The cooperative effect of Lewis and Br?nsted acid sites on Sn-MCM-41 catalysts for the conversion of 1,3-dihydroxyacetone to ethyl lactate

Article abstract of DOI:10.1039/c9gc00820a

Lactic acid and alkyl lactates are widely applied in the production of food, cosmetics, pharmaceuticals, organic synthesis and biodegradable polymers. They can be prepared via one-pot synthesis from renewable trioses, such as dihydroxyacetone (DHA). Br?nsted-Lewis bifunctional solid acids (BAS & LAS) can promote the reaction via a two-step cascade reaction mechanism. BAS catalyses the dehydration of DHA, resulting in the formation of pyruvaldehyde (PA) via the rearrangement of the enol form. Upon alcohol addition, PA can be converted to the desired alkyl lactates at LAS or to pyruvaldehyde hemiacetal (PAHA) at strong BAS. The density and strength control of Br?nsted acid sites (BAS) and Lewis acid sites (LAS) and the optimization of their cooperation are essential for the efficient conversion of trioses to the target products. Here, we prepared a series of Sn-containing mesoporous MCM-41 catalysts with various BAS/LAS ratios by room temperature techniques. Sn-doped [Si]MCM-41 having a lower BAS/LAS ratio in this research shows a high initial selectivity to ethyl lactate (EL) and similar EL yield in 6 hours as the reported best Sn catalyst Sn-grafted [Si]MCM-41/carbon network materials in DHA conversion. A relatively large density of LAS in Sn-doped [Si]MCM-41 causes a fast conversion of PA to EL, while the overall yield has been limited by the BAS density for the DHA conversion. New H-form [Sn]MCM-41, having a suitable density of LAS and weak BAS and an optimized BAS/LAS ratio, provides a 100% yield of ethyl lactate in the catalytic conversion of DHA in ethanol within 30 min, showing a superior performance hitherto.

Full text of DOI:10.1039/c9gc00820a

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: K. Kim, Z. Wang,
Y. Jiang, M. Hunger and J. Huang, Green Chem., 2019, DOI: 10.1039/C9GC00820A.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Page 1 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
The cooperative effect of Lewis and Brønsted acid
sites on Sn-MCM-41catalysts for the conversion of
1,3-dihydroxyacetone to ethyl lactate
Kyung Duk Kim1, Zichun Wang1,2*, Yijiao Jiang2, Michael Hunger3, Jun Huang1,*
1. Laboratory for Catalysis Engineering, School of Chemical and Biomolecular
Engineering, The University of Sydney, NSW 2006, Australia
2. School of Engineering, Macquarie University, Sydney, NSW 2109, Australia
3. Institute of Chemical Technology, University of Stuttgart, 70550 Stuttgart, Germany
1

Green Chemistry
Page 2 of 41
View Article Online
DOI: 10.1039/C9GC00820A
ABSTRACT Lactic acid and alkyl lactates are widely applied in the production of food,
cosmetics, pharmaceuticals, organic synthesis and biodegradable polymers. They can be prepared
via one-pot synthesis from renewable trioses, such as dihydroxyacetone (DHA). Brønsted-
Lewis bifunctional solid acids (BAS & LAS) can promote the reaction via a two-step cascade
reaction mechanism. BAS catalyses the dehydration of DHA, resulting in the formation of
pyrualdehyde (PA) via the rearrangement of enol. Upon alcohol addition, PA can be
converted to the desired alkyl lactates at LAS or to pyruvaldehyde hemiacetal (PAHA) at
strong BAS. The density and strength control of Brønsted acid sites (BAS) and Lewis acid sites
(LAS) and the optimization of their cooperation are essential for the efficient conversion of triose
to the target products. Here, we prepared a series of Sn-containing mesoporous MCM-
41 catalysts with various BAS/LAS ratios by room temperature techniques. Sn-doped
[Si]MCM-41 having the lower BAS/LAS ratio in this research shows the high initial
selectivity to EL and the similar EL yield in 6 hours as the reported best Sn catalyst Sn-
grafted [Si]MCM-41/ carbon network materials in DHA conversion. A relative large density
of LAS in Sn-doped [Si]MCM-41 causes a fast consumption of PA to EL, while the overall
yield has been limited by the BAS density for the DHA conversion. New H-form [Sn]MCM-
2

Page 3 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
41, having the suitable density of LAS and weak BAS and optimized BAS/LAS ratio,
provides a 100 % yield of ethyl lactate in the catalytic conversion of DHA in ethanol within
30 min, showing superior performance hitherto.
KEYWORDS: Acidity cooperativity; Sn-MCM-41; Solid-state NMR; dihydroxyacetone
conversion.
3

Green Chemistry
Page 4 of 41
View Article Online
DOI: 10.1039/C9GC00820A
1. Introduction
Recently, chemical production from renewable biomass has become increasingly
important in the chemical industry due to the depletion of fossil fuels and environmental
concerns.1 In large scale, transesterification of renewable triglycerides (or other esters)
with alcohols (e.g. methanol) has been commercialized to produce alkyl esters (biodiesel)
with glycerol as the main by-product.2 Million tons of glycerol has been generated
annually and still increasing rapidly. As a versatile platform molecule, glycerol can be
converted into various value-added chemicals via dehydration, oxygenation, and
hydrogenation reaction.3-5 Among them, trioses (dihydroxyacetone (DHA) or
glyceraldehyde (GLY)), the primary oxidation products of glycerol,6-10 are the key
intermediates in organic synthesis.11
Trioses can be catalytically converted into lactic acid in water or alkyl lactates in
alcohols,12-13 which is of great interest as a cost-effective alternative to the time-
consuming (2~4 days) fermentation of plant biomass with complicated purification
process.14-16 Lactic acids are key starting precursor for prepare polylactic acid or
12
polylactide (PLA). PLA is one of the main candidates in the production of biodegradable
4

Page 5 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
polymers and their demand is expected to reach 3 million tonnes in 2020.12, 17 Lactic acids
are also interesting building blocks in the preparation of acrylic acid, propylene glycol,
etc.11 Moreover, alkyl lactates are important raw materials used in food, skin care,
cosmetics and healthcare formulations.18
One-pot synthesis of lactic acids and its derivatives from the conversion of trioses is of
great interest. Homogeneous catalysts, such as H2SO4,19 NaOH,20-21 and ZnSO4,22 have
been investigated. The use of homogeneous catalysts is suffered from extremely harsh
conditions (e.g. high temperatures and pressure), toxic or corrosive catalysts and
difficulties of catalyst separation and regeneration. To address these problems,
heterogeneous catalysts are extensively studied. Non-acidic and Brønsted acidic zeolites
are inactive for the reaction.23-25 The conversion of DHA to lactic acids and its derivatives
can be catalyzed by Lewis acidic catalysts, such as metal (Zr, Ti, Sn and Ga)-doped
molecular sieves.24-27 DHA can be converted into methyl lactate in methanol with a yield >
99% over Sn-Beta zeolites at 80 oC after a reaction time of 24 h.24 Moreover, Brønsted-
Lewis bifunctional solid acids (BAS-LAS) like dealuminated zeolites (Si/Al =2.6,
framework Al of 27%) are active for the conversion of DHA in various alcohols at 90 oC
5

Green Chemistry
Page 6 of 41
View Article Online
DOI: 10.1039/C9GC00820A
after 6 h reaction, with alkyl lactate yield of 69-83%.23 Recently, Sn-grafted mesoporous
[Si]MCM-41, filled with a polyaromatic to graphite-like carbon network, was applied as a
BAS-LAS bifunctional catalyst, where Sn(IV) species and oxygen-rich carbons act as LAS
and BAS, respectively.13 It exhibit an extremely high ethyl lactate yield (100 %), compared
with those having exclusive BAS (23.4 %) or LAS (1.25 %) under the same conditions.
Mechanistic studies via isotope labelling experiments suggested that the reaction
proceeds via the dehydration of primary hydroxyl groups of trioses to form pyruvic
aldehyde (PA) on BAS, and then LAS catalyze the conversion of PA into lactic acid and
derivatives via nucleophilic addition of a water or alcohol molecule to carbonyl carbon
atom of aldehyde, followed by a 1,2-hydride shift via an intramolecular Cannizzaro
reaction mechanism.23 This demonstrated the existence of both BAS and LAS is a
prerequisite to a sequential two-step reaction for fast and selective conversion of trioses
into alkyl lactates and lactic acid. Non-acidic to weakly acidic SiOH groups on pure silica
are inactive for the conversion of DHA,25 while strong BAS can promote the DHA
dehydration and also the formation of acetals as the major by-products.23, 28 Therefore, the
6

Page 7 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
density and strength control of BAS and LAS and the optimization of their ratio are essential for
the efficient conversion of triose to lactic acid or alkyl lactate.
In this work, we prepared a series of mesoporous Sn-MCM-41 catalysts through room
temperature synthesis methods. These catalysts are featured with various BAS and LAS
densities and ratios, evidenced by quantitative NMR spectroscopy with probe molecules.
Low BAS/LAS ratios can promote the selectivity in the conversion of trioses to lactates.
H-form Sn-doped MCM-41, having the suitable density of LAS and weak BAS and an
optimized BAS/LAS ratio, provides a 100 % yield of ethyl lactate in the catalytic
conversion of DHA in ethanol within 30 min, showing superior performance hitherto.
2. Experimental
Materials and Agents. All chemicals used for catalyst preparation and catalytic test,
including tetramethylammonium silicate (99.9%), hexadecyltrimethyl-ammonium bromide
(CTABr, 95%), SnCl4·5H2O (98%), tetraethyl orthosilicate (TEOS, 98%), ammonium
hydroxide solution (28% NH3 in H2O), hexadecyltrimethylammoniumchloride solution
(CTACl, ~ 25% in H2O), NaOH, dihydroxyacetone (DHA, 97%), and ethanol (99%), were
purchased from Sigma-Aldrich
7

Green Chemistry
Page 8 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Catalyst preparation. H-[Sn]MCM-41 catalyst: CTACl was mixed with TEOS in a volume
ratio of 1:1 in demineralized water (500 mL) and then mixed with SnCl4 solution.
Ammonium hydroxide solution was dropped into the mixed solution and stirred at room
temperature to form a white gel, followed by vigorous stirring for 1 h. The solids were
recovered by filtration, washing with distilled water, and drying at 333 K in air. Finally, the
as-synthesize samples were calcined at 823 K in air for 6 h to completely remove organic
compounds.
Na-[Sn]MCM-41 catalyst was prepared by the ion-exchange of H/[Sn]MCM-41 (1 g)
with an NaNO3 aqueous solution (0.1 M, 500 ml) at 353 K for 3 h under stirring.
Subsequently, the mixture was filtered and washed with deionized water, followed by
drying at 353 K overnight. After repeating the above mentioned process for 3 times,
Na/[Sn]MCM-41 was obtained.
H-[Si]MCM-41 catalyst and Na-[Si]MCM-41 catalyst were prepared in a similar manner,
by an one-step room-temperature preparation technique.29 simply by using NaOH to
replace ammonium hydroxide solution. Typically for H-[Si]MCM-41, CTACl was mixed
with an ammonium hydroxide solution and TEOS in a volume ratio of 1:1:1 in
8

Page 9 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
demineralized water (500 mL) and stirred at room temperature to form a white gel,
followed by vigorous stirring for 1 h. The resulting solids were collected by filtration,
washed with distilled water, and then dried in an oven at 353 K. Finally, the obtained H-
[Si]MCM-41 catalyst was calcinated at 823 K for 6 h with air. Sn- doped MCM-41 was
obtained by adding Na-[Si]MCM-41 or H-[Si]MCM-41 into a 0.04 M aqueous solution of
SnCl2·2H2O (98% Sigma-Aldrich), with a sample/liquid ratio of 10 g/L. The Sn-exchange
was performed at room temperature under stirring for 24 h. The obtained catalysts (Sn/H-
[Si]MCM-41 and Sn/Na-[Si]MCM-41) were dried at 353 K and then calcined at 823 K for
3h.
Physical characterization. Powder X-ray diffraction (PXRD) patterns were collected by
a Rigaku Multiflex diffractometer using a nickel-filtered Cu Kα radiation beam (40 kV, 30
mA). Nitrogen adsorption-desorption measurements (BET) were performed at 77 K using
a 4000e NOVA system to determine the specific surface areas. Before measurements,
the samples were degassed at 473 K under vacuum to remove adsorbents from the
surface. The primary particle size and specific surface areas were determined by BET
analysis. The elemental analysis of the Si/Al ratios was confirmed by inductively coupled
9

Green Chemistry
Page 10 of 41
View Article Online
DOI: 10.1039/C9GC00820A
plasma atomic emission spectroscopy (ICP/AES) using an Optima 4300 DV instrument
(Perkin-Elmer).
Solid-state NMR investigation. For the 29Si MAS NMR investigations, all samples were
fully hydrated by exposure to the saturated vapour of Ca(NO3)2 solution at ambient
temperature overnight in a desiccator. 29Si MAS NMR measurements were carried out on
a Bruker Avance III 400 WB spectrometer at the resonance frequency of 79.5 MHz and
with the sample spinning rate of 4 kHz using 7 mm MAS rotor. For29Si MAS NMR, single-
pulse π /2 excitation, high-power proton decoupling, and the recycle delay of 20 s were
applied.
Before the 1H and 13C MAS NMR experiments, the samples filled in glass tubes were
dehydrated for 12 h at 723 K under a pressure of less than 10-2 mbar. These dehydrated
samples were sealed in the glass tubes or directly loaded with ammonia or acetone-2-13C
(99.5 % 13C-enriched, Sigma-Aldrich) on a vacuum line. Subsequently, the loaded
samples were evacuated at 373 K for 1 h (for ammonia) or at room temperature for 2 h
(for acetone) to remove weakly physisorbed probe molecules. Subsequently, the samples
10

Page 11 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
were transferred into the MAS NMR rotors under dry nitrogen gas inside a glove box. For
the 31P MAS NMR experiments, the mixture of dehydrated catalysts (50 mg) and a known
amount of trimethylphosphine oxide (TMPO) was sealed inside a rotor with an O-ring-
containing Marcor-Cap. Subsequently, the TMPO-loaded samples inside the sealed
rotors were heated at 433 K for 2 hours for reaching a homogeneous distribution of the
TMPO probe molecules on the sample materials.
1H, 13C and, 31P MAS NMR investigations were performed on the same spectrometer
at resonance frequencies of 400.1, 100.6, and 161.9 MHz, respectively, and with the
sample spinning rate of 8 kHz (1H, 13C) or 10 kHz (31P) using 4 mm MAS rotors. The 31P
MAS NMR spectra were obtained with single π/2 pulse excitation, high-power proton
decoupling, and a repetition time of 30 s. The 1H MAS NMR spectra were obtained after
1
single-pulse π/2 excitation with a repetition time of 20 s. Quantitative H MAS NMR
measurements were conducted using zeolite H,Na-Y (35 % ion-exchanged) as an
external intensity standard. 13C cross-polarization (CP) MAS NMR spectra were recorded
with the contact time of 4 ms and the repetition time of 4 s.
11

Green Chemistry
Page 12 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Catalyst Activity Measurement. The obtained catalysts were applied in the conversion
of DHA to ethyl lactated in ethanol. Prior to the reaction, the catalysts (0.05 g) were
activated at 723 K under a N2 flow of 150 mL/min for 3h. In a typical run, the activated
catalyst (0.05 g) was transferred into a pressure vessel, together with an ethanol solution
(5 mL) containing DHA (0.4 M). The reaction was carried out in a tightly closed pressure
vessel, under controlled temperature for 6 h under stirring. The reaction products were
collected at regular time intervals and prepared in ethanol for GC analysis. A Shimadzu
GCMS-QP2010 Ultra furnished with a RTX-5MS capillary column (30 m × 0.25 mm ×
0.25m) and a Shimadzu GC-FID equipped with a RTX-5 capillary column (25 m × 0.32
mm × 5 m) were employed for the product determination and quantification, respectively.
The selectivity Si to specific product i was calculated as Si (%) = 100 × (i) / [(DHA)0-
(DHA)], where (i) is the molar concentration of the product and (DHA)0 and (DHA)
correspond to the molar concentration of DHA before and after the reaction, respectively.
In the catalyst recycling experiments (five cycles), the catalyst was separated from the
reaction mixture by centrifuge. The catalyst was washed with ethanol for three times to
remove residual and organics, followed by drying at 353 K under vacuum. Then the DHA
12

Page 13 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
conversion experiments were carried out using the recovered catalyst with the same
procedure as described above.
3. Results and Discussions
The structural and textural properties. As shown in Table 1, the BET surface areas (663
m2g-1 and 675 m2g-1) and average pore size (2.5 nm and 2.7 nm) of Sn-doped catalysts
(H-[Sn]MCM-41 and Na-[Sn]MCM-41), prepared by the direct incorporation of Sn into
[Si]MCM-41 framework by sol-gel technique, are typically for [Sn]MCM-41 reported in
literature.30 Sn-exchanged siliceous [Si]MCM-41 catalysts (Sn/H-[Si]MCM-41 and Sn/Na-
[Si]MCM-41) exhibit similar surface areas to the sol-gel prepared [Sn]MCM-41, however,
much lower than the surface area of parent [Si]MCM-41 (1042 m2g-1).29 Significantly, the
average pore sizes of Sn-exchanged [Si]MCM-41 catalysts (1.7 nm for H-form and 1.4
nm for Na-form, respectively) is much lower than those of the sol-gel ones. This could be
attributed to the extra-framework Sn (EFSn) cluster located inside the pores and
channels. The elemental analysis examined by ICP emission spectrometry of samples
did not show a change in tin content. (Table 1 and Table S1).
13

Green Chemistry
Page 14 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Table 1. Summary of surface area, average pore size, distribution of surface OH groups
and Brønsted acid sites on [Sn]MCM-41 catalysts, determined by BET analysis and solid-
state NMR characterization.
Total
Average
pore
Density of Density
of
Si/Sn a
Surfac density of
e area OH
(m2g-1)b groups
(mmol/g)c
Brønsted
Lewis acid
Catalyst
size
(nm)b
acid
sites sites
(mmol/g)c
(mmol/g)d
H-[Sn]MCM-41
Na-[Sn]MCM-41
Sn/H-[Si]MCM-41
9.9
8.2
8.0
2.5
2.7
2.7
663
675
663
2.6
2.0
1.6
4.3 × 10-2
0.72 × 10-2
0.24 × 10-2
1.8 × 10-2
0.6 × 10-2
0.6 × 10-2
Sn/Na-[Si]MCM-41 6.1
2.4
626
1.5
0.12 × 10-2
0.55 × 10-2
a
c
Determined from ICP.b Determined from the adsorption of the N2 isotherms. Total
density of OH groups and density of BAS are determined from 1H MAS NMR spectra of the
d
dehydrated and ammonia-loaded catalysts, respectively.
The density of LAS is
determined from the density of BAS and the BAS/LAS ratio obtained from 31P MAS NMR
spectra.
As shown in Figure 1, the framework structure has been characterized by powder XRD
patterns of all samples. The strong (100) reflections and weak (110) and (200) reflections
of these materials indicates a hexagonal framework structure with long-range order,
which is typically reported for the MCM-41 materials.29, 31 The shift of the broad (100)
14

Page 15 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
reflections in Sn-doped catalysts compared with that of Sn-grafted siliceous catalysts
could be an indication of the successful incorporation of Sn into the silica framework. The
weak (110) and (200) reflections of the Sn-doped MCM-41 compared to Sn-grafted MCM-
41, particularly the absence in Na-[Sn]MCM-41, indicates that the strong disturbance of
long-range order, as often reported for other metal-doped MCM-41 materials.29
100
110
200
H-[Sn]MCM-41
Na-[Sn]MCM-41
Sn/H-[Si]MCM-41
Sn/Na-[Si]MCM-41
1
2
3
4
5
6
7
8
2Theta
Figure 1. Small-angle XRD patterns of Sn-containing samples.
15

Green Chemistry
Page 16 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Local structure and acidity characterization. The effect of Sn-introduction on the local
coordination of Si atoms in the MCM-41 framework was studied by 29Si MAS NMR
spectroscopy as shown in Figure S1. In the deconvolution spectra (Figure S1, bottom),
three signals were identified at δ = -110 ppm, -100 ppm and -91 ppm, which are typically
assigned to Q4 (Si(OSi)4) species, Q3 (Si(OSi)3OH) species and Q2 (Si(OSi)2(OH)2)
species for non-doped [Si]MCM-41 materials.32 A strong broadening of the 29Si MAS NMR
signals in all Sn-containing MCM-41s was observed compared to [Si]MCM-41 materials,
which is similar to that reported for Al-doped [Si]MCM-41 earlier.29 This is due to the
strong disturbance of the local Si structure by the incorporation of heteroatoms into the
[Si]MCM-41 framework, which demonstrates that Sn atoms are incorporated into the
MCM-41 framework.
It has been widely accepted that Al incorporated into the silica framework can interact
with neighbouring silanols to enhance their acid strength, contributing to surface BAS in
MCM-41 catalysts.29, 33 Moreover, BAS can be generated by ion-exchange of parent
materials with multivalent metal cations (e.g. Al3+, Mg2+, Ca2+, La3+, Ce3+, etc.) followed
16

Page 17 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
by calcination,34-36 e.g. the formation of bridging OH groups (Si-OH-Al) by ion-exchanged
of non-acidic Na-form zeolites. Therefore, the formation of BAS in Sn-doped and Sn-
1
exchanged MCM-41 catalysts were investigated by H MAS NMR spectroscopy, a
powerful tool to study surface hydroxyl groups.
1
As shown in Figure 2, the H MAS NMR spectra of all dehydrated samples are
dominated by a strong peak at δ = 1.8 ppm with a weak shoulder at δ1H = 2.8 ppm, which
are assigned to terminal silanols and hydrogen-bonded OH groups, respectively (Figure
2, upper traces).32 Unlike zeolites, no low-field signal due to bridging OH groups at 3.5-
5.2 ppm could be observed in the spectra of these MCM-41 catalysts.3, 32, 37 Loading of
strongly basic molecules, such as ammonia, is a suitable method for a selective
identification of acidic protons.38-39 After ammonia loading, a signal at ca. 6.7 ppm was
observed for all ammonia-loaded Sn-containing MCM-41 catalysts (Figure 2, bottom
traces). The signal at δ1H = 6.7 ppm is assigned to ammonium ions formed by ammonia
protonated on surface BAS of Sn-containing MCM-41 catalysts,29, 40 which can be
generated due to the electronegativity difference between Sn and Si atoms.
17

Green Chemistry
Page 18 of 41
View Article Online
DOI: 10.1039/C9GC00820A
1
The densities of the surface protons were quantified by their corresponding H MAS
NMR signal intensities (see Table 1). H-form MCM-41 materials exhibited much higher
BAS density compared to Na-form MCM-41 materials because of the substitution of
hydroxyl protons by Na+. More significantly, H-[Sn]MCM-41 (4.3 × 10-2 mmol/g) and Na-
[Sn]MCM-41 (0.72 × 10-2 mmol/g) possessed over 6 times higher BAS densities
compared to Sn/H-[Si]MCM-41 (0.25 × 10-2 mmol/g) and Sn/Na-[Si]MCM-41 (0.12 × 10-2
mmol/g). This finding indicates that the incorporation of Sn into the MCM-41 framework
can promote the formation of BAS due to the change with Sn4+ cations under current
conditions. Furthermore, a significant amount of extra-framework Sn species (EFSn) may
be generated in the Sn-doped catalysts, which act as LAS. Ammonia loaded on LAS can
result in a hump at δ1H = 2.8-4.6 ppm.41-44 Therefore, the significant broadening of the 1H MAS
NMR spectra may indicate the formation of LAS on Sn-containing MCM-41 materials.
18

Page 19 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
19

Green Chemistry
Page 20 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Figure 2. 1H MAS NMR spectra of dehydrated a) H-[Sn]MCM-41, b) Na-[Sn]MCM-41, c)
Sn/H-[Si]MCM-41, and d) Sn/Na-[Si]MCM-41, recorded before (top) and after (bottom)
loading with NH3 and subsequent evaculation of physisorbed NH3 at 373 K for 1 h.
13C CP/MAS NMR spectroscopy of solid acids after adsorption of acetone-2-13C
13
(CH3 COCH3) as probe molecule is often used to identify the presence of BAS and
LAS.32, 45-46 Moreover, the acidic strength of surface sites can be scaled by the 13C
13
chemical shift of corresponding signals. For instance, the C MAS NMR spectrum of
acetone-2-13C-loaded [Si]MCM-41 is dominated by a signal at 210 ppm due to
13
CH3 COCH3 adsorbed on non-acidic SiOH groups in pure silica.39 The formation of BAS
13
by Al incorporation into the silica framework results in a low-field shift of the C MAS
NMR signal to δ13C = 213-225 ppm,32 where a larger δ13C values indicate a higher acid
13
strength of BAS. Upon adsorption of CH3 COCH3 at extra-framework Al species, i.e. at
Lewis sites generated e.g. by extra-framework Al species in dealuminated zeolites,
corresponds to δ13C = 233-245 ppm.32 Similarly, strong signals at ca. δ13C = 213 ppm were
observed in the 13C MAS NMR spectra of the all Sn-containing MCM-41 catalysts under
20

Page 21 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
study (Figure 3), indicating the weak acid strength of BAS generated by the substitution
of Si by Sn atoms. The sginals at δ13C = 230-235 ppm are attributed to the existence of
surface LAS due to EFSn. Additional signals at δ13C at ca. 74 ppm and at ca. 160 and 29
ppm are caused by an aldol reaction of the acetone-2-13C on the LAS and 13C atoms in
the non-enriched methyl groups, respectively.42, 46 However, Sn-doped MCM-41 materials
(Figure 3 c and d) showed a much weaker signal of LAS, but stronger signal for aldol
reaction products. This finding is attributed to the rapid condensation of acetone-2-13C on
strong LAS, such as earlier observed on ZrO2.42 Therefore, further characterization of
LAS on Sn-containing MCM-41 catalysts were performed.
213
a)
74
204
160
29
230
213
b)
100
212
212
204
c)
d)
174
280
240
200
160
120
80
40
0
(ppm)
21

Green Chemistry
Page 22 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Figure 3. 13C MAS NMR spectra of a) H-[Sn]MCM-41, b) Na-[Sn]MCM-41, c) Sn/H-
[Si]MCM-41, and d) Sn/Na-[Si]MCM-41.
31P MAS NMR spectroscopy using TMPO as probe molecules is a suitable method for
this purpose.32, 41, 47 As shown in Figure 4, two strong signals at ca. 57 ppm and 46 ppm
dominate the 31P MAS NMR spectra of all Sn-containing MCM-41 materials. The signal
at δ31P = 46 ppm is near the range of TMPO at LAS, such as reported for -Al2O3 and
TiO2/anatase recently.41 Chemical shifts of 60-70 ppm are typical for TMPO protonated
by BAS in zeolites, i.e. large δ31P values represent high acid strength.32, 41, 48 In Figure 4,
the signal at δ31P = 57 ppm was attributed to TMPO at weak BAS, which were generated
on Sn-doped MCM-41 catalysts, similarly to the observation by 13C MAS NMR
spectroscopy. The intensity of the signals at δ31P = 57 ppm strongly decreased in the
order of H-[Sn]MCM-41 > Na-[Sn]MCM-41 > Sn/H-[Si]MCM-41 > Sn/Na-[Si]MCM-41, in
1
line with the amount of BAS quantified by H MAS NMR spectroscopy (Table 1). The
simulation results based on the BAS and LAS signals shown at the bottom of each
spectrum in Figure 4. The increase of the relative signal intensity of TMPO at LAS
22

Page 23 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
correlates well with increasing the ratios of EFSn/framework Sn in Sn-containing MCM-
41s. The densities of LAS were determined by the corresponding BAS densities and
BAS/LAS ratios obtained via adsorption of ammonia and TMPO and subsequent 1H and
31P MAS NMR studies (Figures 2 and 4, as well as Table 1), respectively.
57
46
a) H-[Sn]MCM-41
experiment
simulation
components
BAS : LAS = 71 : 29
b) Na-[Sn]MCM-41
experiment
simulation
components
BAS : LAS = 55 : 46
c) Sn/H-[Si]MCM-41
experiment
simulation
components
BAS : LAS = 29 : 71
d) Sn/Na-[Si]MCM-41
experiment
simulation
components
BAS : LAS = 18 : 82
200
150
100
-100
50
0
-50

31P / ppm
23

Green Chemistry
Page 24 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Figure 4. 31P MAS NMR spectra of TMPO adsorbed on (a) H-SnMCM-41, (b) Na-SnMCM-
41, (c) Sn/H-[Si]MCM-41, and (d) Sn/Na-[Si]MCM-41.
Catalytic conversion of DHA to ethyl lactate over Sn-containing MCM-41 catalysts.
Recently, Sn-MCM-41s were demonstrated to be promising catalysts in the conversion
of DHA to alkyl lactates in alcohols, which requires the presence of both BAS and LAS.
Here, we prepared a series of Sn-MCM-41 materials with various BAS densities and
BAS/LAS ratios, which enabled the investigation of the cooperative effect of BAS and
LAS on the catalytic conversion of DHA. We have studied the conversion of DHA to ethyl
lactate over Sn-MCM-41 in ethanol as a solvent. Scheme 1 shows the proposed reaction
pathway of the conversion of DHA to the products. Initially, BAS catalyse the dehydration
of DHA, resulting in the formation of pyrualdehyde (PA) via the rearrangement of enol.
Upon alcohol addition, PA can be converted to the desired alkyl lactates at LAS or to
pyruvaldehyde hemiacetal (PAHA) at strong BAS. The latter can result in pyruvaldehyde
diacetal, a major by-product. On the other hand, is not well known, whether PAHA can be
converted to alkyl lactate.
24

Page 25 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
O
LAS
+
O
OH
O
O
LAS
rearrangement
BAS
O
ROH
O
OH
HO
+ ROH
- H2O
H
Pyruvaldehyde
(PA)
1,3-dihydroxyacetone
(DHA)
BAS
+ ROH
O
O
OH
OR
O
BAS
OH
? BAS
+ ROH
OR
OR
OR
Pyruvaldehyde
diacetal (PADA)
Pyruvaldehyde
Hemiacetal (PAHA)
Alkyl lactate
Scheme 1. Proposed mechanism for the conversion of dihydroxyacetone (DHA) to alkyl
lactate.
In the present work, the catalytic performance of all the obtained MCM-41 materials
were firstly evaluated under the same conditions like reported in literature.13, 23 H-ZSM-5,
as a reference material having exclusively BAS,3 led to a DHA conversion of 41 % after
a reaction time of 6 h (Supporting Information Figure S2a). DHA was completely
converted over Sn-doped MCM-41 catalysts within 30 min reaction time, while DHA
conversions of 96% and 70 % were obtained with Sn/H-[Si]MCM-41 and Sn/Na-[Si]MCM-
41, respectively. With much lower densities of acid sites, the significant higher activity of
mesoporous Sn-containing MCM-41s than microporous zeolite H-ZSM-5 is explained by
their larger pore size, which can improve the diffusion of reactants and products inside
25

Green Chemistry
Page 26 of 41
View Article Online
DOI: 10.1039/C9GC00820A
the pores and channels.49 Mainly the desired product, ethyl lactate (EL), could be
detected for the Sn-containing MCM-41 catalysts during the whole reaction time, while a
small amount of PA was only observed for the Sn-doped MCM-41 catalysts. As reported
in literature,13 nearly no EL, but PAHA and PADA, could be detected for the Brønsted
acidic H-ZSM-5 zeolite in the first hours. After a reaction time of 6 h, an EL yield of 4.5%
was achieved and increased to over 50% with a DHA conversion of 100 % after 40 h
reaction time. This indicates that strong BAS are able to convert DHA into EL beyond
PADA, however, less effective compared to LAS. Therefore, the tuning and control of
surface BAS and LAS on the catalysts is of great importance for the selective conversion
of DHA to alkyl lactate.
To gain more insight into the acidity-activity behavior of Sn-containing MCM-41
o
catalysts, lower reaction temperatures, e.g. 65 and 45 C, were applied (Figure 5a and
S3). The [Sn]MCM-41 materials showed a much higher conversion of DHA than the Sn-
[Si]MCM-41 materials under the same conditions, which was similar to their properties
observed at 90 oC. With similar acid strength (Figure 4) and density of LAS (Table 1), Na-
26

Page 27 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
[Sn]MCM-41 gave a DHA conversion of 100 %, but only 73% and 51% were observed for
Sn/H-[Si]MCM-41 and Sn/Na-[Si]MCM-41, respectively, after 4 h reaction time. H-
[Sn]MCM-41, providing the highest BAS density (4.3×10-2 mmol/g) among the MCM-41
materials, could completely convert DHA even within a reaction time of 2 h. The BAS of
this catalyst are very active sites in the DHA dehydration to PA,13 and thus, the higher
BAS density of the [Sn]-MCM-41 catalysts compared with the Sn-[Si]MCM-41 catalysts
leads to a higher yield of desired EL.
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
a)
b)
Sn/H-[Sn] Sn/Na-[Sn]
MCM-41
MCM-41
H-[Sn]
MCM-41
H-[Sn]
MCM-41
0
50
100 150 200 250 300 350
Time (min)
Figure 5. (a) DHA conversion as a function of the reaction time over various Sn-containing
MCM-41s (H-[Sn]MCM-41 (black), Na-[Sn]MCM-41 (red), Sn/H-[Si]MCM-41 (blue), and
Sn/Na-[Si]MCM-41 (pink)), and (b) yield (black) and selectivity (grey) to ethyl lactate.
27

Green Chemistry
Page 28 of 41
View Article Online
DOI: 10.1039/C9GC00820A
Reaction conditions: 0.05g of the MCM-41 catalysts, 5 mL ethanol solution containing 0.4
M DHA at 65 ℃ after a reaction time of 6 h under stirring.
PA, as an intermediate, can be converted to alkyl lactate via the intramolecular
Cannizzaro reaction.50 The redox disproportionative conversion of PA involves a ketone
reduction and an aldehyde oxidation. It can be promoted in the presence of Lewis acidic
catalysts (Scheme 1), much faster than on Brønsted acid catalyst, such as zeolite H-ZSM-
5. Under the same conditions of the DHA conversion, Sn/Na-[Si]MCM-41 with the lowest
BAS/LAS ratio results in the highest selectivity to EL (94.5%), which decreases with
reducing BAS/LAS ratio (89.4-80.8%) (Figure S4). An increasing BAS/LAS ratio leads to
a higher amount of PADA (4.8-15.8%) as the main by-product (Table S2).
For demonstrating the effect of BAS and LAS, the selectivities to each product
generated during the DHA conversion as a function of the reaction time is shown in Figure
6. The highest initial selectivity to EL (> 98%) was obtained with Sn/Na-[Si]MCM-41
(Figure 6a), having the lowest BAS/LAS ratio. The lowest BAS density is accompanied
by a decrease of the conversion DHA to PA, while a relative large density of LAS causes
28

Page 29 of 41
Green Chemistry
View Article Online
DOI: 10.1039/C9GC00820A
a fast conversion of PA to EL. An increasing amount of PA and a small amount of PADA
were observed for longer reaction times, indicating the faster conversion of DHA to PA
on BAS than the PA is converted on LAS. With similar densities of LAS, the lower initial
selectivity to EL decreased, but a higher selectivity to PA and PADA was obtained after
6 h reaction time over Na-[Sn]MCM-41 (Figure 6b). The higher amount of BAS enables
Na-[Sn]MCM-41 to completely convert DHA after 4 h reaction time. In this case, more
BAS contributed to the reverse reaction to generate PA from PADA. The formed PA can
compensate the PA converted at LAS, and thus shifts the reaction balance to promote
the formation of EL. The latter is indicated by the initial increase and subsequent decrease
of the amount of PA and PADA. This was similarly observed with H-[Sn]MCM-41 having
both the highest BAS and LAS density. On the other hand, the larger BAS/LAS ratio of
this material leads to an even lower initial EL selectivity.
29