Conversion of Dihydroxyacetone to Methyl Pyruvate Catalyzed by Hybrid Molecular Sieves at Low Temperature: A Strategy for the Green Utilization of Glycerol

Article abstract of DOI:10.1007/s10562-019-03064-3

Abstract: Methyl pyruvate (MPA) was synthesized from dihydroxyacetone (DHA), the oxidation product of glycerol, over Sn-β and TS-1 hybrid molecular sieves at low temperature. Sn-β and TS-1 provide the active sites for conversion of DHA to methyl lactate (MLA) and oxidation of MLA to MPA, respectively. Synergism of Lewis acid sites on Sn-β and TS-1 realize the production of MPA from DHA. After optimization, 71% yield of MPA can be obtained under 50?°C. This is the first report for the synthesis of MPA from DHA directly. Graphic Abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-019-03064-3

Catalysis Letters
https://doi.org/10.1007/s10562-019-03064-3
Conversion of Dihydroxyacetone to Methyl Pyruvate Catalyzed
by Hybrid Molecular Sieves at Low Temperature: A Strategy
for the Green Utilization of Glycerol
Qianqian Luo1 · Tianliang Lu1,2 · Jun Xu1 · Xiaomei Yang2 · Lipeng Zhou2
Received: 27 September 2019 / Accepted: 2 December 2019
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
Methyl pyruvate (MPA) was synthesized from dihydroxyacetone (DHA), the oxidation product of glycerol, over Sn-β and
TS-1 hybrid molecular sieves at low temperature. Sn-β and TS-1 provide the active sites for conversion of DHA to methyl
lactate (MLA) and oxidation of MLA to MPA, respectively. Synergism of Lewis acid sites on Sn-β and TS-1 realize the
production of MPA from DHA. After optimization, 71% yield of MPA can be obtained under 50 °C. This is the frst report
for the synthesis of MPA from DHA directly.
Graphic Abstract
Keywords Catalysis · Biomass conversion · Zeolites · Methyl pyruvate · Dihydroxyacetone
1 Introduction
KHSO4 is used as dehydrating agent in this route resulting
in low atom-efciency and severe environmental pollution.
Synthesis of pyruvates can also be realized through carbo-
hydrates fermentation [5], but conditions including tem-
perature, pH value, and so on should be regulated precisely
to keep the microbe active. Furthermore, the space time
yield of product is low while the separation and purifca-
tion cost of the product is high. These drawbacks limit the
commercialization of this process. Lactates (lactic acid and
its esters), the biomass derivates, have similar molecular to
pyruvates. For example, the only diference of molecular
structure between methyl lactate (MLA) and methyl pyru-
vate (MPA) is that MLA have 2 more H atoms than MPA.
So, preparation of pyruvates from the corresponding lac-
tates by catalytic oxidation attracts much attention due to
the excellent atom-efciency, and various catalysts have
been developed [6–13]. Our group had realized the highly
As important chemicals, pyruvates including pyruvic acid
and esters are widely adopted in various felds such as plas-
tic, food, cosmetic, pesticidal, and pharmaceutical industries
[1–3]. Meanwhile, applications of pyruvates as raw materials
in synthesis of many bioactive substances are attractive [4].
Pyruvates can be produced from tartaric acid by dehydra-
tive decarboxylation. This is the conventional and commer-
cial method for the synthesis of pyruvates. However, excess
* Tianliang Lu
lutianliang@zzu.edu.cn
1
School of Chemical Engineering, Zhengzhou
University, 100 Kexue Road, Zhengzhou 450001,
People’s Republic of China
2
College of Chemistry, Zhengzhou University, 100 Kexue
Road, Zhengzhou 450001, People’s Republic of China
Vol.:(0123456789)
1 3

Q. Luo et al.
efcient synthesis of ethyl pyruvate from ethyl lactate over
TS-1 using H2O2 as oxidant without solvent, and 97.8% ethyl
pyruvate yield can be obtained at 50 °C [13].
2.2 Synthesis of TS‑1
Molar composition of the gel is n(SiO2):n(TiO2):n(TPAB
r):n(ethanolamine):n(H2O) = 1.0:0.033:0.1:0.5:30. After
addition of colloidal Si into the beaker containing ethanola-
mine, TPABr and H2O with stirring, TBOT-isopropanol was
dropped into the mixture. Then, isopropanol was evaporated.
And crystallization of the gel was carried out by hydrother-
mal treatment at 175 °C for 72 h in a stainless autoclave.
After fltration, washing (with H2O), drying (120 °C, 12 h),
and calcination (550 °C, 6 h), white solid of TS-1 was got-
ten. To make the synthesized TS-1 more stable, further treat-
ment was carried out. The TS-1 was added to the autoclave
containing ethanolamine and TPABr. After treatment at
175 °C for 48 h, the TS-1 was separated, and washed with
H2O. Furthermore, the dilute sulphuric acid and hydrogen
peroxide solution were used to wash the TS-1. After that, the
solid of TS-1 was dried (120 °C, 12 h) and calcined (550 °C,
6 h) for using.
Glycerol is the by-product of biodiesel industry. Devel-
opment of the strategy for the utilization of glycerol attracts
much attention from researchers [14–16]. Dihydroxyacetone
(DHA) can be obtained from glycerol [17–21]. Xu’s group
had reported that DHA can be produced from glycerol under
mild conditions with high yields. For example, 80% yields
of DHA can be ofered by Au/CuO at 40–50 °C [21]. Mean-
while, DHA is the precursor for production of lactates which
can be further oxidized to pyruvates as mentioned above
[22–32]. Thus, realization of direct conversion of DHA to
pyruvates under mild conditions is a strategy for the green
utilization of glycerol.
Sn containing solid materials [24–32], especially the Sn
doped zeolites [27–32], were usually used as catalysts in
conversion of DHA to lactates. Our group had prepared hier-
archical Sn-USY for this reaction. 100% DHA conversion
and higher than 95% MLA yield were obtained at 25–40 °C
[32]. Sn-β also shows excellent catalytic performance for the
conversion of DHA to lactate [28–31]. Tarnning et al. had
investigated the performance of Tin-containing silicates on
conversion of DHA to MLA. In the investigated zeolites of
Sn-β, Sn-MFI, Sn-MCM-41 and Sn-SBA-15, Sn-β displayed
the best performance. Higher than 90% MLA yield achieved
at 40 °C [31]. Wang had also applied Sn-β to catalyze the
conversion of DHA to MLA, and 92% MLA yield can be
obtained at 60 °C [29].
2.3 Synthesis of Sn‑β and Sn‑USY
The Tin containing zeolites were synthesized by the solid-
state ion-exchange method. Take the synthesis of Sn-β as
an example. Firstly, Al-β (Si/Al = 13.8) was added to the
HNO3 solution (13 mol L−1) with a solid-to-liquid ratio of
1 g 20 mL−1. After refuxing at 100 °C for 20 h, the solid was
separated by centrifugation. Then, the solid was washed by
deionized water to neutral. After drying at 100 °C overnight,
the obtained solid was transferred to a mortar containing
SnCl4·5H2O. The amount of SnCl4·5H2O was calculated
according to the theoretical Sn content of 2 wt%. After
grounding for 1 h, the solid was dried at 100 °C overnight.
Finally, after calcination at 550 °C for 3 h, the Sn-β was
obtained. Sn-USY was also synthesized by this method, but
the concentration of HNO3 solution is 8 mol L−1.
In this work, one-pot conversion of DHA to pyruvate was
realized under mild conditions. Based on previous work,
Sn containing zeolites was applied as active component for
conversion of DHA to lactate, and TS-1 was used to catalyze
the conversion of lactate to pyruvate in presence of H2O2.
After optimization, 100% DHA conversion and 71% MPA
yield were obtained at 50 °C over Sn-β and TS-1 hybrid
catalyst. This is the frst report for the synthesis of MPA
from DHA directly.
2.4 Catalyst Characterization
2 Experimental
Crystalline structures of all the samples were determined
by powder X-ray difraction (XRD) on a Panalytical X’pert
PRO instrument. Radiation of Cu Kα (λ = 0.15418 nm),
40 kV of tube voltage, and 40 mA of tube current were
applied in this measurement. Surface area and pore vol-
ume were characterized by physical adsorption method on
Quantachrome Autosorb. Adsorption/desorption isotherms
of N2 were measured at −196 °C. Pretreatment of the sam-
ple was carried out at 300 °C for 3 h to desorb the H2O and
impurity on the surface. The surface area was calculated
by using Brunauer–Emmett–Teller (BET) method. Ti con-
tent in TS-1 was measured by a X-ray fuorescence (XRF)
2.1 Materials
Zeolites of β and USY were obtained from Nankai Uni-
versity Catalyst Co. (People’s Republic of China). DHA,
MLA, MPA, and 30% H2O2 were purchased from Sinopharm
Chemical Reagent Co., Ltd. (People’s Republic of China).
Tetrabutyl titanate (TBOT), tetrapropylammonium bromide
(TPABr), and SnCl4·5H2O were purchased from Aladdin
Chemical Reagent Corporation. Ethanolamine, isopropyl
alcohol, naphthalene, colloidal silica, HNO3, and H2SO4
were used as received.
1 3

Conversion of Dihydroxyacetone to Methyl Pyruvate Catalyzed by Hybrid Molecular Sieves at…
spectrometer, and Sn contents in Sn-β and Sn-USY were
determined by induced coupled plasma-atomic emission
spectroscopy (ICP-AES). UV–Vis spectra of the samples
were collected on a Shimadzu UV-2600 spectrophotometer
with the wavelength range of 800–200 nm. Bruker Ten-
sor II spectrometer was used to collect the FTIR spectra of
the samples with adsorption of pyridine (Pyridine-FTIR).
Pretreatment of the sample (9–15 mg) was carried out at
450 °C for 3 h after the analysis pool was evacuated. Then,
the spectrum as the background reference was collected at
room temperature. The resolution was 4 cm−1. After that,
pyridine was introduced to the system to adsorb saturately
on the catalyst (about 15 min). Elevate the temperature to
Sn-USY
Sn-β
TS-1
desired temperature, and evacuate the system to desorb the
10
20
30
40
50
2Theta (o)
pyridine. After evacuation for 30 min, the FTIR spectrum
was recorded.
Fig. 1 XRD patterns of TS-1, Sn-β, and Sn-USY
2.5 Catalytic Tests
A round bottom fask with refux condenser was used as the
reactor in this work. Firstly, 0.2 g of DHA along with desired
amount of Sn-β and TS-1 were added to the reactor contain-
ing 5 mL of methanol under stirring. Then, the reactor was
heated to the desired temperature. After reaction for a default
time, 30% H2O2 was added to the system. Finally, stop the
reaction after some hours. Analysis of the reaction mixture
was carried out by an Agilent 7890A GC equipped an HP-5
column. Naphthalene was used as the internal standard. The
identifcation of the products was realized by Agilent 6890 N
GC/5973 MS and comparison with the authentic samples.
Table 1 Textural properties of the samples
Sample
SBET (m2 g−1
)
Total pore vol- SExternal
Mesopore
volume
ume (mL g−1
)
(m2 g−1
)
(mL g−1
)
TS-1
Sn-β
Sn-USY 444
380
435
0.27
0.31
0.35
45
98
98
0.10
0.13
0.17
SExternal SBET −SMicropore
,
mesopore volume total pore vol-
ume − micropore volume, where the micropore surface area and vol-
ume were determined by the t-plot method at a relative pressure of
0.05–0.70
3 Results and Discussion
of tetrahedrally coordinated SnIV [33]. For Sn-β, no obvious
band can be observed at around 300 nm, which ascribed to
the absorption of SnO2 as shown in Fig. 2a, implying no
aggregated Sn species appeared in synthesis of Sn-β. No
diference between spectra of Sn-USY and H-USY can be
observed at around 300 nm implying no SnO2 existed in
Sn-USY. For TS-1, an intense absorption band at 202 nm,
which attributed to the presence of framework Ti4+ with
tetrahedral coordination, is observed [34]. Furthermore, a
little absorption band appears at around 313 nm, which is
the main absorption band in spectrum of TiO2 as shown in
Fig. 2b, indicating the presence of some aggregated TiO2.
Presence of some TiO2 is caused by that the Ti content actu-
ally applied in the synthesis of TS-1 is slightly higher than
the theoretical maximum Ti content can be introduced to
the framework (2.5%, molar fraction). Excess Ti was used
in the synthesis of TS-1 to ensure the maximum framework
Ti content. FT-IR spectroscopy of absorbed base molecules
was used to study the acidic properties of samples. As shown
in Fig. 3, IR bands at ~1447 cm−1 observed in the spectra
of the TS-1 can be ascribed to pyridine adsorption on Lewis
3.1 Catalyst Characterization
Sn-β and Sn-USY were applied for the conversion of DHA
to MLA, and TS-1 was used for the oxidation of MLA to
MPA. As shown in Fig. 1, XRD patterns confrm the typical
BEA structure of Sn-β, FAU of Sn-USY, and MFI of TS-1.
As listed in Table 1, all the samples have high total surface
area. Both Sn-β and Sn-USY have higher external surface
area and mesopore pore volume than TS-1 due to the dea-
lumination procedure for introduction of Sn. Ti content in
TS-1 was tested by XRF method, the Si/Ti molar ratio is
31 which is close to the initial Si/Ti molar ratio in the gel
for synthesis of TS-1. Sn contents in Sn-β and Sn-USY are
1.91 wt% and 1.88 wt%, respectively, which are close to the
theoretical values (2 wt%). Presences of Sn in framework
of Sn-β and Sn-USY, and Ti in framework of TS-1 were
proved by UV–Vis spectra (Fig. 2). A sharp and intense
absorption band at ca. 208 nm is observed in spectra of both
Sn-β and Sn-USY. This is usually ascribed to the presence
1 3

Q. Luo et al.
(L) acid sites [35]. The appearance of band at 1491 cm−1 for
TS-1 indicates pyridine adsorption on L acid sites [35]. For
Sn-β and Sn-USY, the IR band at 1491 cm−1 was ascribed
to pyridine adsorption on both L and Brønsted (B) acid sites
[30, 33]. Both Sn-β and Sn-USY provided the strong absorp-
tion band at 1452 cm−1, assigned to the pyridine bound with
the L acid Sn sites [30, 33]. The similarity of FT-IR spectra
between Sn-β and Sn-USY indicates the similar acid prop-
erties between Sn-β and Sn-USY. A very weak absorption
feature at 1547 cm−1 related to the B acid was also observed
in the spectra of Sn-β and Sn-USY [30, 33]. Location of the
band at 1447 cm−1 for TS-1 is diferent from the locations of
the bands at 1452 cm−1 for Sn-β and Sn-USY indicating the
diferent L acid properties of TS-1 from Sn-β and Sn-USY.
a
SnO2
208
Sn-USY
H-USY
Sn-β
200
300
400
500
600
Wavelength (nm)
3.2 Catalytic Conversion of DHA to MLA Over Sn‑β
b
As reported in previous literature, both Sn-β and Sn-USY
have good catalytic performance on conversion of DHA to
MLA in methanol. In this work, Sn-β and Sn-USY were
tested as catalyst frstly to compare their catalytic perfor-
mance in the presence of TS-1 which would be used as
oxidative catalyst in the following conversion of MLA to
MPA (Table 2). Interestingly, Sn-USY and Sn-β showed the
similar catalytic performance. After reaction at 50 °C for
5 h, 85% MLA yield with 100% DHA conversion and 86%
MLA yield with 100% DHA conversion were obtained over
Sn-USY and Sn-β, respectively. The similarity of the cata-
lytic performance between Sn-USY and Sn-β is caused by
their similar acid properties (Fig. 3). In this work, Sn-β was
selected as the catalyst for conversion of DHA to MLA in
the following study. The reaction includes two successive
steps: (1) conversion of DHA to MLA over Sn-β (desired
amount of TS-1 was also added at beginning of this step), (2)
oxidation of MLA to MPA over TS-1 started by addition of
H2O2. To simplify the operation, Sn-β and TS-1 were added
to the reactor simultaneously at the frst step. As shown in
Table 2, the addition of TS-1 in the frst step does not afect
the MLA yield under the same conditions.
TiO2
202
TS-1
313
200
300
400
500
600
Wavelength (nm)
Fig. 2 UV–Vis spectra of Sn-β, Sn-USY, and SnO2 (a), and TS-1, and
TiO2 (b)
1452
1491
1547
Sn-USY
Table 2 Catalytic conversion of DHA to MLA over Sn-USY and
Sn-β
1447
Sn-β in the presence of TS-1
Sample
Conversion of DHA (%)
Yield of
MLA
(%)
TS-1
Sn-USY
Sn-β
100
100
100
85
86
85
1550
1500
Wavenumber (cm-1)
1450
1400
Sn-βa
Reaction conditions: 5 mL of methanol, 0.2 g of DHA, 0.1 g of TS-1,
0.2 g of Sn containing zeolite (Sn-β or Sn-USY), 50 °C, 5 h
Fig. 3 FT-IR spectra of Sn-β, Sn-USY and TS-1 after pyridine
aNo TS-1 was added
adsorption and evacuation at 150 °C for 30 min
1 3

Conversion of Dihydroxyacetone to Methyl Pyruvate Catalyzed by Hybrid Molecular Sieves at…
100
3.3 Conversion of DHA to MPA Over Hybrid Sn‑β
and TS‑1
80
Yield of MLA
Yield of MPA
To get more information about the catalytic performance for
Conversion of DHA
Sn-β and TS-1, the efects of reaction conditions including
60
the amount of methanol, amount of Sn-β and TS-1, reac-
tion temperature, amount of H2O2, and reaction time were
investigated. In this study, 100% DHA conversion were
40
obtained for all the catalytic tests, however, total yield of
MLA and MPA was lower than 100% implying the forma-
20
tion of byproducts which was not detected. Figure 4 shows
the efect of methanol volume on the conversion of DHA to
0
MPA at 50 °C. When the methanol amount increased from
0.1
0.2
0.3
0.4
1 to 2 mL, the yields of MLA and MPA increased. When
the methanol was small (<2 mL), methanol cannot totally
immerse the solid catalyst and the mass transfer in the slurry
of the reaction mixture is poor. With the further increase of
the methanol amount to 3.5 mL, MLA yield increased but
the MPA yield decreased. This implies that more methanol
facilitated the conversion of DHA to MLA, but conversion
of MLA to MPA was suppressed. Methanol amount afects
the concentration of H2O2 which was used as oxidant in the
oxidation of MLA to MPA. With the increase of methanol
amount, concentration of H2O2 would decrease which might
lead to the suppression the formation of MPA. As shown in
Fig. 4, the yield of MPA reached a maximum of 54% when
the methanol volume was 2 mL which was selected in the
following study.
Amount of Sn-β (g)
Fig. 5 Efect of Sn-β amount on the conversion of DHA to MPA.
Reaction conditions: 2 mL of methanol, 0.2 g of DHA, 0.2 g of TS-1,
2 mL of H2O2, 50 °C, reaction time before addition of H2O2 was 12 h,
reaction time after addition of H2O2 was 5 h
of MPA was caused by further oxidation of MLA. When the
Sn-β amount was increased from 0.1 to 0.4 g, the total yields
of MLA and MPA were comparable. 0.3 g of Sn-β was used
in the following study, because the yield of MPA reached
a maximum of 59% when the Sn-β amount was 0.3 g. Fig-
ure 6 shows the efect of TS-1 amount on the conversion of
DHA to MPA. Without TS-1, only MLA (78% yield) was
obtained and no MPA formed although H2O2 was added.
This means that Sn-β and TS-1 played distinctively diferent
roles in the conversion of DHA to MPA. Sn-β provides only
the active sites for the conversion of DHA to MLA and TS-1
The efect of Sn-β amount on the conversion of DHA to
MPA is shown in Fig. 5. Sn-β provides the active sites for the
conversion of DHA to MLA. Total yield of MLA and MPA
refects the catalytic performance of Sn-β, because formation
100
100
Yield of MLA
Yield of MPA
Yield of MLA
80
80
60
40
20
0
Conversion of DHA
Yield of MPA
Conversion of DHA
60
40
20
0
1
1.5
2
2.5
3.5
0
0.1
0.15
0.25
0.3
0.35
0.4
0.5
Amount of methanol (mL)
Amount of TS-1 (g)
Fig. 4 Efect of methanol amount on the conversion of DHA to MPA.
Reaction conditions: 0.2 g of DHA, 0.4 g of Sn-β, 0.2 g of TS-1,
2 mL of H2O2, 50 °C, reaction time before addition of H2O2 was 12 h,
reaction time after addition of H2O2 was 5 h
Fig. 6 Efect of TS-1 amount on the conversion of DHA to MPA.
Reaction conditions: 2 mL of methanol, 0.2 g of DHA, 0.3 g of Sn-β,
2 mL of H2O2, 50 °C, reaction time before addition of H2O2 was 12 h,
reaction time after addition of H2O2 was 5 h
1 3

Q. Luo et al.
is indispensable for the oxidation of MLA to MPA. Both
TS-1 and Sn-β are L acid catalysts, however, the diferent
electronic structure of Ti4+ and Sn4+ leads to the diferent
active sites in TS-1 and Sn-β (Fig. 3). With the increase of
TS-1 amount from 0 to 0.35 g, yield of MLA decreased from
78 to 6% and yield of MPA increased from 0 to 71%. With
the increase of TS-1 amount, amount of L acid sites for the
conversion of MLA to MPA increased, and this is favorable
for the conversion of MLA to MPA. Further increased the
TS-1 amount to 0.5 g, all the MLA converted to MPA. How-
ever, yield of MPA decreased to 64%. When TS-1 amount is
too large, contact between Sn-β and DHA would be hindered
because of the adsorption of DHA on TS-1 leading to the
decrease of the catalytic activity of Sn-β. This futher caused
the decrease of the MPA yield. Thus, 0.35 g of TS-1 was
selected in the following study. Based on the reaction scheme
for the conversion of DHA to MLA over Sn-β proposed in
the literature and the scheme for the oxidation of MLA to
MPA over TS-1 illustrated in our previous report [13, 15, 25,
26], the reaction scheme for the conversion of DHA to MPA
in methanol over Sn-β and TS-1 was proposed (Scheme 1).
Firstly, pyruvic aldehyde (PA) forms by dehydration of DHA
and/or glyceraldehyde (GLA) over B and/or L acid sites of
Sn-Beta. Then, isomerization and alcohol-addition of PA
lead to the formation of MLA over L acid sites of Sn-Beta.
When the H2O2 is added to the reactor, Ti(OOH) species
form on the surface of TS-1. The MLA desorbs from the
Sn-Beta, and adsorbs on TS-1. The hydroxyl group of MLA
reacts with Ti(OOH) leading to dehydration. After that, des-
orption of MPA from TS-1 occurs, and TS-1 is recovered.
This reaction pathway explains the roles of Sn-β, TS-1 and
H2O2 in this reaction. As listed in Table 2, further conver-
sion of MLA cannot proceed without H2O2 leading to the
high yield of MLA. When the H2O2 was added, oxidation of
MLA to MPA would occur in presence of Ti(OOH). In this
case, yield of MLA decreased and yield of MPA increased.
As mentioned above, side reactions occurred in this catalytic
system. For conversion of DHA to MLA, the by-products
might be the condensation and/or polymerization products
of the DHA and the intermediates [15, 25, 26]. Detection
of these by-products by GC is difcult. This might be why
conversions of DHA are 100%, but yields of MLA are about
85% in Table 2. However, for most reactions after addition
of H2O2, total yield of MPA and MLA is lower than 80%.
This implies that there is side reaction in the oxidation of
MLA. We had proved in previous work that MPA is unstable
Scheme 1 Proposed reaction scheme for the conversion of DHA to MPA over Sn-β and TS-1
1 3

Conversion of Dihydroxyacetone to Methyl Pyruvate Catalyzed by Hybrid Molecular Sieves at…
100
[13]. In the presence of H2O2, decomposition of MPA might
occur through hydrolysis and decarboxylation (Scheme 1).
Yield of MLA
H2O2 is the oxidant for the oxidation of MLA to MPA.
Yield of MPA
80
Conversion of DHA
Efect of H2O2 amount on the conversion of DHA to MPA
is shown in Fig. 7. When the 0.5 mL of H2O2 was applied,
60
yields of MLA and MPA were 43% and 40%, respectively.
This indicates that the oxidation of MLA is insufcient.
Theoretically, 0.11 mL of H2O2 can realize the complete
40
conversion of 0.2 g of DHA to MPA. However, 2 mL of
methanol was used as solvent in the reaction, and the added
20
H2O2 aqueous would be diluted. Lower addition amount
of H2O2 leads to the lower concentration of H2O2 in the
0
reaction solution. Oxidation ability of H2O2 is related to
45
50
55
60
the concentration. That is why MLA yield decreased to 6%
and MPA yield increased to 71% when the H2O2 amount
increased from 0.5 to 2 mL. Further increased the amount of
H2O2 to 3 mL would lead to the decrease of the MPA yield.
Besides increase the H2O2 concentration in the reaction mix-
ture, the increase of the addition amount of H2O2 also leads
to the increase of the total volume of the reaction mixture,
and this would cause the decrease of the concentration of
reactants and intermediates. Increase of H2O2 concentration
and decrease of the reactants and intermediates concentra-
tion are two contrary factors for the conversion of DHA to
MPA. When the amount of H2O2 is in the range of 2–3 mL,
decrease of the reactants and intermediates concentration
is the major factor which caused the decrease of the MPA
yield.
Reaction temperature (oC)
Fig. 8 Efect of reaction temperature on the conversion of DHA to
MPA. Reaction conditions: 2 mL of methanol, 0.2 g of DHA, 0.3 g of
Sn-β, 0.35 g of TS-1, 2 mL of H2O2, reaction time before addition of
H2O2 was 12 h, reaction time after addition of H2O2 was 5 h
temperature increased from 45 to 50 °C, the MLA yield
decreased from 20% to 6%, and the MPA yield increased
from 59% to 71%. Further elevated the temperature, the
MPA yield decreased, which might be caused by the insta-
bility of MPA. We had proved that decomposition of MPA
would occur through hydrolysis and decarboxylation in
the presence of H2O2 [13]. As shown in Fig. 9, with the
increase of the reaction time before addition of H2O2 from
1 to 5 h, MLA yield kept comparable, and MPA yield
increased from 35% to 69%. Further increased the reaction
time to 12 h, MPA yield increased slightly to 71%. Thus,
Reaction temperature and reaction time were opti-
mized. Figure 8 shows efect of the reaction temperature
on the conversion of DHA to MPA. This reaction is very
sensitive to the reaction temperature. When the reaction
100
100
Yield of MLA
Yield of MPA
Conversion of DHA
Yield of MLA
Yield of MPA
80
80
Conversion of DHA
60
40
20
0
60
40
20
0
1
3
5
12
0.5
0.8
1
1.5
2
2.5
3
Reaction time before addition of H2O2 (h)
Amount of H
2O
2
(mL)
Fig. 7 Efect of H2O2 amount on the conversion of DHA to MPA.
Reaction conditions: 2 mL of methanol, 0.2 g of DHA, 0.3 g of Sn-β,
0.35 g of TS-1, 50 °C, reaction time before addition of H2O2 was
12 h, reaction time after addition of H2O2 was 5 h
Fig. 9 Efect of the reaction time before addition of H2O2 on the
conversion of DHA to MPA. Reaction condition: 2 mL of methanol,
0.2 g of DHA, 0.3 g of Sn-β, 0.35 g of TS-1, 2 mL of H2O2, 50 °C,
reaction time after addition of H2O2 was 5 h
1 3

Q. Luo et al.
100
80
60
40
20
0
Compliance with Ethical Standards
Yield of MLA
Conflict of interest The authors declare that they have no confict of
Yield of MPA
interest.
Conversion of DHA
References
1. Bozkurt F, Tornuk F, Toker OS, Karasu S, Arici M, Durak MZ
(2016) Efect of vaporized ethyl pyruvate as a novel preserva-
tion agent for control of postharvest quality and fungal damage
of strawberry and cherry fruits. LWT Food Sci Technol 65:1044–
1049. https://doi.org/10.1016/j.lwt.2015.09.043
2. Wang Y, Lewis JD, Román-Leshkov Y (2016) Synthesis of ita-
conic acid ester analogues via self-aldol condensation of ethyl
pyruvate catalyzed by hafnium BEA zeolites. ACS Catal 6:2739–
2744. https://doi.org/10.1021/acscatal.6b00561
3
5
7
12
Reaction time after additioin of H2O2 (h)
3. Dusselier M, Wouwe PV, Dewaele A, Makshina E, Sels BF (2013)
Lactic acid as a platform chemical in the biobased economy: the
role of chemocatalysis. Energy Environ Sci 6:1415–1442. https://
doi.org/10.1039/C3EE00069A
Fig. 10 Efect of the reaction time after addition of H2O2 on the con-
version of DHA to MPA. Reaction condition: 2 mL of methanol,
0.2 g of DHA, 0.3 g of Sn-β, 0.35 g of TS-1, 2 mL of H2O2, 50 °C,
reaction time before addition of H2O2 was 5 h
4. Hernández K, Gómez A, Joglar J, Bujons J, Parella T, Clapés P
(2017) 2-Keto-3-deoxy-L-rhamnonate aldolase (YfaU) as catalyst
in aldol additions of pyruvate to amino aldehyde derivatives. Adv
Synth Catal 359:2090–2100. https://doi.org/10.1002/adsc.20170
0360
before addition of H2O2, there must be enough reaction
time (≥5 h) for the conversion of DHA to MLA. Figure 10
shows the efect of reaction time after addition of H2O2.
For this step, when the reaction time is shorter, for exam-
ple 3 h, oxidation of MLA to MPA would be insufcient.
When the reaction time is longer, for example ≥7 h, MLA
can be oxidized to MPA completely, however, decrease
of MPA yield would occur because the MPA is unstable.
5. Zhang B, Zhu Y, Zhang J, Wang D, Sun L, Hong J (2017)
Engineered Kluyveromyces marxianus for pyruvate produc-
tion at elevated temperature with simultaneous consumption of
xylose and glucose. Bioresour Technol 224:553–562. https://doi.
org/10.1016/j.biortech.2016.11.110
6. Zhang W, Innocenti G, Oulego P, Gitis V, Wu H, Ensing B, Cavani
F, Rothenberg G, Shiju NR (2018) Highly selective oxidation of
ethyl lactate to ethyl pyruvate catalyzed by mesoporous vanadia–
titania. ACS Catal 8:2365–2374. https://doi.org/10.1021/acsca
tal.7b03843
7. Zhang L, Wang R, Song L, Zhao X, Fan Q, Li H, Yu Q, Li X, Zeng
J, Zhang C, Liu T, Wang Z (2019) Aerobic oxidative dehydroge-
nation of ethyl lactate over reduced MoVNbOx catalysts. Catal
Lett 149:840–850. https://doi.org/10.1007/s10562-018-2616-9
8. Liu K, Huang X, Pidko EA, Hensen EJM (2017) MoO3–TiO2
synergy in oxidative dehydrogenation of lactic acid to pyruvic
acid. Green Chem 19:3014–3022. https://doi.org/10.1039/C7GC0
0807D
4 Conclusions
MPA was synthesized from DHA over Sn-β and TS-1
hybrid molecular sieves at low temperature. Sn-β and TS-1
provide the active sites for conversion of DHA to MLA
and oxidation of MLA to MPA, respectively. Synergism of
Lewis acid sites on Sn-β and TS-1 realize the production
of MPA from DHA. After optimization, 71% yield of MPA
can be obtained under 50 °C. This is the frst report for the
synthesis of MPA from DHA directly, and this work will
contribute to the green utilization of glycerol.
9. Zhao X, Zhang C, Xu C, Li H, Huang H, Song L, Li X (2016)
Kinetics study for the oxidative dehydrogenation of ethyl lactate
to ethyl pyruvate over MoVNbOx based catalysts. Chem Eng J
296:217–224. https://doi.org/10.1016/j.cej.2016.03.088
10. Hayashi H, Sugiyama S, Katayama Y, Sakai K, Sugino M, Shige-
moto N (1993) Liquid-phase oxidation of ethyl lactate to pyruvate
over suspended oxide catalysts. J Mol Catal 83:207–217. https://
doi.org/10.1016/0304-5102(93)87020-9
11. Sugiyama S, Kikumoto T, Tanaka H, Nakagawa K, Sotowa KI,
Maehara K, Himeno Y, Ninomiya W (2009) Enhancement of
catalytic activity on Pd/C and Te-Pd/C during the oxidative dehy-
drogenation of sodium lactate to pyruvate in an aqueous phase
under pressurized oxygen. Catal Lett 131:129–134. https://doi.
org/10.1007/s10562-009-9920-3
Acknowledgements We are grateful to the National Natural Science
Foundation of China (Grant Nos. 21802125, 21875222, 21871236),
China Postdoctoral Science Foundation (Grant No. 2017M612418),
the fund of the State Key Laboratory of Catalysis in DICP (Grant
No. N-16-02), the Henan Science and Technology Project (Grant No.
172102210490) and Program of Processing and Efcient Utilization of
Biomass Resources of Henan Center for Outstanding Overseas Scien-
tists (Grant No. GZS2018004) for the fnancial supports.
12. Zhang C, Wang T, Ding Y (2017) Oxidative dehydrogenation of
lactic acid to pyruvic acid over Pb–Pt bimetallic supported on car-
bon materials. Appl Catal A 533:59–65. https://doi.org/10.1016/j.
apcata.2017.01.010
13. Lu T, Zou J, Zhan Y, Yang X, Wen Y, Wang X, Zhou L, Xu J
(2018) Highly efcient oxidation of ethyl lactate to ethyl pyruvate
1 3

Conversion of Dihydroxyacetone to Methyl Pyruvate Catalyzed by Hybrid Molecular Sieves at…
catalyzed by TS-1 under mild conditions. ACS Catal 8:1287–
1296. https://doi.org/10.1021/acscatal.7b03558
conversion of dihydroxyacetone to lactic acid. Catal Sci Technol
6:6551–6560. https://doi.org/10.1039/C6CY00553E
26. Pighin E, Díez VK, Di Cosimo JI (2017) Kinetic study of the ethyl
lactate synthesis from triose sugars on Sn/Al2O3 catalysts. Catal
Today 289:29–37. https://doi.org/10.1016/j.cattod.2016.10.002
27. Guo Q, Fan DF, Pidko DEA, Graaf WNP, Feng PDZ, Li PDC,
Hensen PDEM (2013) Highly active and recyclable Sn-MWW
zeolite catalyst sugar conversion to methyl lactate and lactic acid.
Chemsuschem 6:1352–1356. https://doi.org/10.1002/cssc.20130
0160
14. Skaf DW, Natrin NG, Brodwater KC, Bongo CR (2012) Com-
parison of phototcatalytic hydrogen production from glycerol
and crude glycerol obtained from biodiesel processing. Catal Lett
142:1175–1179. https://doi.org/10.1007/s10562-012-0886-1
15. Lu T, Fu X, Zhou L, Su Y, Yang X, Han L, Wang J, Song C (2017)
Promotion efect of Sn on Au/Sn-USY catalysts for one-pot con-
version of glycerol to methyl lactate. ACS Catal 7:7274–7284.
https://doi.org/10.1021/acscatal.7b02254
16. Zhou L, Xu Y, Yang X, Lu T, Han L (2019) Utilization of bio-
diesel byproduct glycerol: production of methyl lactate over Au/
CuO and Sn-Beta binary catalyst under mild reaction conditions.
Energ Convers Manage 196:277–285. https://doi.org/10.1016/j.
enconman.2019.06.011
28. Zhu Z, Xu H, Jiang J, Guan Y, Wu P (2017) Sn-Beta zeolite hydro-
thermally synthesized via interzeolite transformation as efcient
Lewis acid catalyst. J Catal 352:1–12. https://doi.org/10.1016/j.
jcat.2017.04.031
29. Yao JK, Fu KR, Wang YC, Li TD, Liu HX, Wang JG (2017)
Hierarchically porous Sn-β-zeolites via an OSDA-free synthe-
sis. Green Chem 19:3214–3218. https://doi.org/10.1039/C7GC0
1040K
17. Wang X, Wu G, Wang F, Ding K, Zhang F, Liu X, Xue Y (2012)
Base-free selective oxidation of glycerol with 3% H2O2 catalyzed
by sulphonato–salen–chromium(III) intercalated LDH. Catal
Commun 28:73–76. https://doi.org/10.1016/j.catcom.2012.08.014
18. Rodrigues EG, Carabineiro SAC, Chen X, Delgado JJ, Figue-
iredo JL, Pereira MFR (2011) Selective oxidation of glycerol
catalyzed by Rh/activated carbon: importance of support surface
chemistry. Catal Lett 141:420–431. https://doi.org/10.1007/s1056
2-010-0515-9
30. Dijkmans J, Dusselier M, Gabriëls D, Houthoofd K, Magusin
PCMM, Huang S, Pontikes Y, Trekels M, Vantomme A, Giebeler
L, Oswald S, Sels BF (2015) Cooperative catalysis for multistep
biomass conversion with Sn/Al beta zeolite. ACS Catal 5:928–
940. https://doi.org/10.1021/cs501388e
31. Osmundsen CM, Holm MS, Dahl S, Taarning E (2012) Tin-
containing silicates: structure–activity relations. Proc R Soc A
468:2000–2016. https://doi.org/10.1098/rspa.2012.0047
32. Yang X, Wu L, Wang Z, Bian J, Lu T, Zhou L, Chen C, Xu J
(2016) Conversion of dihydroxyacetone to methyl lactate cata-
lyzed by highly active hierarchical Sn-USY at room temperature.
Catal Sci Technol 6:1757–1763. https://doi.org/10.1039/C5CY0
1037C
19. Ning X, Li Y, Yu H, Peng F, Wang H, Yang Y (2016) Promoting
role of bismuth and antimony on Pt catalysts for the selective
oxidation of glycerol to dihydroxyacetone. J Catal 335:95–104.
https://doi.org/10.1016/j.jcat.2015.12.020
20. Lee S, Kim HJ, Lim EJ, Kim Y, Noh Y, Huber GW, Kim WB
(2016) Highly selective transformation of glycerol to dihydroxyac-
etone without using oxidants by a PtSb/C-catalyzed electrooxida-
tion process. Green Chem 18:2877–2887. https://doi.org/10.1039/
C5GC02865E
33. Dijkmans J, Gabriëls D, Dusselier M, Clippel F, Vanelderen P,
Houthoofd K, Malfiet A, Pontikes Y, Sels BF (2013) Produc-
tive sugar isomerization with highly active Sn in dealuminated
β zeolites. Green Chem 15:2777–2785. https://doi.org/10.1039/
C3GC41239C
21. Liu SS, Sun KQ, Xu BQ (2014) Specifc selectivity of Au-cata-
lyzed oxidation of glycerol and other C3-polyols in water with-
out the presence of a base. ACS Catal 4:2226–2230. https://doi.
org/10.1021/cs5005568
34. Maurelli S, Vishnuvarthan M, Chiesa M, Berlier G, Doorslaer SV
(2011) Elucidating the nature and reactivity of Ti ions incorpo-
rated in the framework of AIPO-5 molecular sieves. New evidence
from 31PHYSCORE spectroscopy. J Am Chem Soc 133:7340–
7343. https://doi.org/10.1021/ja202210g
22. Oberhauser W, Evangelisti C, Caporali S, Santo VD, Bossola
F, Vizza F (2017) Ethyl lactate from dihydroxyacetone by a
montmorillonite-supported Pt(II) diphosphane complex. J Catal
350:133–140. https://doi.org/10.1016/j.jcat.2017.04.006
23. Hayashi H, Shigemoto N, Sugiyama S, Masaoka N, Saitoh K
(1993) X-ray photoelectron spectra for the oxidation state of
TeO2–MoO3 catalyst in the vapor-phase selective oxidation of
ethyl lactate to pyruvate. Catal Lett 19:273–277. https://doi.
org/10.1007/BF00771764
35. Bonino F, Damin A, Bordiga S, Lamberti C, Zecchina A (2003)
Interaction of CD3CN and pyridine with the Ti(IV) centers of
TS-1 catalysts: a spectroscopic and computational study. Lang-
muir 19:2155–2161. https://doi.org/10.1021/la0262194
24. Godard N, Collard X, Vivian A, Bivona LA, Fiorilli S, Fusaro
L, Aprile C (2018) Rapid room temperature synthesis of tin-
based mesoporous solids: infuence of the particle size on the
production of ethyl lactate. Appl Catal A 556:73–80. https://doi.
org/10.1016/j.apcata.2018.02.014
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
25. Wang X, Liang F, Huang C, Li Y, Chen B (2016) Siliceous
tin phosphates as efective bifunctional catalysts for selective
1 3