Synthetic and Catalytic Potential of Amorphous Mesoporous Aluminosilicates Prepared by Postsynthetic Aluminations of Silica in Aqueous Media

Article abstract of DOI:10.1002/cctc.201701660

Amorphous aluminosilicate catalysts have been used industrially on a large scale for almost a century. However, the influence of the pH on the alumination of silica in aqueous solutions has remained largely unclear. Herein, room temperature aluminations of different mesoporous amorphous silicas (fumed silica, dried silica gel, SBA-15, MCM-41, and COK-12) with aqueous solutions of various pH (3–13) are explored. The aqueous solutions are prepared using different aluminum sources (Al(NO₃)₃ or NaAlO₂) and alkaline additives (NaOH or NH₄OH). The decoupling of pH and Al source using alkaline additives results in a vast experimental potential to prepare unique aluminosilicates, whereby an important role is played by the pH development during the treatment. The bulk and surface composition, acidity, aluminum coordination, morphology, hydrothermal stability, and porosity of the obtained materials are characterized. Optimal samples possess large surface areas and superior acidities (up to 50 % higher) and outstanding stabilities compared to aluminosilicates prepared with state-of-the-art methods. The obtained materials are evaluated in a series of acid-catalyzed model reactions involving substrates of various chemical reactivity and size, enabling insight in the catalytic functionality of the introduced Br?nsted and Lewis sites. The potential of the obtained materials is emphasized by the similar or superior acidity and catalytic performance compared to several benchmark industrial silica–alumina-based catalysts.

Full text of DOI:10.1002/cctc.201701660

Accepted Article
Title: Synthetic & Catalytic Potential of Amorphous Mesoporous
Aluminosilicates Prepared by Post-Synthetic Aluminations of
Silica in Aqueous Media
Authors: Roel Locus, Danny Verboekend, Martin d'Halluin, Michiel
Dusselier, Yuhe Liao, Nicolas Nuttens, Tony Jaumann,
Steffen Oswald, Luis Mafra, Lars Giebeler, and Bert Sels
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201701660
Link to VoR: http://dx.doi.org/10.1002/cctc.201701660
A Journal of
www.chemcatchem.org

10.1002/cctc.201701660
ChemCatChem
DOI: 10.1002/ ((please add manuscript number))
Article type: Full Paper
Synthetic & Catalytic Potential of Amorphous Mesoporous Aluminosilicates Prepared
by Post-Synthetic Aluminations of Silica in Aqueous Media
Roel Locus, Danny Verboekend*, Martin d’Halluin, Michiel Dusselier, Yuhe Liao, Nicolas
Nuttens, Tony Jaumann, Steffen Oswald, Luís Mafra, Lars Giebeler, Bert Sels*
Dr. R. Locus, Dr. D. Verboekend, Dr. M. d’Halluin, Dr. M. Dusselier, Y. Liao, N. Nuttens,
Prof. B. F. Sels
Center for Surface Chemistry and Catalysis, KU Leuven,
Celestijnenlaan 200F, Bus 2461, B-3001 Heverlee, Belgium
E-mail: danny.verboekend@kuleuven.be, bert.sels@kuleuven.be
Dr. T. Jaumann, Dr. S. Oswald, Dr. L. Giebeler
Leibniz-Institute for Solid State and Materials Research (IFW) Dresden e.V., Institute for
Complex Materials, Helmholtzstr. 20, D-01069 Dresden, Germany
Dr. L. Mafra
CICECO - Aveiro Institute of Materials, Department of Chemistry, University of Aveiro,
Campus Universitário de Santiago, 3810-193 Aveiro, Portugal
Keywords: amorphous aluminosilicates, silica, alumination, acidity, catalysis
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Abstract
Amorphous aluminosilicates catalysts have been used industrially on a large scale for almost a
century. However, the influence of the pH on the alumination of silica in aqueous solutions
has remained largely unclear. Herein, room temperature aluminations of different mesoporous
amorphous silicas (fumed silica, dried silica gel, SBA-15, MCM-41, and COK-12) with
aqueous solutions of various pH (3-13) are explored. The aqueous solutions are prepared
using different aluminum sources (Al(NO3)3 or NaAlO2) and alkaline additives (NaOH or
NH4OH). The decoupling of pH and Al source using alkaline additives results in a vast
experimental potential to prepare unique aluminosilicates, where an important role is played
by the pH development during the treatment. The bulk and surface composition, acidity,
aluminum coordination, morphology, hydrothermal stability, and porosity of the obtained
materials are characterized. Optimal samples possess large surface areas and superior acidities
(up to 50% higher) and outstanding stabilities compared to aluminosilicates prepared via state
of the art methods. The obtained materials are evaluated in a series of acid-catalyzed model
reactions involving substrates of various chemical reactivity and size, enabling insight in the
catalytic functionality of the introduced Brønsted and Lewis sites. The potential of the
obtained materials is emphasized by the similar or superior acidity and catalytic performance
compared to several benchmark industrial silica-alumina-based catalysts.
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
1. Introduction
Porous amorphous aluminosilicates are amongst the most important heterogeneous acid
catalysts because of their active acid sites, a facile recoverability, high regeneration ability,
and a relatively easy synthesis.[1] Combined with the large abundance of silicon and
aluminum, these properties render them environmentally-friendly and industrially-viable
catalyst.[1,2] Mesoporous amorphous silica-alumina (ASA) are accordingly used on large scale
in the hydrocracking of heavy oil fractions to middle distillates, as active matrix for pre-
cracking of heavy hydrocarbon fractions in fluidized catalytic cracking particles, in the
dehydrochlorination of chlorinated hydrocarbons, and numerous others.[2-5]
The advantage of ASA over (purely Lewis acidic) alumina is that Brønsted acid sites are
formed when monomeric aluminum is incorporated in the silica framework with a tetrahedral
coordination (isomorphous substitution).[6,7] Still, the exact nature of the acidity of ASAs can
be rather complex due to their large heterogeneity.[8-16] Not only the relative quantity of silica
and alumina can strongly vary, but also their distribution (zoning, gradients, etc.), their
coordination (in the case of Al: tetrahedral/framework versus octahedral/extra-framework),
and accessibility (isolated aluminum atoms within the bulk or accessible on the surface).
Based on these issues, the total acidity of the ASA is usually much lower (generally <100
µmol g-1, Table S1), compared to the total aluminum content in the aluminosilicate catalysts.
ASAs can be synthesized using a variety of bottom-up and top-down methods, such as co-
precipitation, grafting, and deposition, all of which involve the contacting of a silica source
with an aluminum source or vice versa.[8-16] One of the most used (and commercially
attractive) methods may be the deposition of Al by subjecting silica to Al-containing aqueous
solutions, taking advantage of the hydrolysis-condensation behavior of Al3+ ions in water. The
latter is typically performed using various Al salts such as Al2(SO4)3, Al(NO3)3, and NaAlO2.
Unlike the aforementioned salts, NaAlO2 forms an alkaline solution after dissolution, and has
yielded materials with relatively high acidity.[14,16,17] This observation implies that control of
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
the pH may be beneficial for the creation of acid sites. However, despite an early indication of
this beneficiary effect,[18,19] the influence of the pH on the alumination of silica in aqueous
solutions has never been subjected to a dedicated study.
Herein, we present the post-synthetic alumination of silica using aqueous solutions of varying
pH as a facile technique to prepare aluminosilicate catalysts featuring superior physico-
chemicals properties and outstanding catalytic performance compared to conventional silica-
aluminas. In a case study on the widely-studied model system SBA-15, the effects of the
alumination step on the porous, compositional, structural, and acidic properties are disclosed.
Within the compositional range studied, high solid yields are combined with total acidities
that exceed the maximum values reported in the literature by ca. 50%. In addition, the overall
porosity is largely preserved and the hydrothermal stability is enhanced. The aluminated
SBA-15 samples showed exceptional activities and selectivities in various Lewis and
Brønsted acid-catalyzed model reactions, such as alkylation, isomerization, and cracking.
Furthermore, the versatility of the method is emphasized by the successful extrapolation
towards various ordered (MCM-41 and COK-12) and non-ordered silica sources, such as
fumed silica and dried silica gel.
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
2. Results and Discussion
The pH development prior to and during the alumination of silicas in aqueous media is
discussed in Section 2.1. Next, in Section 2.2, the synthesis and characterization of the
aluminated SBA-15 materials is discussed. SBA-15 was selected as a model type for
amorphous mesoporous silica because of the well-defined 1D ordered pore system. In Section
2.3 an extrapolation to other ordered and non-ordered amorphous silicas is presented. Next, in
Section 2.4 selected samples are catalytically evaluated and compared to several commercial
amorphous and crystalline aluminosilicates standards. Finally, the relationships between the
active solid catalyst, pH, acidity, and activity are discussed (Section 2.5).
2.1. The Role of pH during the Alumination of Silica in Aqueous Media
The pH in the used aqueous solutions depends on the nature and quantity of the type of Al salt
and the base (Figure 1a). For example, the addition of Al(NO3)3 to distilled water causes a
drop in pH to a plateau value of ~3, due to complexation of the OH- ions by Al3+ ions. To
create alumination solutions with higher pH three other types of solutions were employed.
’
First, solutions of NaAlO2 were studied (‘0-x-NaAlO2 samples, where ‘x’ refers to the applied
Al concentration [Alinitial]). These solutions display an increasing pH as a function of [Al]initial
due to the increased release of OH- ions. Nevertheless, as with Al(NO3)3, the alkalinity in
these solutions remains directly connected to the amount of Al. Conversely, by
complementing the Al source (Al(NO3)3) with a certain amount of base, the pH and the Al
concentration are decoupled. NaOH and NH4OH were selected as references for strong and
weak bases, respectively. NaOH completely dissociates in water, whereas NH4OH yields a
partial dissociation (Kb = 1.8 · 10-5) and has a buffering effect on the pH.[22] For the sake of
conciseness, we have fixed the base concentrations of NaOH (0.1 M) and NH4OH (0.5 M)
within this contribution. These base concentrations were selected based on the recent work on
the enhancement of the acidity of Al-containing MCM-41 by base treatments.[20] In the case
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
of the solutions with NaOH (‘0-x-NaOH’ samples) the pH reduces due to the complexation of
-
hydroxyls to form the aluminate ions (Al(OH)4 ). For aluminations with NH4OH on the other
hand, a buffering effect is observed in the form of a plateau at a pH of ~ 10 for increasing
aluminum concentrations (‘0-x-NH4OH’ samples). The measured pH curves in Figure 1a
agree well with the calculated ones (Figure S1). It should be noted that a solid alumina phase
can form in Al-containing solutions when the pH is reduced to ~10 and lower.[21] The
formation of alumina (and its hydroxides) occurred in samples 0-0.03-NaOH, 0-0.02-NH4OH
and 0-0.03-NH4OH, and was observed as a transformation in the solution from clear to
colloidal white (Figure S2). Accordingly, aluminations of silicas were systematically
performed by mixing the silica and the Al source directly with the base, rather than first
mixing the Al source with the base, followed by the addition of the silica.
During the aluminations of silica SBA-15, the pH decreases as a function of time (Figure 1b-
c). This change is explained partially by the low isoelectric point of silica (about 2-3),
combined with the consumption of hydroxyl ions by the gradual dissolution of silica.[23] The
latter is proven by the pH-time profile of S-0-NaOH, in which the absence of Al implies that
the reduction of the pH is largely due to the consumption of hydroxyl ions by the silica
dissolution. When aluminum is present, the pH lowering is more pronounced. For S-0.03-
NaOH the pH dropped to almost neutral (pH ~ 6) and for S-0.04-NaOH (not shown) even to
3.7. For the alumination with NH4OH, the buffering effect on the pH is obvious (Figure 1c).
Only a small initial change of pH is observed after the addition of the Al and the silica (0-2
min) to the ammonium hydroxide solution, and the pH remained constant (around 10.5)
during the remainder of the treatment. During aluminations with NaAlO2, the pH increased to
above 10 within the first minute after the addition of distilled water to the mixture of silica
and NaAlO2. Afterwards this initial increase the pH reduced gradually, reaching
approximately 9 after 30 min.
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
2.2. Synthesis and characterization of aluminated ordered mesoporous silica
The different alumination techniques and the properties of the parental and the resulting
materials are summarized in Table 1. A first important criterion after any post-synthetic
modification is the mass yield (Y), which is particularly relevant in this case as silica readily
dissolve in alkaline solutions (pH > 10). Mass yields of S-x-NaOH and S-x-NH4OH for
aluminum concentrations ([Al]initial) lower than 0.006 M are around 75%, but for higher
[Al]initial mass yields nearly reach 100% indicating a negligible loss of silica (Figure 2a). The
increasing yields likely relate to the lower pH at increasing [Al]initial. Still, the increased solid
yield may also be due the increased resistance of the solid to alkaline leaching as the covering
of silica by Al surface species.[24] For example, even though the pH of solutions with NaAlO2
increases with Al content, aluminations with NaAlO2 did not show a reduced mass yield as a
function [Al]initial. The absence of such trend confirms that, besides the pH, also other factors
such as the Al content have a pronounced influence on the dissolution of silica.
Bulk (SiICP/AlICP) and surface (SiXPS/AlXPS) composition evidenced the absence of Al in the
parent SBA-15 (S-P, Table 1 and Figure 2b). An Al incorporation of ~90% into the samples
S-x-NaOH and ~85% in S-x-NH4OH was achieved (Figure 2b,c). On the other hand,
alumination with NaAlO2-based solutions led to an incorporation of less than 60% of the
initial Al into the respective silica. The Si/Al ratios are lower for the surface(-near) species
compared to the bulk, suggesting that most Al is incorporated near the surface of the pores
and/or the outer surface of the particles. This result follows logically from post-synthesis
alumination by a solution, differing to aluminations during silica synthesis (as for example in
Al-MCM-41). In the latter case, the surface Si/Al ratios are often higher than the bulk ratios
because of the incorporation of Al into the pore walls rather than on the surface.[20]
The porous properties of the samples before and after aluminations were analyzed with N2
physisorption (Figures 3a,b and S3). The typical type IV nitrogen isotherms with type A
hysteresis loops were observed, indicating capillary condensation in (cylindrical) mesopores
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
in the parent SBA-15.[25] A sharp pore size distribution (PSD) of ~7 nm was combined with a
high specific surface area
500 m2 g-1 and high pore volumes > 0.5 cm3 g-1. For S-x-NaOH
>
samples, the average pore size (dp) did not change after alumination, but the PSDs became
broader, forming pores up to 20 nm for [Al]initial lower than 0.015 M. The amount of
micropores decreased after this type of alumination on SBA-15, as is indicated by the lower
N2 uptake at p/p0<0.2. These trends are explained by Ostwald ripening, which is catalyzed by
OH- ions. Pores with larger sizes expanded at the expense of smaller pores, which were filled
up with a portion of the etched silica from the larger pores.[20,21] The isotherms and pore size
distributions of S-x-NH4OH and S-x-NaAlO2 samples remained mostly the same upon
treatment (Figure S3). As for the S-x-NaOH samples, the largest changes in pore size
occurred in aluminations with the highest pH (and thus the lowest [Al]initial). The evolution in
the PSDs after different aluminations was in good resemblance with the trends obtained on
mildly alkaline-treated Al-MCM-41.[20]
The pore volume (Vpore) and specific surface area (SBET) showed comparable trends after
aluminations (Table 1). Generally, Vpore and SBET decreased by ca. 20% in alkaline media.
This behavior is partially explained by the filling of smaller micro- and mesopores by
Ostwald ripening and the deposition of Al(-oxide) species in these smaller pores. For the S-x-
NaOH samples an increasing Al content reduced the negative influence on Vpore and SBET by
an increased neutralization of the solution. This neutralization did not occur for the S-x-
NH4OH and S-x-NaAlO2 samples. In the former, the buffering effect caused an almost
constant pH and in the latter the pH increased with the aluminum concentration. In general,
the changes in porosity related well with the yield. Moreover, a general observation is that as
long as the yield exceeded 90%, a limited effect on the porosity is achieved.
The 1D-hexagonal pore ordering in SBA-15 enables a straightforward analysis of the effect of
the aluminations on the porous structure using transmission electron microscopy (TEM). In
the parent material the ordered mesopores were clearly visible (Figure 4). After aluminations
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
in NaOH with [Al]initial of 0.002 M and 0.015 M domains significantly appeared less ordered,
especially near the edges of the particles. This observation suggests that the highest reactivity
of the hydroxyls occurs at the outside of the silica particles. The latter may relate to the
existence of accessibility limitations, inhibiting the hydroxyls to reach bulk regions in the
unidirectional SBA-15 pores. In the sample S-0.03-NaOH no pore transformation was
observed, in agreement with the observed high mass yields and preserved porosities. No
significant effect on the particle size was detected by TEM.
Alumination of amorphous silica samples can induce two major types of Al coordination.
Tetrahedral Al (AlIV) relates to Al species fully incorporated into the silica network, whereas
octahedral Al (AlVI) indicates the presence of extra-structure Al2O3 domains.[6] In the treated
27
samples, at least two types of Al coordination can be distinguished using Al MAS NMR:
AlIV sites resonating at ~54 ppm, while AlVI sites resonate at ~0 ppm.[6] The NMR spectrum
of S-0.03-NaOH reveals around 70% tetrahedral aluminum, which is similar to the
commercial benchmark aluminosilicate (ASA-6, Figure S4).
To monitor the introduced acidity, pyridine-probed Fourier transformed infra-red
spectroscopy (PP-FTIR) experiments were conducted (Table 2 and Figure 5). In the S-x-
NaAlO2 samples a plateau value was achieved both in Brønsted (B) and in Lewis (L) acidity
(~30 and ~70 µmol g-1 respectively) for [Al]initial ≥ 0.015 M. For the S-x-NaOH samples on
,
the other hand, B and L acidity were lower compared to the SBA-15 treated with the solution
of NaAlO2 for [Al]initial < 0.03 M. However, a higher acidity was reached for S-0.03-NaOH (B
= 60 µmol g-1 and L = 78 µmol g-1). With higher concentrations of aluminum (S-0.04-NaOH)
the acidity was substantially lower (B = 31 µmol g-1, L = 76 µmol g-1). Compared to the S-x-
NaOH samples, the increase in acidity with increasing [Al]initial was lower for the S-x-NH4OH
samples (B = 39 µmol g-1 and L = 78 µmol g-1, for S-0.03-NH4OH). However, the latter
acidity of S-0.03-NH4OH was higher compared to S-0.03-NaAlO2. In general, the ratio of B/L
was roughly constant at 0.4 for the NaAlO2-treated samples. The NH4OH and NaOH-treated
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
samples yielded B/L ratios of ≥0.5, which were less constant over the Al concentration
applied. Both bases yielded a relatively high B/L ratio (about 0.7) at 0.006 M, which reduced
to 0.5 at higher concentrations ([Alinitial]>0.006 M). Exceptionally, S-0.03-NaOH yielded a
B/L of nearly 0.8, which is tentatively attributed to the significant pH change during the
treatment (Figure 1c, vide supra). The Brønsted and the total acidity in S-0.03-NaOH are
amongst the highest values achieved for amorphous silica-alumina both commercial and in
literature (Figure 6a,b). Moreover, for Si/Al ratios exceeding 15, sample S-0.03-NaOH
features over 50% higher Bronsted and 30% higher total acidity compared to the state of the
art. The latter is a significant result, particularly taking into account the affordable nature of
the reagents, and the ease of the procedure, and the vast potential to further optimize the
alumination protocol.
A measure to assess the accessibility of the aluminum atoms introduced into the silica is to
normalize the number of formed acid sites (B + L) by the total amount of aluminum (AlICP).
This measure, also referred to as the effective acidity (EA), was recently used to highlight
acidity changes upon pH-controlled activation of Al-containing MCM-41.[20] At
[Al]initial ≤ 0.02 M, the EA values are highest for S-x-NaAlO2 samples followed by samples
obtained by treatment in NH4OH and NaOH, respectively (Figure 5d). Nevertheless, at
increasing [Al]initial the effective acidities converge to around 20% (at 0.03 M), with S-0.03-
NaOH displaying the highest EA. This lowering of the EA as a function of [Al]initial is
attributed to the decreased number of isolated acidic Al located on the external surface, which
relates well with the increased SiXPS/AlXPS ratios (Figure 2c). Such trend accords with the
existing literature, where the highest reported EA values follow a quasi-linear increase from
about 5% at Si/Al = 10 to about 30% at Si/Al = 60 (Figure 6c). Within these references, the
effective acidity of the samples prepared by the alumination display among the highest,
showing that the described techniques yield rather accessible Al sites. In addition, sample
‘M50-0-NaOH’, derived from ref. [20] and prepared by optimized mild alkaline treatment of
10
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Al-containing MCM-41, confirms that preparing aluminosilicates at the appropriate pH yields
the highest effective acidities.
The hydrothermal stability of aluminosilicates is an important parameter in many catalytic
applications. Stability tests in boiling water form a suitable test to assess hydrothermal
stability,[27] and were executed on selected samples. After the treatment, the porosity of the
samples was measured (Table 1). The parent silica (S-P) showed a pronounced one-third
reduction in BET surface area. Conversely, the aluminated samples treated with [Al]initial of
0.015 to 0.03 M (Si/Al ratios of ca. 20-50) featured similar or even slightly enhanced BET
surface areas (increase up to 10%). Moreover, the aluminated samples displayed a
substantially higher stability in boiling water compared to a commercial aluminosilicate
(ASA-6), which showed a 16% decrease in BET surface area.
An increase of the stability of the aluminated materials compared to the parental silicas may
be expected based on the higher Al content.[27] However, the Al content alone cannot explain
the superior stability of the aluminated samples compared to the commercial reference ASA-6.
In this case, the superior stability can be attributed to the more selective deposition of Al on
the external surface. As mentioned above, in ASA-6 more aluminum is likely localized in the
bulk as typical for silica syntheses with both Al and Si precursors, where it is unable to
protect the silica framework in many cases.[27] As the effective acidity is directly linked to the
Al distribution, the stability test results can be related to the effective acidity. Hence, the
effective acidity can act as an indirect indicator of the hydrothermal stability of amorphous
aluminosilicates. Naturally, more dedicated study is required to verify the latter hypothesis.
Nevertheless, the potential relationship implies that even when the porosity and acidity of the
aluminosilicates prepared by post-synthetic alumination are similar to that of a commercial
sample, the use of the post-synthetic technology remains advantageous as the solids are more
likely to maintain their physico-chemical properties in the catalytic cycle.
11
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
2.3. Extrapolation to Ordered and Non-Ordered Amorphous Silica
Other ordered and non-ordered mesoporous silica were exposed to selected alumination
conditions in NaOH. Like for SBA-15, treatment of the ordered silicas COK-12 and MCM-41
(Table S2) resulted in a yield of about 75%, which increased to about 100% with an increased
[Al]initial. Additionally, for these materials the base-only treatment leads to a reduction of the
overall BET surface area, and an enlarged average pore diameter (dp) (Figure S5). Similar to
the yield, these effects decreased with increased [Al]initial
.
For non-ordered amorphous silica (AS1-P and AS2-P), the weight loss at low [Al]initial was
about two-fold (ca. 50%, Table 1, Figure 2a). Intuitively, the reactivity of silica towards
alkaline media may be directly related to its overall (BET) surface area. However, the degree
of dissolution in the absence of aluminum does not relate well with the total BET surface area
(Table 2). Nevertheless, a trend was attained between the treatment solid mass yield (Y) and
the pore size of the parent silica (Figure S6), suggesting a minimum (meso)pore size to
achieve a larger extent of dissolution. This implies that, under the studied conditions, the
higher solid yields of the ordered silica may be caused by the limited access of the base
hydroxyls to the center part of the unidirectionally structured MCM-41, SBA-15, and COK-
12 particles. This hypothesis agrees well with the local reduction at the edges of the SBA
particles as evidenced by TEM (vide supra). Alike SBA-15, for high [Al]initial high yields (up
to 100%) were attained for the non-ordered silica, whereas the porosity was largely preserved.
In dried silica gel-derived samples (‘AS1-x-b’) the pore size distributions remained almost
unchanged with only slightly larger tail towards larger pore sizes (Figure 3d). Typical pore
sizes occurred between 5 and 20 nm with a maximum around 10 nm. On the other hand,
fumed silica-derived samples (‘AS2-x-b’), displayed a more pronounced increase in the
number of larger mesopores after alumination (25-35 nm, Figure S5d).
The treatments successfully introduced substantial Brønsted and Lewis acidity into the
materials. About 30% lower acidities were obtained compared to SBA-15 (Table 2). The
12
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
AS1-derived samples typically displayed about 40-50% lower B/L ratios compared to SBA-
15-derived samples, and, in this regard, the acidic properties are closer to the commercial
standard ASA materials.[8,9,28] Compared to the acidities of these commercial materials (ASA-
6 and ASA-0.4), the total acidities of ~60 µmol g-1 obtained for the AS1 and AS2 samples are
slightly lower. Still, AS1 and AS2-based samples feature considerably higher Si/Al ratios
(based on the maximal theoretical amount of incorporated aluminum), and therefore should
display substantially higher effective acidities (Table 2). Moreover, optimizations on the
alumination of these amorphous silicas may further enhance the acidity of these materials.
Finally, in-line with the SBA-15-related materials, hydrothermal treatment of AS1-0.03-
NaOH yielded a preservation of the BET surface area.
2.4. Catalytic Functionality
The functionality of the formed acid sites in selected samples was evaluated in several acid-
catalyzed reactions (Figure 7). These regard reactions involving substrates of different sizes
catalyzed by either Lewis, Brønsted, or both types of acid sites, therefore enabling a thorough
insight in the catalytic potential of the aluminosilicate catalysts. The Friedel-Crafts alkylation
of toluene with benzyl alcohol (BA) is catalyzed by moderate and strong Brønsted acid sites
(Figure 7a-b), and represents a frequently-used model reaction for acid catalysis in the
condensed phase.[29] An increase of the conversion of benzyl alcohol (ABA) was observed for
increasing [Al]initial in NaOH-treated SBA-15 (S-x-NaOH, Figure 7a). On the other hand, the
S-x-NH4OH samples display an optimum in ABA of 19 mmolBA gcat-1 h-1 at [Al]initial = 0.02 M.
Aluminations with NaAlO2 enhance the activity with increasing [Al]initial, however, this
enhancement is smaller compared to the S-x-NaOH samples. For example, the maximum
activity (S-0.03-NaAlO2) was around 75% compared to S-0.03-NaOH. S-0.03-H2O and S-
-1
0.04-NaOH performed poorly showing an activity of about 10 mmolBA gcat h-1, further
highlighting the importance of optimizing the Al concentration in the alumination process.
13
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Based on the performance of the S-x-b samples, several AS-1-derived samples were selected
for catalytic evaluation. AS1-0.03-NaOH and AS1-0.02-NH4OH samples displayed an ABA of
15 and 16 mmolBA gcat-1 h-1, which is slightly higher than for commercial ASA-6 (14 mmolBA
gcat-1 h-1), and significantly higher compared to γ-alumina (2 mmolBA gcat-1 h-1). Otherwise, the
-1
sample AS1-0.015-NaAlO2 resulted in an activity of 9 mmolBA gcat h-1. An approximately
linear relation was found between Brønsted acidity and ABA for Brønsted acidities below 30
µmol g-1 (Figure 7b). In general, the aluminated samples displayed a higher performance per
Brønsted acid site compared to ASA-6, suggesting that the acid sites (particularly in the S-x-
NaOH samples) offer higher functionality.
The catalytic pyrolysis of low-density polyethylene (LDPE) -an important challenge in the
repurposing of plastic waste- was applied as model reaction to evaluate the ability of the
catalyst to convert high molecular-weight hydrocarbons (Figure 7e-f).[30-31] The LDPE
pyrolysis was studied by comparing the temperature where 50% (T50) of the mass was
converted to volatile products. Figure 7e shows T50 decreasing as a function of the applied
initial aluminum concentration down to 375 °C for an [Al]initial of 0.02 M. At higher aluminum
concentrations, the T50 remained largely constant for the S-0.03-NH4OH and S-0.03-NaAlO2
samples. In contrast, for S-0.03-NaOH, the T50 decreases to about 336 °C. The latter
conversion is lower compared to the commercial, strongly-acidic USY zeolites with Si/Al
ratio of 40 (USY-40) and 2.6 (USY-2.6), which reached a T50 of 341 °C and 377 °C,
respectively. T50 values of non-ordered AS1 samples after alumination were comparable (ca.
360 °C) to the commercial amorphous aluminosilicate ASA-6. An inverse correlation was
visible between T50 and the total acidity (Figure 7f). Based on the lower position of the
profile, like in the case of the aforementioned alkylation, the acid sites in the S-x-NaOH
samples seem to achieve the most efficient catalytic performance.
The aluminated samples were also tested in the reaction of 1,3-dihydroxyacetone (DHA) to
ethyl lactate (ELA) via pyruvic aldehyde (PAL) (Figure 7g-h). Lactic acid (and its ester
14
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
derivatives, such as ELA) is emerging as a renewable chemical platform for solvents, acrylic
acid and poly-lactic acid, an important biodegradable plastic.[32-35] Often the rate-determining
step in the conversion of DHA to ELA is the dehydration and subsequent rearrangement of
DHA to PAL, which is catalyzed by both Brønsted and Lewis acid sites. The second step,
PAL to ELA with incorporation of ethanol solvent, is catalyzed primarily by Lewis acid sites
at low temperature, while strong Brønsted acidity leads to side products and low product
selectivity.[32] The initial rate ELA formation (RELA) showed that the S-0.03-NaOH sample
was most active, plotted in function of [Alinitial] (Figure 7g), e.g. yielding 15% of ELA after 1
h compared to 9% and 13% in S-0.03-NaAlO2 and ASA-6, respectively. Other alumination
techniques showed lower ELA formation rates. A roughly linear relation was observed
between the RELA and the total amount of acid sites (Figure 7h). The latter suggest that the
Lewis acid sites in the materials are active in the conversion of PAL to ELA, as the sole
activity Brønsted acid sites would have resulted in a much higher selectivity to side-products
at the expense of ELA formation.[32]
The valorization of α-pinene was used as a second test reaction in the field of bio-based
renewables (Figure 7c-d). α-pinene is the main component of turpentine oil, which is a
significant side product in the paper and the medium density fiberboard industry, which both
use (pine) trees as biomass source.[36] In this reaction, the rate determining step is based on the
Brønsted acid catalyzed conversion of α-pinene to the pinanyl carbocation, that further
rearranges into mono- and bicyclic products. Fast product diffusion (and thus acidity in large
accessible pores) is essential to avoid polymeric hydrocarbon formation. The activity (Aα-p)
-1
and showed an optimum around [Al]initial of 0.015 M, where activities of ca. 400 mmolα-p gcat
h-1 are reached (Figure 7c). The latter indicates that, for this reaction, an optimum of acidity
exist for the aluminated samples. The activity and productivity of the aluminated MCM-41
(M-x-NaOH) samples showed roughly similar performance compared to the S-x-NaOH
samples, yielding an activity of ~400 mmolα-p gcat-1 h-1 at [Al]initial = 0.02 M.
15
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Finally, selected aluminated samples were tested in the cracking (dealkylation) reaction of
propyl phenol. The cracking of alkylated aromatics plays an important role in modern day
petroleum refineries, and is likely to gain high attention in future bio-refineries. This reaction
is catalyzed by Brønsted and Lewis acid sites in gas phase, but zeolites are required to achieve
high phenol selectivity (>90%).[37] The aluminated materials were active in the dealkylation
of propylphenol (Figure 8). S-x-b samples display similar conversions of propylphenol (Xpp)
compared to the sample ASA-6, whereas AS1-x-b samples resulted in slightly lower
conversions (Figure 8a,b). The samples aluminated with NaAlO2 systematically displayed
the lowest Xpp. Despite the lower conversions compared to ASA-6, the selectivity towards
phenol and propylene (Sda in Figure 8c,d) was about 15% higher for the AS1-x-b samples. S-
x-b samples also displayed higher selectivities than ASA-6, comparable to the AS1-x-b
samples. The highest Sda was achieved in the AS1-0.03-NaOH (Sda~90%, at Xpp~70%). This
sample approached the very high shape selectivity of the zeolite ZSM-5 (> 95%) which is
remarkable, given its amorphous nature and inexpensive synthesis.
2.5. Relationship between solid, pH, acidity, and activity
The implications of aluminations of silica should not only be evaluated by the used aluminum
and silica source, but also by the involved pH (Figure 9a). Herein, the use of base additives
can play an important role, as they allow tailoring of the pH to that point, at which the Al is
most efficiently deposited on the external surface of silica. In addition to the initial pH of the
alumination solution, the pH development during the treatment also matters. In fact, the pH at
the end of the treatments (pHfinal, Table S3) may be even more important, as judged by the
volcano-type relation of both the total acidity (B+L) and the activity in alkylation to the pHfinal
(Figure 9b). Both quantities are optimal when pHfinal is neutral (pH 5-8). Hence, if the pH
strongly varies in time during the alumination (as with strong bases such as NaOH, Figure
1b), control of the treatment time becomes important. In this context, weak bases, such as
16
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
NH4OH (Figure 1c), provide the additional advantage that the pH during the treatment can be
buffered and more easily controlled. In addition, using the latter base enables to obtain the
active protonic form of the catalysts by a single calcination step, and makes the otherwise-
required ion exchange obsolete.
3. Conclusions
Despite the extended history of the synthesis and application of amorphous silica-alumina, the
role of the pH in the post-synthetic alumination of silica materials has never been studied. We
demonstrated within this contribution that by decoupling the pH from the Al concentration in
post-synthetic alumination of silica, unique aluminosilicate catalysts at high solid yields are
obtained. In a case study on widely-studied SBA-15, materials with preserved porosity,
unprecedented acidities, superior hydrothermal stability and superior catalytic performance
were prepared. The catalytic diversity of controlled aluminations was shown in several acid-
catalyzed reactions (toluene alkylation, α-pinene isomerization, alkylphenol dealkylation, and
ethyl lactate synthesis), where the materials outperformed established amorphous
aluminosilicate materials. Moreover, in the pyrolysis of LDPE plastics an enhanced
performance compared to both amorphous and crystalline aluminosilicates (zeolites) was
demonstrated. The developed pathways can be applied to any porous silica, as we have
verified by preparing aluminosilicate catalysts from several other ordered and non-ordered
amorphous silica sources. The presented synthetic insights open the door to the synthesis, in-
depth characterization, and application of a novel family of amorphous aluminosilicate
materials, in which a more efficient use of aluminum ensures an enhanced activity, selectivity,
and stability in catalytic conversions.
4. Experimental Section
17
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Synthesis of ordered silicas: SBA-15 was synthesized as stated in literature.[25] 20.0 g of
Pluronic P123 (Sigma-Aldrich) was dissolved in 635 g distilled H2O and 115 g concentrated
HCl (Fisher) (37 wt.%) at 35 °C. After a full P123 dissolution 43 g of tetraethyl orthosilicate
(TEOS, Acros) was added for synthesis of ~12 g of silica. This mixture was stirred for 20 h at
35 °C and afterwards filtered and washed. The recovered powder was dried overnight at 80 °C
and finally calcined at 550 °C for 6 h (ramp rate: 1 °C min-1). The synthesis procedures of
other ordered mesoporous silica MCM-41 and COK-12 are described in the supporting
information (Table S2). Parent materials are denoted as ‘a-P’. Herein ‘a’ is the type of parent
material: ‘S’ for SBA-15, ‘M’ for MCM-41 and ‘C’ for COK-12. The ‘P’ indicates that it
regards the parental untreated silica.
Commercial silicas and aluminosilicates: Two commercial amorphous (non-ordered) silicas
were used as parent material: silica gel (6 nm pore size, 230-400 mesh particle size, 40-63 µm
particle size; Sigma-Aldrich, denoted as ‘AS1-P’) and fumed silica (Aerosil 380, denoted as
‘AS2-P’). Several commercial aluminosilicate references for acidity and catalytic benchmarks
were used: amorphous silica alumina (Grace, Si/Al=6, denoted as ASA-6’), USY zeolite
CBV-780 (Zeolyst, Si/Al=40, denoted as ‘USY-40’), and USY zeolite CBV-600 (Zeolyst,
Si/Al=2.6, denoted as ‘USY-2.6’). In the case of these aluminosilicates the number in the
label corresponds to their Si/Al ratio.
Post-synthetic aluminations: Aluminations were performed at room temperature by adding 10
ml solutions of distilled water, or 0.1 M sodium hydroxide (Fisher), or 0.5 M ammonium
hydroxide (Chemlab) to 0.333 g of silica source and the appropriate amount of
Al(NO3)3·9H2O (Chemlab) or NaAlO2 (Sigma Aldrich), followed by stirring (300 rpm) for 30
min. In the case of aluminations with NaOH and NH4OH, Al(NO3)3·9H2O was used.
Aluminations with NaAlO2 were performed in the absence of an additional base. After the
alumination, samples were immediately suspended in 400 ml H2O, followed by a filtering and
washing step. The recovered powder was dried overnight at 80 °C and calcined at 550 °C for
18
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
6 h (ramp rate: 1°C min-1). The mass yields after alumination were calculated by dividing the
final mass after calcination by the product of the mass before alumination (0.333 g) and the
maximal amount of Al2O3 incorporated in the solid based on the applied aluminum
concentration (up to 0.040 g). The samples were converted to the protonic form by three
+
consecutive NH4 -ion exchanges of 6 h (1 g of silica per 100 ml NH4NO3 (Acros) solution of
0.1 M) with intermediate filtering and washing followed by a calcination as described above.
The hydrothermal stability of the selected samples was assessed by refluxing the solids (100
mg) in a round-bottom flask (100 ml) filled with distilled water (10 mL) at 100 °C for 22 h.
After treatment, solids were isolated by Büchner funnel filtration, followed by washing and
drying overnight at 60°C.
Sample coding after aluminations: Samples were labeled by the generic formula ‘a-x-b’. The
‘a’ refers to the used starting silica source. In the case no silica was added in the alumination
solution (Figure 1a), the ‘a’ is substituted with a ‘0’. The ‘x’ in ‘a-x-b’ refers to the initial
concentration of Al in the alumination solution ([Al]initial in mol l-1). Finally, ‘b’ refers to the
type of base used in the alumination: ‘NaOH’ for aluminations in aqueous NaOH, ‘NH4OH’
for aluminations in aqueous NH4OH, and ‘NaAlO2’ for aluminations with NaAlO2 in distilled
water. In the case no alkaline source was present, i.e. aluminations with Al(NO3)3·9H2O in
distilled water, ‘b’ was replaced with ‘H2O’. For example, SBA-15 treated in a 0.1 M NaOH
solution with 0.006 M Al(NO3)3·9H2O is denoted as ‘S-0.006-NaOH’. Similarly, SBA-15
aluminated in distilled water complemented with 0.03 M of Al(NO3)3·9H2O is referred to as
‘S-0.03-H2O’. Finally, silica gel treated in distilled water complemented with 0.02 M of
NaAlO2 is referred to as ‘AS1-0.02-NaAlO2’. An additional overview of the sample
nomenclature is provided in Table S4.
Characterization: Nitrogen physisorption isotherms were determined using a Tristar 3000
(Micromeritics) at -196 °C. Samples were degassed prior to analysis under N2 flow at 300 °C
overnight. Pore size distributions (PSDs) were determined using the NL-DFT (non-local
19
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
density functional theory) model for cylindrical pores in metal oxides for the adsorption
isotherm.[38] The standard deviations on the model were between 2 and 5 cm3 g-1. Specific
surface areas were determined using the BET model (Brunauer, Emmett, Teller).
Transmission electron microscopy (TEM) was carried out on a FEI Tecnai F30 with a field
emission gun at 300 kV. Prior to analysis, samples were dispersed in ethanol and dropped
onto a copper grid with a lacey carbon layer. For image processing the program Digital
Micrograph (Gatan Inc.) was used.
Bulk Si and Al contents (SiICP/AlICP) were determined using inductively coupled plasma
optical emission spectroscopy (ICP-OES) using analytical wavelengths of λ = 251.611 nm for
Si and λ = 396.153 nm for Al on a Varian 720-ES instrument. 50 mg of the samples was
mixed with 250 mg LiBO2 and heated for 10 min at 1000 °C. The molten material was then
dissolved in 0.42 M HNO3 and further diluted. The water content in the material was
measured on several accounts and an average of 10 wt.% was taken into account for
determination of the Al and Si content.
Surface Si and Al contents (SiXPS/AlXPS) were probed with X-ray photoelectron spectroscopy
(XPS) using a Specs Phoibos 100 with Mg Kα radiation (1253.6 eV) at 300 W in an energy
range of 0 – 1000 eV. The spectrometer is equipped with a hemispherical analyzer allowing
high sensitivity and high resolution experiments. Energy scale and binding energy were
calibrated with Cu and Au foils at the binding energies of Cu 2p3/2 (932.67 eV) and Au 4f7/2
(84.00 eV), respectively. Because of large gas adsorption, the sample chamber could only be
held at a base pressure of around 3·10-8 mbar after 1 h of pumping to degas the samples in a
pre-vacuum chamber. Spectra were acquired with a pass energy of 15 eV and a step size of
0.1 eV. The spectra were analyzed with the software CASA-XPS and the element
concentrations are calculated with the standard sensitivity factors provided by the software.
27Al magic angle spinning (MAS) NMR spectra were collected using a Bruker Avance III 400
MHz spectrometer equipped with a 9.4 T wide-bore magnet operating at a Larmor frequency
20
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
of 104.3 MHz. The samples were hydrated for 48 h over a saturated salt solution prior to
27
analysis and packed in 4-mm ZrO2 MAS rotors. Al MAS NMR spectra were recorded in a
double resonance probe at a spinning rate of 14 KHz. Spectra were obtained using a π/18
short rf pulse (~0.3 µs) calibrated using an aqueous solution of Al(NO3)3, corresponding to an
rf field strength of 104 kHz. The recycle delay was set to 1 s and the number of scans was
between 10 k and 30 k. Chemical shifts are quoted in ppm from the aqueous solution of
Al(NO3)3 (0 ppm).
Pyridine-probed Fourier transformed infrared (PP-FTIR) spectroscopy was executed using a
Nicolet 6700 spectrometer equipped with a DTGS detector. Samples were pressed into self-
supporting plates and degassed at 400 °C for 1 h in vacuum before measurements. Lewis and
Brønsted acid sites were analyzed using a pyridine probe. After evacuation, the samples were
saturated by 4–5 pulses of ~25 mbar of pyridine at 50 °C for 1 min. Physisorbed pyridine was
removed by heating to 150 °C. The spectra were collected at 150 °C after 20 min of
equilibration. The absorptions at 1450 and 1550 cm-1 correlated to the amount of Lewis (L)
and Brønsted (B) acid sites, respectively. The extinction coefficients used were similar to
those determined by Emeis.[39]
Catalysis: Alkylations of toluene with benzyl alcohol were performed by mixing 0.025 g of
catalyst with 4.98 ml (47 mmol) toluene (Fisher), 0.062 ml (0.6 mmol) benzyl alcohol (Sigma
Aldrich) and 0.048 ml (0.3 mmol) propyl cyclohexane (TCI Europe) as internal standard. The
catalyst powder was predried for 2 h at 300 °C (ramp 5 °C min-1) and the reaction was carried
out in closed Schott bottles (Duran) for 1 h under stirring at 120 °C. The reaction mixture was
analyzed with gas chromatography (Agilent 6850 series).
Low density polyethylene (Alfa Aesar, particle size = 500 µm, density = 0.92 g cm-3) catalytic
cracking was carried out in thermogravimetric analysis (TGA, Q500 TA Instruments). The
powdered polymer (6 mg) and catalyst (2 mg) were carefully weighed in a TGA crucible and
21
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
suitably mixed in order to obtain an intimate contact. The reaction was performed in N2 flow
(70 cm3 min-1) ramping the temperature from room temperature to 700 °C with 10 °C min-1.
The isomerization of α-pinene (Sigma-Aldrich) was carried out in a 50 ml Parr reactor with a
sampling device at 150 °C under 6 bar of nitrogen with a stirring speed of 750 rpm. A mixture
of substrate (20 g α-pinene) and catalyst (0.4 g) was heated to 100 °C, the first liquid sample
was taken subsequently. The reaction mixture was then further heated to 150 °C and more
samples were taken 10, 30, and 60 min after the first sample. All samples were analyzed on a
gas chromatograph (5890, Hewlett Packard) equipped with an HP1 column and a flame
ionization detector. Tetradecane (Sigma-Aldrich) was used as an external standard. Activity
(Aα-p) was determined by using the slope of the linear part of the conversion of α-pinene
versus the contact time.
For the synthesis of ethyl lactate (ELA, Sigma-Aldrich), 5 ml ethanol (Fisher) solutions with
0.4 M 1,3-dihydroxyacetone (Sigma-Aldrich) and naphthalene (internal standard, Alfa-Aesar)
were added to 0.2 g of catalyst. The reaction was performed in glass crimp cap vials at 110 °C
under stirring. The reaction mixture was analyzed with a gas chromatograph (Agilent 6850
series) equipped with an FID detector and a 30 m Agilent HP-5 column,. A calibration curve
versus naphthalene was used for the response factor. Initial rates of ELA formation (RELA
)
were determined by using the slope of the linear part of the ELA yield vs. contact time,
assessed at < 21% yield.
Dealkylation experiments were performed in a custom-built plug-flow fixed-bed reactor
equipped with 4 parallel quartz reactors (inner diameter of 3 mm). Typically, the quartz
reactors were filled with in total 120 mg of catalysts (sieve fraction: 0.125–0.250 mm),
yielding a catalyst bed height of ca. 13 mm. Propyl phenol (Sigma-Aldrich, >97%) and water
were brought to the gas phase using a nitrogen flow (20 ml min-1) passed through a saturator.
Afterwards these flows were mixed, yielding a gas mixture of the molar composition
0.86/0.12/0.02 (N2/H2O/alkylphenol), and passed through the reactors. Effluent gases were
22
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
characterized using an in-line gas chromatograph (Agilent 6850 series) equipped with a HP1
column and a FID detector.
Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.
Acknowledgements
D. V. and M. D. acknowledge support from an FWO post-doctoral fellowship. R. L.
acknowledges the IAP. N. N. acknowledges the KU Leuven (FLOF). L. M. thanks CICECO-
Aveiro Institute of Materials, POCI-01-0145-FEDER-007679 (FCT Ref. UID /CTM /
50011/2013 and PTDC/QEQ-QAN/6373/2014), financed by national funds through the
FCT/MEC and when appropriate co-financed by FEDER under the PT2020 Partnership
Agreement. The authors are also grateful to the Portuguese NMR Network (RNRMN).
Received: ((will be filled in by the editorial staff))
Revised: ((will be filled in by the editorial staff))
Published online: ((will be filled in by the editorial staff))
References
[1]
[2]
[3]
G. Busca, Chem. Rev. 2007, 107, 5366-5410.
W. Vermeiren, J.-P. Gilson, Top. Catal. 2009, 52, 1131-1161.
J. A. Scherzer, A. J. Gruia, in Hydrocracking Science and Technology, Marcel Dekker,
New York, 1996, p. 215.
[4]
C. Pistarino, E. Finocchio, M. A. Larrubia, B. Serra, S. Braggio, G. Busca, M. Baldi,
Ind. Eng. Chem. Res. 2001, 40, 3262-3269.
R. Mann, Catal. Today 1993, 18, 509-528.
E. Garrone, F. Fajula, in Acidity and Basicity (Eds.: H. G. Karge, J. Weitkamp),
Springer-Verlag Berlin Heidelberg, 2008.
[5]
[6]
[7]
[8]
G. Busca, in Metal Oxides (Ed.: J. L. G. Fierro), CRC Press, 2005, pp. 247-318.
E. J. M. Hensen, D. G. Poduval, P. C. M. M. Magusin, A. E. Coumans, J. A. R. van
Veen, J. Catal. 2010, 269, 201-218.
[9]
T. C. Keller, J. Arras, M. O. Haus, R. Hauert, A. Kenvin, J. Kenvin, J. Pérez-Ramírez,
J. Catal. 2016, 344, 757-767.
[10] V. C. F. Holm, G. C. Bailey, A. Clark, J. Phys. Chem. 1959, 63, 129-133.
[11] M. Caillot, A. Chaumonnot, M. Digne, J. A. Van Bokhoven, ChemCatChem 2014, 6,
832-841.
[12] M. Caillot, A. Chaumonnot, M. Digne, C. Poleunis, D. P. Debecker, J. A. van
Bokhoven, Microp. Mesop. Mater. 2014, 185, 179-189.
[13] R. Mokaya, W. Jones, Z. Luan, M. Alba, J. Klinowski, Catal. Lett. 1996, 37, 113-120.
[14] L. Y. Chen, Z. Ping, G. K. Chuah, S. Jaenicke, G. Simon, Microp. Mesop. Mater. 1999,
27, 231-242.
[15] R. Mokaya, W. Jones, J. Mater. Chem. 1999, 9, 555-561.
[16] Z. Luan, M. Hartmann, D. Zhao, W. Zhou, L. Kevan, Chem. Mater. 1999, 11, 1621-
1627.
[17] S. Biz, M. G. White, J. Phys. Chem. B 1999, 103, 8432-8442.
[18] K. G. Miesserov, Russ. Chem. Bull. 1956, 5, 1207-1211.
23
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
[19] H.-M. Kao, C.-C. Ting, S.-W. Chao, J. Mol. Catal. A. 2005, 235, 200-208.
[20] R. Locus, D. Verboekend, R. Zhong, K. Houthoofd, T. Jaumann, S. Oswald, L.
Giebeler, G. Baron, B. F. Sels, Chem. Mater. 2016, 28, 7731-7743.
[21] C. J. Brinker, G. W. Scherer, Sol-gel Science: The Physics and Chemistry of Sol-gel
Processing, Academic Press, San Diego, 1990.
[22] J. Van Aelst, et al., Adv. Funct. Mater. 2015, 25, 7130-7144.
[23] D. Zhai, L. Zhao, Y. Liu, J. Xu, B. Shen, J. Gao, Chem. Mater. 2015, 27, 67-74.
[24] D. Verboekend, J. Pérez-Ramírez, Chem. Eur. J. 2011, 17, 1137-1147.
[25] D. Zhao, Q. Huo, J. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc. 1998, 120,
6024-6036.
[26] T. Ennaert, J. Geboers, E. Gobechiya, C. M. Courtin, M. Kurttepeli, K. Houthoofd, C.
E. A. Kirschhock, P. C. M. M. Magusin, S. Bals, P. A. Jacobs, B. F. Sels, ACS Catal.
2015, 5, 754-768.
[27] R. Mokaya, J. Phys. Chem. B, 2000, 104, 8279–8286 .
[28] E. J. M. Hensen, D. G. Poduval, V. Degirmenci, D. A. J. M. Ligthart, W. Chen, F.
Maugé, M. S. Rigutto, J. A. R. van Veen, J. Phys. Chem. C 2012, 116, 21416-21429.
[29] D. Verboekend, M. Milina, S. Mitchell, J. Pérez-Ramírez, Cryst. Growth Des. 2013,
13, 5025-5035.
[30] J. Pérez-Ramírez, S. Abelló, A. Bonilla, J. C. Groen, Adv. Funct. Mater. 2009, 19,
164-172.
[31] D. W. Park, E. Y. Hwang, J. R. Kim, J. K. Choi, Y. A. Kim, H. C. Woo, Polym.
Degrad. Stab. 1999, 65, 193-198.
[32] F. de Clippel, M. Dusselier, R. Van Rompaey, P. Vanelderen, J. Dijkmans, E.
Makshina, L. Giebeler, S. Oswald, G. V. Baron, J. F. M. Denayer, P. P. Pescarmona, P.
A. Jacobs, B. F. Sels, J. Am. Chem. Soc. 2012, 134, 10089-10101.
[33] P. P. Pescarmona, K. P. F. Janssen, C. Delaet, C. Stroobants, K. Houthoofd, A.
Philippaerts, C. De Jonghe, J. S. Paul, P. A. Jacobs, B. F. Sels, Green Chem. 2010, 12,
1083-1089.
[34] J. Dijkmans, M. Dusselier, D. Gabriëls, K. Houthoofd, P. C. M. M. Magusin, S. Huang,
Y. Pontikes, M. Trekels, A. Vantomme, L. Giebeler, S. Oswald, B. F. Sels, ACS Catal.
2015, 5, 928-940.
[35] M. Dusselier, P. Van Wouwe, A. Dewaele, E. Makshina, B. F. Sels, Energy Environ.
Sci. 2013, 6, 1415-1442.
[36] N. Nuttens, D. Verboekend, A. Deneyer, J. Van Aelst, B. F. Sels, ChemSusChem 2015,
8, 1197-1205.
[37] D. Verboekend, Y. Liao, W. Schutyser, B. F. Sels, Green Chem. 2016, 18, 297-306.
[38] M. Jaroniec, M. Kruk, J. P. Olivier, S. Koch, in Stud. Surf. Sci. Catal., Vol. 128 (Eds.:
G. Kreysa, J. P. Baselt, K. K. Unger), Elsevier, 2000, pp. 71-80.
[39] C. A. Emeis, J. Catal. 1993, 141, 347-354.
24
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Figure 1. (a) Measured pH values in aqueous solutions of different Al sources, with and
without base addition such as NaOH (0-x-NaOH), NaAlO2 (0-x-NaAlO2), or NH4OH (0-x-
NH4OH). The ‘x’ refers to the applied concentration of aluminum ([Al]initial). (b,c) pH
evolution during aluminations of SBA-15. In the case H2O, NaOH, or NH4OH is used in a
synthesis or solution, [Al]initial refers to the Al(NO3)3 concentration. Conversely, in the case
NaAlO2 is used, Al(NO3)3 is not used and [Al]initial refers to the NaAlO2 concentration.
25
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Table 1. Mass yields, composition, and porous properties of parental and aluminated silica.
e
c
d
Sample
Yielda
(%)
Si/Alb
SBET
dpore
Vpore
(mol mol-1)
(m2 g-1)
679 (-36%)g
352
(nm)
7
(cm3 g-1)
0.63
0.46
0.57
0.50
0.44
0.41
0.51
0.72
0.43
0.56
0.46
0.43
0.53
0.48
0.48
0.46
0.74
0.76
0.61
0.69
0.60
0.74
0.55
1.08
1.13
0.99
1.24
S-P
100
77
80
82
91
89
100
100
97
87
94
95
95
94
95
97
100
51
74
100
99
97
100
52
83
90
-
no Al
no Al
243 (45)
81 (-)
39 (21)
30 (14)
20 (15)
-
S-0-NaOH
8
S-0.002-NaOH
S-0.006-NaOH
S-0.015-NaOH
S-0.02-NaOH
S-0.03-NaOH
S-0.04-NaOH
S-0.03-H2O
327
8
301
7
292
6
308
7
497 (+2%)
770
7
7
-
418
6
S-0.006-NH4OH
S-0.02-NH4OH
S-0.03-NH4OH
S-0.006-NaAlO2
S-0.015-NaAlO2
S-0.02-NaAlO2
S-0.03-NaAlO2
AS1-P
87
404
8
29
354 (+1%)
340
7
19
7
137
54
465
8
413 (+10%)
408
7
41
7
-
364
7
-
473 (-37%)
395
10
12
10
10
12
12
26
32
32
32
26
AS1-0.006-NaOH
AS1-0.02-NaOH
AS1-0.03-NaOH
AS1-0.02-NH4OH
AS1-0.015-NaAlO2
AS2-P
81h
30h
22h
29h
55h
-
375
370 (+2%)
321
387
316
AS2-0.006-NaOH
AS2-0.015-NaOH
AS2-0.03-NaOH
ASA-6
81h
41h
22h
6
239
262
237
403 (-16%)
aMass yield after alumination divided by mass before alumination combined with the maximal
b
amount of Al2O3 incorporated. Si/Al ratio of the bulk (SiICP/AlICP) and the surface
(SiXPS/AlXPS, the latter between brackets). cSpecific surface area (BET method). dAverage pore
diameter(s) on the basis of the adsorption NL-DFT pore size distribution. eTotal pore volume.
g
c-fN2 physisorption. The values in brackets indicate the change in BET surface area after a
hydrothermal treatment (conditions in the experimental section).hValues were not measured,
but calculated assuming the Al incorporation was the same (91%) as in the corresponding
SBA-15 samples (following Figure 2c).
26
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Figure 2. (a) Mass yields (Y) after alumination of SBA-15 (S-x-b) and amorphous silica
(AS1-x-b) using various aqueous solutions. The ‘x’ refers to the applied initial Al
concentration in solution ([Al]initial), the ‘b’ to the base present in solution. (b) Amount of
aluminum incorporated into the bulk of the silica material (AlICP) compared to the theoretical
maximal amount of Al that is incorporated at the obtained mass yield (Almax). (c) AlICP/Almax
as a function of [Al]initial. The legend in (a) also applies in (b,c). (d) Bulk and surface Si/Al
ratios (SiICP/AlICP and SiXPS/AlXPS, respectively) in S-x-NaOH samples.
27
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Figure 3. N2 physisorption isotherms of (a) SBA-15 (S-x-NaOH) and (c) amorphous silica
(AS1-x-b) samples before and after aluminations, and the accompanying adsorption NL-DFT
PSDs (b, d, respectively). The ‘x’ refers to the applied initial Al concentration in solution
([Al]initial), the ‘b’ to the base present in solution.
28
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701660
ChemCatChem
Figure 4. Bright-field TEM micrographs of selected SBA-15-derived samples. The scale bars
are 50 nm.
29
This article is protected by copyright. All rights reserved.