1
32
T. Moteki et al. / Journal of Catalysis 378 (2019) 131–139
of SBA-15 is slightly slower rather than that of another representa-
tive 2D-hexagonal mesoporous silica, MCM-41, maybe due to its
thicker pore wall) [19,20].
Two decades ago, Xia and Mokaya demonstrated the catalytic
reaction over nitrided mesoporous silica MCM-48 as a base catalyst
would be explained by the formation of imine species. The forma-
tion of imine species was confirmed by FT-IR measurement. In
other reactions, the catalytic activity was in the order of pri-
mary > secondary ꢀ tertiary silanamine sites. Due to the difference
of electronegativity between C and Si atoms, the basicity of silana-
mine species should be different from the conventional organic
amines. Furthermore, the catalytic activity did not follow the basic-
ity generally assumed by the electronegativity of H and Si atoms.
The order would be explained by steric effect originated from the
formation of internal secondary and tertiary silanamine site inside
the pore wall, which was also suggested by spectroscopic studies.
[
[
22]. Since then, several reactions (e.g., Knoevenagel condensation,
17,18,22,24,31,33] transesterification, [31] Morita-Baylis-Hillman
reaction [29]) has been catalyzed by nitirded mesoporous silicas.
Moreover, further methyl-functionalization over silanamine sites
enabled to catalyze Knoevenagel condensation of high pK
a
reactant
(
i.e., diethyl malonate) [28] and CO addition reaction [30]. It has
2
been shown that the amount of introduced nitrogen did not
explain their catalytic activities in a simple manner [17,31,33].
Instead of the sample that contains more nitrogen sites prepared
at severe nitridation condition, the sample that contains less nitro-
gen sites prepared at low or moderate nitridation conditions
showed better catalytic results [17,31,33]. Thus, catalytic activity
of each site would assume to be different, and large contribution
of primary silanamine sites has been supposed in some papers
2. Experimental
2.1. SBA-15 preparation
The mother mesoporous silica, SBA-15, was prepared by a con-
ventional hydrothermal method. All the commercial chemicals
were used without purification. In a typical synthesis, 1.64 g of tri-
block copolymer, PluronicÒ P123 (Sigma-Aldrich, average Mn of
5800), was dissolved into a 60.63 g of aqueous hydrochloric acid
solution (FUJIFILM Wako Pure Chemical, volumetric analysis
grade) in a 250 ml PP bottle. The mixture was stirred with a stirring
bar at 313 K for 3 h to prepare a homogeneous solution. Then,
3.47 g of tetraethoxysilane (FUJIFILM Wako Pure Chemical,
>95.0%) was added dropwise to the solution under stirring and
hydrolyzed at 313 K for 24 h. The final chemical composition was
[
31,33]. Detailed quantitative analysis of each silanamine site
and their contribution calculation toward catalytic reaction would
be necessary for further design of the catalyst.
The formation of different silanamine species has been investi-
gated via direct and indirect approaches, e.g., FT-IR, [17,18,21–
2
3,25,26,31] solid-state MAS NMR, [14–19,24–27,31–33] ESR,
27] CO probe technique, [20,21,24] and XPS [17,19,31,33].
Although solid-state Si MAS NMR has been well conducted to
detect SiO species, it only gives indirect information in the for-
[
2
2
9
N
x y
2 2 2 5
1SiO /0.017P123/6HCl/190H O/4C H OH. The mixture was sub-
mation of silanamine species. Thus, it should be studied together
with other techniques. FT-IR measurement has also been well con-
ducted to prove the formation of silanamine species by the obser-
jected to a hydrothermal treatment at 373 K for 24 h under static
condition in an electric oven. The obtained solid precipitate was fil-
tered and washed with deionized water. The samples were cal-
cined in air using a muffle furnace to remove the occluded
vation of N–H related vibrations of primary and secondary
silanamins [17,18,21–23,25,26,31]. The presence of tertiary silana-
mine was, however, not confirmed due to the lack of unique vibra-
tion. In 2015, detailed analysis via solid-state 15N MAS NMR was
conducted by Polshettiwar and coworkers [32,33]. Due to the very
low relative signal intensity of 15N (about one-thousandth of H),
ꢁ1
surfactant, P123, at 823 K for 6 h with a ramping rate of 1 K min
.
2.2. NSBA-15 preparation
1
Nitridation of the prepared SBA-15 sample was conducted
under high temperature with a flow of pure ammonia gas using a
home-made horizontal tubular reactor (/ 50 mm) with PID tem-
perature controller. In a typical procedure, 1.0 g of calcined SBA-
15 sample was put in a ceramic boat and placed in the tubular
1
dynamic nuclear polarization technique [32] or H-detected
approach [33] was used for the measurement. The technique
directly and successfully revealed the formation of secondary and
tertiary silanamines, and together with catalytic results, primary
silanamine was supposed to be the most active catalytic site for
one of Knoevenagel condensation [32,33]. The quantitative analy-
sis of each silanamine site is, however, not yet insufficient, and
more general understanding would be necessary for the further
design and use of nitrided silica catalysts.
Here in this study, a series of nitrided mesoporous silica SBA-15
samples at different nitridation temperature was synthesized, and
quantitative analysis of formed silanamine species was conducted.
The amount of nitrogen content was determined by both bulk and
surface analyses (i.e., chemical analysis and XPS, respectively). For
the quantitative analysis of each silanamine species, we used nar-
row scan spectra of XPS. Although XPS is a strong technique to
directly identify the chemical state of an element, the previous
studies conducting the measurement have lacked the detailed dis-
cussions [17,19,31,33]. We have conducted the careful deconvolu-
tion of the spectra of N 1s level and the amount of each silanamine
species was decided. The prepared samples were applied to several
aldol and Knoevenagel condensation reactions as catalysts. The
contribution of each silanamine site to the reaction was decon-
volved by mathematical least square method using a determinant
of matrix representation. Primary silanamine site showed two
order of magnitude high activity for the aldol reaction involving
aldehydes than secondary and tertiary silanamine sites, which
reactor. After the reactor was purged by flow in-house N
2
gas, it
was heated with a ramping rate of 5 K min . At 573 K, the flow
gas was switched from in-house N to ammonia (Jyotou Gas,
ꢁ1
2
ꢁ1
99.999%) with a flow rate of 2.0 L min . The temperature was
increased up to 823–1273 K and kept for 10 h. The obtained sam-
ples are denoted as ‘‘NSBA-15-X”, where X represents the nitrida-
tion temperature.
2.3. Catalyst characterizations
Powder X-ray diffraction (XRD) measurement was conducted
on a Rint 2000 (Rigaku) using a CuKa radiation (40 kV, 40 mA) with
a step size of 0.02° and a 4 s per step between 0.5° and 3° (2h).
Nitrogen sorption measurement was conducted on a Quadrasorb-
evo (Quantachrome) at 77 K. Samples were pre-treated at 573 K
for 3 h under vacuum. Thermal gravimetric analysis was conducted
on a Thermo Plus TG 8120 (Rigaku) with a ramping rate of
ꢁ1
2 3
1 K min from 298 to 1173 K under air condition using Al O as
a reference. Elemental analysis (CHN analysis) was conducted on
a Vario Micro Cube (Elementar Analysensysteme GmbH). X-ray
photoelectron spectroscopy was conducted on a PHI Quantera
SXM (ULVAC-PHI Inc.). Pelletized samples were hold on a sample
stage with a silver paste. The spectra were calibrated with a peak