the initial stages of MSN formation, especially concerning the nucleation
and Gaussian fitting, it is observed a decreasing tendency in the AUC of and growth processes. Therefore, we have demonstrated that the for­
Q2 with the progression of the reaction time (32.8 > 17.1 > 10.2 > 4.6) mation of MSN is not only affected by the pH, the temperature or the
with a concomitant increase of the AUC of Q3 (41.7 < 47.4 < 48.6 < solvent, as it has been widely studied, but also by the amount of TEOS
74.0) and Q4, except for MSN-4hours (25.5 < 35.6 < 41.2 > 21.4). The added, the stirring rate and the TEOS addition rate. We have observed
degree of condensation (DC) was also calculated, according to the that these parameters influence the silica available to be hydrolyzed and
equation reported by Barczak (Fig. 14A) [90]. The variation of the DC then, affect the hydrolysis and condensation rates, which play a central
parameter reflects that the connectivity of silica in the lattice increases role in the development of MSN. Our observations can be explained by
progressively with the reaction time, except for MSN-4hours (73.2 < the LaMer-based growth model, by which nucleation or growth are
79.6 < 82.8 > 79.2). The lower condensation degree observed in the predominant depending on the self-nucleation threshold. We have also
case of MSN-4hours can be explained by some hypothesis. On the one described the phenomenon of sequential nucleation cycles that occurs in
hand, it must be considered that the hydrolysis and condensation of the reaction and produces heterogeneous samples when the TEOS is
silanol groups are both continuous reactions during the synthesis. slowly available to be hydrolyzed.
Therefore, it can be hypothesized that in the interval from 2 h to 4 h, On the other hand, we have analyzed the evolution and aging of
regarding the silica species equilibrium, the hydrolysis can surpass the MSN. We have observed the rapid formation of MCM41-type MSN,
condensation, decreasing the condensation rate. Considering the suggesting that the templating, nucleation and growth steps occur in less
changes in hydrolysis/condensation tendency, similar behavior can be than 5 min. Then, the evolution and aging of MSN after 5 min mainly
observed in Ref. [45], in which nanoparticles grow from 2 to 4 h, but evolves towards the increase of the condensation inside the nano­
become smaller from 4 to 10 h. On the other hand, the reduction of particles and their strengthening. Also, aging implies that silica is
condensation degree can be explained due to the mild growth in the redistributed between nanoparticles (Ostwald ripening).
nanoparticles from 2 h to 4 h. In this context, the silica deposited to grow In the literature, we can find very complex and smart nanodevices
the nanoparticles can be less condensed than the nanoparticle already developed from MSN, including MCM-41 type ones, with a wide range of
formed at 2 h, reducing the global condensation degree of the nano­ functionalities, and they have been applied to many fields. They have
particles. Finally, it can be also hypothesized that the narrow shape of Q4 been functionalized with alkoxysilanes, molecular gates, dyes and bio­
peak in MSN-4hours provokes that, after the deconvolution, a portion of molecules; loaded with drugs, medical image contrasts, dyes and nucleic
Q4 peak is wrongly integrated as Q3 peak, and therefore, the AUC of Q4 acids, doped with metallic and semimetallic ions, organometallic com­
peak decreases. However, qualitatively the Q4 peak height can indicate pounds, etc. Nevertheless, the detailed information about the synthesis
an increase in Q4 connections in comparison to MSN-2hours, in which conditions of MSN is scorned in most of the reported works. However, as
the peak height is much lower. we demonstrate herein, small changes in the synthesis process can have
These results confirm that once the nanoparticles are formed, the a high impact in the final material obtained. Through our study, a more
silica condensation inside the material is a slow process that can take profound knowledge about the synthesis mechanism of MCM-41-type
several hours. This agrees with previous reports in which a partial MSN has been obtained, especially concerning the parameters that can
condensation was reported for a solid obtained after 3 min of stirring modulate the hydrolysis and condensation rate, as well as the nucleation
[24] that after aging resulted in a more stable and well-ordered MCM-41 and growth processes. This understanding could provide a solid basis to
structure [48,51] due to a gradual increase in the degree of condensa­ improve the control of the synthesis of nanoparticles and obtain a more
tion, which confers the nanoparticles greater resistance to the shrinkage reproducible and controlled MCM-41-type MSN.
caused by calcination [65].
CRediT authorship contribution statement

Vicente Candela-Noguera: Writing – original draft, Investigation,

12
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

Data curation, Conceptualization. María Alfonso: Investigation. Pedro detection based on receptor-gated mesoporous nanoparticles, Nanoscale 14 (2022)
13505–13513, [Link]
Amorós: Investigation, Data curation. Elena Aznar: Supervision,
[13] F. Sancenõn, L. Pascual, M. Oroval, E. Aznar, R. Martínez-Máñez, Gated silica
Investigation. María Dolores Marcos: Writing – review & editing, Su­ mesoporous materials in sensing applications, ChemistryOpen 4 (2015) 418–437.
pervision, Investigation, Funding acquisition, Data curation. Ramón [14] E. Climent, M. Biyikal, D. Gröninger, M.G. Weller, R. Martínez-Máñez, K. Rurack,
Martínez-Máñez: Writing – review & editing, Supervision, Investiga­ Multiplexed detection of analytes on single test strips with antibody-gated
indicator-releasing mesoporous nanoparticles, Angew. Chem. Int. Ed. 59 (2020)
tion, Funding acquisition. 23862–23869, [Link]
[15] E. Garrido, et al., A sensitive nanosensor for the in situ detection of the cannibal
drug, ACS Sens. 5 (2020) 2966–2972, [Link]
Declaration of competing interest acssensors.0c01553.
[16] A. Estepa-Fernández, et al., Engineering nanoparticle communication in living
systems by stigmergy: an application to enhance antitumor therapy in triple-
The authors declare that they have no known competing financial
negative breast cancer, Nano Today 48 (2023), 101692, [Link]
interests or personal relationships that could have appeared to influence nantod.2022.101692.
the work reported in this paper. [17] B. de Luis, et al., Nanoprogrammed cross-kingdom communication between living
microorganisms, Nano Lett. 22 (2022) 1836–1844, [Link]
nanolett.1c02435.
Data availability [18] B. de Luis, A. Llopis-Lorente, F. Sancenón, R. Martínez-Máñez, Engineering
chemical communication between micro/nanosystems, Chem. Soc. Rev. 50 (2021)
Data will be made available on request. 8829–8856, [Link]
[19] B. de Luis, et al., A chemical circular communication network at the nanoscale,
Chem. Sci. 12 (2021) 1551–1559, [Link]
Acknowledgment [20] K. Zhang, et al., Facile large-scale synthesis of monodisperse mesoporous silica
nanospheres with tunable pore structure, J. Am. Chem. Soc. 135 (2013)
2427–2430, [Link]
Vicente Candela-Noguera thanks the Spanish Government for his [21] S. Malekmohammadi, et al., Nonordered dendritic mesoporous silica nanoparticles
fellowship (FPU15/02753). This research was supported by projects as promising platforms for advanced methods of diagnosis and therapies, Mater.
PID2021-126304OB-C41, PID2021-126304OB-C43 and PID2021- Today Chem. 26 (2022), 101144, [Link]
mtchem.2022.101144.
128141OB-C22 funded by MCIN/AEI/10.13039/501100011033/ and [22] G. Lelong, S. Bhattacharyya, S. Kline, T. Cacciaguerra, M.A. Gonzalez, M.-L.
by European Regional Development Fund - A way of doing Europe. This L. Saboungi, Effect of surfactant concentration on the morphology and texture of
study also forms part of the Advanced Materials programme (MFA/ MCM-41 materials, J. Phys. Chem. 112 (2008) 10674–10680, [Link]
10.1021/jp800898n.
2022/049 and MFA/2022/053) and was supported by MCIN with [23] S.H. Wu, H.P. Lin, Synthesis of mesoporous silica nanoparticles, Chem. Soc. Rev. 42
funding from European Union NextGenerationEU (PRTR-C17.I1) and by (2013) 3862–3875, [Link]
Generalitat Valenciana. This study was also supported by Generalitat [24] L. Jin, S.M. Auerbach, P.A. Monson, Simulating the formation of surfactant-
templated mesoporous silica materials: a model with both surfactant self-assembly
Valenciana (CIPROM/2021/007). The authors also thank the Electron
and silica polymerization, Langmuir 29 (2013) 766–780, [Link]
Microscopy Service at UPV for support. la304475j.
[25] J.S. Beck, et al., A new family of mesoporous molecular sieves prepared with liquid
crystal templates, J. Am. Chem. Soc. 114 (1992) 10834–10843, [Link]
Appendix A. Supplementary data 10.1021/ja00053a020.
[26] A. Monnier, et al., Cooperative formation of inorganic-organic interfaces in the
Supplementary data to this article can be found online at [Link] synthesis of silicate mesostructures, Science 261 (1993) 1299–1303, [Link]
org/10.1126/science.261.5126.1299.
org/10.1016/[Link].2023.112840.
[27] S. Sadasivan, C.E. Fowler, D. Khushalani, S. Mann, Nucleation of MCM-41
nanoparticles by internal reorganization of disordered and nematic-like silica -
References surfactant clusters, Angew. Chemie - Int. Ed vol. 41 (2002) 2151–2153, https://
[Link]/10.1002/1521-3773(20020617)41:12<2151::AID-ANIE2151>[Link];2-
U.
[1] F. Sancenón, L. Pascual, M. Oroval, E. Aznar, R. Martínez-Máñez, Gated silica
[28] J. Frasch, B. Lebeau, M. Soulard, J. Patarin, R. Zana, In situ investigations on
mesoporous materials in sensing applications, ChemistryOpen 4 (2015) 418–437,
cetyltrimethylammonium surfactant/silicate systems, precursors of organized
[Link]
mesoporous MCM-41-type siliceous materials, Langmuir 16 (2000) 9049–9057,
[2] C.M.A. Parlett, K. Wilson, A.F. Lee, Hierarchical porous materials: catalytic
[Link]
applications, Chem. Soc. Rev. 42 (2013) 3876–3893, [Link]
[29] Z. Yi, et al., A new insight into growth mechanism and kinetics of mesoporous silica
c2cs35378d.
nanoparticles by in situ small angle X-ray scattering, Langmuir 31 (2015)
[3] A. García-Fernández, E. Aznar, R. Martínez-Máñez, F. Sancenón, New advances in
8478–8487, [Link]
in vivo applications of gated mesoporous silica as drug delivery nanocarriers, Small
[30] K.J. Edler, Current understanding of formation mechanisms in surfactant-
16 (2020), 1902242, [Link]
templated materials, Aust. J. Chem. 58 (2005) 627–643, [Link]
[4] M. Manzano, M. Vallet-regí, Mesoporous silica nanoparticles in nanomedicine
CH05141.
applications, J. Mater. Sci. Mater. Med. 29 (2018) 65, [Link]
[31] F. Lu, S.H. Wu, Y. Hung, C.Y. Mou, Size effect on cell uptake in well-suspended,
s10856-018-6069-x.
uniform mesoporous silica nanoparticles, Small 5 (2009) 1408–1413, [Link]
[5] I.I. Slowing, J.L. Vivero-Escoto, C.W. Wu, V.S.Y. Lin, Mesoporous silica
org/10.1002/smll.200900005.
nanoparticles as controlled release drug delivery and gene transfection carriers,
[32] S.C. Chien, et al., Molecular simulations of the synthesis of periodic mesoporous
Adv. Drug Deliv. Rev. 60 (2008) 1278–1288, [Link]
silica phases at high surfactant concentrations, J. Phys. Chem. C 121 (2017)
addr.2008.03.012.
4564–4575, [Link]
[6] E. Garrido, et al., A sensitive nanosensor for the in situ detection of the cannibal
[33] J. Gu, W. Fan, A. Shimojima, T. Okubo, Organic–inorganic mesoporous
drug, ACS Sens. 5 (2020) 2966–2972, [Link]
nanocarriers integrated with biogenic ligands, Small 3 (2007) 1740–1744, https://
acssensors.0c01553.
[Link]/10.1002/smll.200700311.
[7] F. Catalano, P.P. Pompa, Design rules for mesoporous silica toward the nanosize: a
[34] R. Narayan, U.Y. Nayak, A.M. Raichur, S. Garg, Mesoporous silica nanoparticles: a
systematic study, ACS Appl. Mater. Interfaces 11 (2019) 47237–47246, [Link]
comprehensive review on synthesis and recent advances, Pharmaceutics 10 (2018)
org/10.1021/acsami.9b16135.
1–49, [Link]
[8] Y. Chen, et al., The complete control for the nanosize of spherical MCM-41,
[35] R.I. Nooney, D. Thirunavukkarasu, Y. Chen, R. Josephs, A.E. Ostafin, Synthesis of
J. Nanosci. Nanotechnol. 12 (2012) 7239–7249, [Link]
nanoscale mesoporous silica spheres with controlled particle size, Nature 14
jnn.2012.6459.
(2002) 4721–4728, [Link]
[9] A. Llopis-Lorente, B. Lozano-Torres, A. Bernardos, R. Martínez-Máñez, F. Sancenón,
[36] J. Schmitt, et al., Outset of the morphology of nanostructured silica particles during
Mesoporous silica materials for controlled delivery based on enzymes, J. Mater.
nucleation followed by ultrasmall-angle X-ray scattering, Langmuir 32 (2016)
Chem. B 5 (2017) 3069–3083, [Link]
5162–5172, [Link]
[10] M. Vallet-Regí, F. Schüth, D. Lozano, M. Colilla, M. Manzano, Engineering
[37] Q. Cai, Z.S. Luo, W.Q. Pang, Y.W. Fan, X.H. Chen, F.Z. Cui, Dilute solution routes to
mesoporous silica nanoparticles for drug delivery: where are we after two decades?
various controllable morphologies of MCM-41 silica with a basic medium, Chem.
Chem. Soc. Rev. 51 (2022) 5365–5451, [Link]
Mater. 13 (2001) 258–263, [Link]
[11] F. Tang, L. Li, D. Chen, Mesoporous silica nanoparticles: synthesis, biocompatibility
[38] J. Zhang, Z. Luz, D. Goldfarb, EPR studies of the formation mechanism of the
and drug delivery, Adv. Mater. 24 (2012) 1504–1534, [Link]
mesoporous materials MCM-41 and MCM-50, J. Phys. Chem. B 101 (1997)
adma.201104763.
7087–7094, [Link]
[12] E. Garrido, E. Climent, M.D. Marcos, F. Sancenón, K. Rurack, R. Martínez-Máñez,
Dualplex lateral flow assay for simultaneous scopolamine and “cannibal drug”

13
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

[39] C.J. Brinker, Hydrolysis and condensation of silicates: effects on structure, J. Non- [64] Á.A. Beltrán-Osuna, J.L. Gómez Ribelles, J.E. Perilla, Effects of commonly
Cryst. Solids 100 (1988) 31–50, [Link] unreported physicochemical variables on the morphology of mesoporous silica
[40] Z.A. Qiao, L. Zhang, M. Guo, Y. Liu, Q. Huo, Synthesis of mesoporous silica nanoparticles type MCM-41, J. Nanoparticle Res. 19 (2017) 1–14, [Link]
nanoparticles via controlled hydrolysis and condensation of silicon alkoxide, 10.1007/s11051-017-4077-2.
Chem. Mater. 21 (2009) 3823–3829, [Link] [65] T.W. Kim, P.W. Chung, V.S.Y. Lin, Facile synthesis of monodisperse spherical
[41] E. Yamamoto, K. Kuroda, Colloidal mesoporous silica nanoparticles, Bull. Chem. MCM-48 mesoporous silica nanoparticles with controlled particle size, Chem.
Soc. Jpn. 89 (2016) 501–539, [Link] Mater. 22 (2010) 5093–5104, [Link]
[42] E. Yamamoto, K. Kuroda, Preparation and controllability of mesoporous silica [66] T. Yokoi, T. Karouji, S. Ohta, J.N. Kondo, T. Tatsumi, Synthesis of mesoporous
nanoparticles, in: Enzymes, Elsevier Inc., 2018, pp. 1–10, [Link] silica nanospheres promoted by basic amino acids and their catalytic application,
[Link].2018.09.001. Chem. Mater. 22 (2010) 3900–3908, [Link]
[43] F. Rizzi, et al., High surface area mesoporous silica nanoparticles with tunable size [67] W.Y. Sang, O.P. Ching, Tailoring MCM-41 mesoporous silica particles through
in the sub-micrometer regime: insights on the size and porosity control modified sol-gel process for gas separation, Oct, in: AIP Conference Proceedings,
mechanisms, Molecules 26 (2021) 4247, [Link] American Institute of Physics Inc., 2017, [Link]
molecules26144247. 020147.
[44] J. Rathousky, M. Zukalova, P.J. Kooyman, A. Zukal, Synthesis and characterization [68] T. Yokoi, et al., Mechanism of formation of uniform-sized silica nanospheres
of colloidal MCM-41, Colloids Surfaces A Physicochem. Eng. Asp. 241 (2004) catalyzed by basic amino acids, Chem. Mater. 21 (2009) 3719–3729, [Link]
81–86, [Link] org/10.1021/cm900993b.
[45] Y.D. Chiang, H.Y. Lian, S.Y. Leo, S.G. Wang, Y. Yamauchi, K.C.W. Wu, Controlling [69] K. Kuperkar, et al., Viscoelastic micellar water/CTAB/NaNO3 solutions: rheology,
particle size and structural properties of mesoporous silica nanoparticles using the SANS and cryo-TEM analysis, J. Colloid Interface Sci. 323 (2008) 403–409,
taguchi method, J. Phys. Chem. C 115 (2011) 13158–13165, [Link] [Link]
10.1021/jp201017e. [70] V. Agarwal, M. Singh, G. McPherson, V. John, A. Bose, Microstructure evolution in
[46] M. Varache, I. Bezverkhyy, L. Saviot, F. Bouyer, F. Baras, F. Bouyer, Optimization aqueous solutions of cetyl trimethylammonium bromide (CTAB) and phenol
of MCM-41 type silica nanoparticles for biological applications: control of size and derivatives, Colloids Surfaces A Physicochem. Eng. Asp. 281 (2006) 246–253,
absence of aggregation and cell cytotoxicity, J. Non-Cryst. Solids 408 (2015) [Link]
87–97, [Link] [71] R.M. Pallares, X. Su, S.H. Lim, N.T.K. Thanh, Fine-tuning of gold nanorod
[47] X. Lv, L. Zhang, F. Xing, H. Lin, Controlled synthesis of monodispersed mesoporous dimensions and plasmonic properties using the Hofmeister effects, J. Mater. Chem.
silica nanoparticles: particle size tuning and formation mechanism investigation, C 4 (2016) 53–61, [Link]
Microporous Mesoporous Mater. 225 (2016) 238–244, [Link] [72] K. Möller, J. Kobler, T. Bein, Colloidal suspensions of nanometer-sized mesoporous
micromeso.2015.12.024. silica, Adv. Funct. Mater. 17 (2007) 605–612, [Link]
[48] M. Yu, et al., A simple approach to prepare monodisperse mesoporous silica adfm.200600578.
nanospheres with adjustable sizes, J. Colloid Interface Sci. 376 (2012) 67–75, [73] P.P. Ghimire, M. Jaroniec, Renaissance of Stöber method for synthesis of colloidal
[Link] particles: new developments and opportunities, J. Colloid Interface Sci. 584 (2021)
[49] N.A. Zainal, S.R.A. Shukor, H.A.A. Wab, K.A. Razak, Study on the effect of 838–865, [Link]
synthesis parameters of silica nanoparticles entrapped with rifampicin, Chem. Eng. [74] S. Fouilloux, O. Taché, O. Spalla, A. Thill, Nucleation of silica nanoparticles
Trans. 32 (2013) 2245–2250, [Link] measured in situ during controlled supersaturation increase. Restructuring toward
[50] P. Khodaee, N. Najmoddin, S. Shahrad, The effect of ethanol and temperature on a monodisperse nonspherical shape, Langmuir 27 (2011) 12304–12311, https://
the structural properties of mesoporous silica synthesized by the sol-gel method, [Link]/10.1021/la2013842.
J. Tissues Mater. 1 (2018) 10–17, [Link] [75] R. Kumar, H.T. Chen, J.L.V. Escoto, V.S.Y. Lin, M. Pruski, Template removal and
[51] L. Li, J. Song, J. Wang, W. Fan, Controllable synthesis of mesoporous silica with thermal stability of organically functionalized mesoporous silica nanoparticles,
hierarchical architecture, Mater. Express 7 (2017) 283–290, [Link] Chem. Mater. 18 (2006) 4319–4327, [Link]
10.1166/mex.2017.1381. [76] A. Lodha, et al., Synthesis of mesoporous silica nanoparticles and drug loading of
[52] S. Ma, Y. Wang, Y. Zhu, A simple room temperature synthesis of mesoporous silica poorly water soluble drug cyclosporin A, J. Pharm. BioAllied Sci. 4 (2012) 92–94,
nanoparticles for drug storage and pressure pulsed delivery, J. Porous Mater. 18 [Link]
(2011) 233–239, [Link] [77] J. Kobler, K. Möller, T. Bein, Colloidal suspensions of functionalized mesoporous
[53] T.N.M. Bernards, M.J. van Bommel, A.H. Boonstra, Hydrolysis-condensation silica nanoparticles, ACS Nano 2 (2008) 791–799, [Link]
processes of the tetra-alkoxysilanes TPOS, TEOS and TMOS in some alcoholic nn700008s.
solvents, J. Non-Cryst. Solids 134 (1991) 1–13, [Link] [78] W. Stöber, A. Fink, E. Bohn, Controlled growth of monodisperse silica spheres in
3093(91)90005-Q. the micron size range, J. Colloid Interface Sci. 26 (1968) 62–69, [Link]
[54] H. Yamada, C. Urata, E. Yamamoto, S. Higashitamori, Y. Yamauchi, K. Kuroda, 10.1016/0021-9797(68)90272-5.
Effective use of alkoxysilanes with different hydrolysis rates for particle size [79] K.W. Gallis, C.C. Landry, Synthesis of MCM-48 by a phase transformation process,
control of aqueous colloidal mesostructured and mesoporous silica nanoparticles Chem. Mater. 9 (1997) 2035–2038, [Link]
by the seed-growth method, ChemNanoMat 1 (2015) 194–202, [Link] [80] W.H. Fu, Y. Guan, Y.M. Wang, M.Y. He, A facile synthesis of monodispersed
10.1002/cnma.201500010. mesoporous silica nanospheres with Pm3n structure, Microporous Mesoporous
[55] M.J. Hollamby, D. Borisova, P. Brown, J. Eastoe, I. Grillo, D. Shchukin, Growth of Mater. 220 (2016) 168–174, [Link]
mesoporous silica nanoparticles monitored by time-resolved small-angle neutron [81] B.N. Khlebtsov, V.A. Khanadeev, N.G. Khlebtsov, Determination of the size,
scattering, Langmuir 28 (2012) 4425–4433, [Link] concentration, and refractive index of silica nanoparticles from turbidity Sp[1] S.
[56] H. Yamada, C. Urata, Y. Aoyama, S. Osada, Y. Yamauchi, K. Kuroda, Preparation of M. Mousavi Elyerdi, M. N. Sarvi, and A. J. O’Connor, “synthesis of ultra small
colloidal mesoporous silica nanoparticles with different diameters and their unique nanoparticles (< 50 nm) of mesoporous MCM-48 for bio-adsorption,”, Langmuir 24
degradation behavior in static aqueous systems, Chem. Mater. 24 (2012) (2008) 8964–8970, [Link]
1462–1471, [Link] [82] Y.S. Lin, C.L. Haynes, Impacts of mesoporous silica nanoparticle size, pore
[57] H.I. Meléndez-Ortiz, L.A. García-Cerda, Y. Olivares-Maldonado, G. Castruita, J. ordering, and pore integrity on hemolytic activity, J. Am. Chem. Soc. 132 (2010)
A. Mercado-Silva, Y.A. Perera-Mercado, Preparation of spherical MCM-41 4834–4842, [Link]
molecular sieve at room temperature: influence of the synthesis conditions in the [83] S. Lechevallier, et al., Luminescence properties of mesoporous silica nanoparticles
structural properties, Ceram. Int. 38 (2012) 6353–6358, [Link] encapsulating different europium complexes: application for biolabelling,
[Link].2012.05.007. J. Nanomater. 2013 (2013) 11, [Link]
[58] L. Luo, Y. Liang, E.S. Erichsen, R. Anwander, Monodisperse mesoporous silica [84] Q. He, et al., The effect of PEGylation of mesoporous silica nanoparticles on
nanoparticles of distinct topology, J. Colloid Interface Sci. 495 (2017) 84–93, nonspecific binding of serum proteins and cellular responses, Biomaterials 31
[Link] (2010) 1085–1092, [Link]
[59] S. Liu, et al., The influence of the alcohol concentration on the structural ordering [85] M. Varache, I. Bezverkhyy, F. Bouyer, R. Chassagnon, F. Baras, F. Bouyer,
of mesoporous silica: cosurfactant versus cosolvent, J. Phys. Chem. B 107 (2003) Improving structural stability of water-dispersed MCM-41 silica nanoparticles
10405–10411, [Link] through post-synthesis pH aging process, J. Nanoparticle Res. 17 (2015) 356,
[60] S. Kachbouri, N. Mnasri, E. Elaloui, Y. Moussaoui, Tuning particle morphology of [Link]
mesoporous silica nanoparticles for adsorption of dyes from aqueous solution, [86] B. Moongraksathum, Y.W. Chen, Synthesis and size control of uniform, spherically
J. Saudi Chem. Soc. 22 (2018) 405–415, [Link] shaped hexagonal mesoporous silica, J. Porous Mater. 26 (2019) 51–58, https://
jscs.2017.08.005. [Link]/10.1007/s10934-018-0609-0.
[61] K.S. Rao, K. El-Hami, T. Kodaki, K. Matsushige, K. Makino, A novel method for [87] D.M. Oliveira, A.S. Andrada, Synthesis of ordered mesoporous silica MCM-41 with
synthesis of silica nanoparticles, J. Colloid Interface Sci. 289 (2005) 125–131, controlled morphology for potential application in controlled drug delivery
[Link] systems, Cerâmica 65 (2019) 170–179, [Link]
[62] L.M. Rossi, L. Shi, F.H. Quina, Z. Rosenzweig, Stöber synthesis of monodispersed 69132019653742509.
luminescent silica nanoparticles for bioanalytical assays, Langmuir 21 (2005) [88] M. Lindén, S.A. Schunk, F. Schüth, In situ X-ray diffraction study of the initial
4277–4280, [Link] stages of formation of MCM-41 in a tubular reactor, Angew. Chem. Int. Ed. 37
[63] H. Yamada, H. Ujiie, C. Urata, E. Yamamoto, Y. Yamauchi, K. Kuroda,
A multifunctional role of trialkylbenzenes for the preparation of aqueous colloidal
mesostructured/mesoporous silica nanoparticles with controlled pore size, particle
diameter, and morphology, Nanoscale 7 (2015) 19557–19567, [Link]
10.1039/C5NR04465K.

14
V. Candela-Noguera et al. Microporous and Mesoporous Materials 363 (2024) 112840

(1998) 821–823, [Link] [90] M. Barczak, Template removal from mesoporous silicas using different methods as
6<821::AID-ANIE821>[Link];2-I. a tool for adjusting their properties, New J. Chem. 42 (2018) 4182–4191, https://
[89] N.T.K. Thanh, N. Maclean, S. Mahiddine, Mechanisms of nucleation and growth of [Link]/10.1039/c7nj04642a.
nanoparticles in solution, Chem. Rev. 114 (2014) 7610–7630, [Link]
10.1021/cr400544s.

15