A Polyoxometalate-Based Microfluidic Device for Liquid-Phase Oxidation of Glycerol

Article abstract of DOI:10.1002/cssc.201901057

Peroxidation of glycerol has been carried out in a polyoxometalate (POM)-based microfluidic reactor, which was fabricated on a capillary by using a layer-by-layer strategy. Lactic acid (LA) is produced selectively in high yield with a TOF as high as 20 000 h^?1, compared to a TOF of 200 h^?1 in batch mode. This POM microfluidic reactor is readily prepared, scalable, highly stable, reusable, and also potentially applicable to selective oxidation of other bio-wastes.

Full text of DOI:10.1002/cssc.201901057

Accepted Article
Title: A Polyoxometalate-Based Microfluidic Device for Liquid Phase
Oxidation of Glycerol
Authors: Xiaohong Wang, Meilin Tao, Yiming Li, Xueyan Zhang,
Zonghang Li, and Craig L Hill
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: ChemSusChem 10.1002/cssc.201901057
Link to VoR: http://dx.doi.org/10.1002/cssc.201901057
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201901057
COMMUNICATION
A Polyoxometalate-Based Microfluidic Device for Liquid Phase
Oxidation of Glycerol
Meilin Tao,[a,b] Yiming Li, Xueyan Zhang, Zonghang Li, Craig L. Hill, Xiaohong Wang*[a]
[a]
[a]
[a]
*[b]
Abstract: Peroxidation of glycerol has been performed in
microfluidic reactor based on polyoxometalates (POMs), which has
been fabricated on a capillary using a layer-by-layer strategy. Lactic
acid (LA) is produced selectively in high yield with a TOF as high as
(LA), which proceeds in a continuous flow, an approach that is
potentially scale and safer than under conventional reaction
conditions.
-
1
-1
20,000 h compared to a TOF of 200 h in batch mode. This POM
The fabrication procedure for the H PMo O (henceforth
3
12 40
microfluidic reactor is readily prepared, scalable, highly stable,
reusable, and also potentially applicable to selective oxidation of
other bio-wastes.
“HPMo”)-MR(2) is as follows (Scheme 1):
(1) A commercial glass capillary of 0.5-mm wall thickness
(Yongnian Company) is rinsed with 0.1 M NaOH for 1 h then
with deionized water for 30 min at room temperature (RT).
(
2) The capillary is sequentially filled and flushed with a poly
dimethyl diallyl ammonium chloride (PDDA) solution (2 wt% in
.1 M NaCl) for 1 h at RT to generate the capillary-PDDA
Liquid phase oxidation is central to the conversion of petroleum
or biomass-based materials into important chemicals, especially
fine chemicals,[1] It is also important for the treatment of
environmental pollutants.[2] Polyoxometalates (POMs), which are
multi-metal-oxide cluster anions of W, Mo or V in their highest
0
assembly. The interface of capillary is negative charged, which
can connect with the positive charged PDDA successfully
through electrostatic interaction.
oxidation states,[ are well-established to be important catalysts
3]
(
3) Subsequently, HPMo (0.5 mg/mL) is dispersed on the
2 2 2
for the selective oxidation of organic substrates by O H O
,[4]
[5]
surface of the capillary-PDDA by rinsing at RT for 30 min. The
HPMo trianion is readily bound electrostatically to the highly
positive PDDA-decorated surface forming the capillary-PDDA-
single-layer-HPMo assembly (denoted as HPMo-MR(1), where“1”
represents one layer of HPMo). Finally, repeating the above
treatment several times affords HPMo-MRs(n). Between each
repeat, the capillary is washed with deionized water for 5 min.
The HPMo-MRs(n) are stored in deionized water at 4 °C after
each use until next use.
and organic peroxides.[6] Some of the long-time and general
problems in catalytic oxidation are (a) selectivity (avoiding
overoxidation, competing low-selectivity autoxidation side-
reactions, etc.), (b) safety, particularly in peroxide-based
2
oxidations and O -based ones with high-vapor-pressure organic
substrates under pressure, and (c) catalyst stability.[7] (d) In
addition, green heterogeneous catalytic reactions in
conventional batch or semi-batch mode frequently have mass
transfer challenges involving multiple phases (substrates,
oxidants and solid catalysts). Catalyst leaching and recovery can
also be issues in batch reactors.[8] Microfluidic systems or
continuous-flow microreactors (MRs) have some competitive
advantages over microscopic batch or semi-batch reactors: MRs
(
a) are frequently more environmentally benign, (b) have higher
[
9]
mass and heat transfer efficiencies, (c) exhibit higher surface-
to-volume ratios, and (d) generate higher outputs at higher
levels of safety. Such benefits have led to increasing use of MRs
by both academic and industrial investigations.[10] Micro-reactor
technology has been reviewed for use in both oxidative or
hazardous reactions.[11]
Scheme 1. Schematic for preparation of HPMo-MR(2) based on electrostatic
assembly.
In this communication, we describe for the first time a micro-
fluidic catalytic system based on POMs for organic cascade
reactions. We henceforth refer to this device as POM-
microfluidic reactor (POM-MR). The H O oxidation of glycerol, a
2 2
large-tonnage by-product of biodiesel production, that is highly
efficient and highly selective for the desired product, lactic acid
Features of these HPMo-containing microfluidic reactors
include: (1) easy fabrication of POM active sites via layer-by-
layer incorporation of polyoxometalate anions to PDDA
polycation polymers; (2) excellent retention of the POM by the
PDDA-surface, a feature illustrated recently by quantitative
removal of POM water oxidation catalysts from aqueous
solutions by semi-conductor photo-electrode surfaces bearing
surface-bound ammonium cations;[12] (3) low restriction of
substrate fluid flowing through the reactor; (4) a very large
specific interfacial area per unit of volume in the MR, which
leads to efficient liquid-solid reaction.
＋
＋
[
a]
M. Tao , Y. Li , X. Zhang, Z. Li, Prof. X. Wang
Key Lab of Polyoxometalate Science of Ministry of Education
Northeast Normal University
Changchun 130024, P. R. China
E-mail: wangxh665@nenu.edu.cn
Prof. C. L. Hill
Department of Chemistry
Emory University
[
[
b]
+]
The fabrication procedure can be followed to some extent by
the SEM images (Figure 1). The initial smooth surface of the
bare capillary becomes covered with small mounds of uniform
depth, width, and shape in the capillary-PDDA. The surface of
1515 Dickey Dr., Atlanta, 30322 GA, USA
These authors contributed equally to this work
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201901057
COMMUNICATION
the capillary-PDDA-HPMo bears some uniform and well defined
particles with a size of ~10 nm. When this MR has two and three
HPMo coating layers, HPMo-MRs(2 or 3), more particles with
(1 M) for a while under vacuum until the capillary was dissolved,
and then the final solutions were tested by Uv-vis. The
concentration of HPMo-MR(1), HPMo-MR(2) and HPMo-MR(3)
-7
-7
30-nm diameters are observed.
divided by capillary internal area were 2.8×10 , 6.7×10 , and
.3×10-7 mol/cm2, respectively. Thermogravimetric analysis
TGA) was used to confirm the successfully fabrication of HPMo-
MR via the layer-by-layer method. The evident weight reduction
8
(
a
b
[
15]
around 550 °C is due to HPMo decomposition. The TGA data
-7
-7
-7
2
indicate that a loss of 2.7×10 , 6.4×10 , and 8.9×10 mol/cm in
HPMo-MR(1), HPMo-MR(2), and HPMo-MR(3), respectively
(
Figure S3).
2 2
Initially the influence of the molar ratio of H O to glycerol was
evaluated (Table 1) at 2:1, 4:1, 6:1, and 8:1 values. As expected,
increasing H enhances the conversion with a maximum value
c
d
e
2
O
2
of 77 % in the case of HPMo-MR(2) at the molar ratio of 4:1. The
maximum yield to LA (54 %) is obtained at this molar ratio, and
decreases at higher peroxide concentrations due to
overoxidation. The products of overoxidation include acetic acid,
formic acid, CO
2
, et al.[ In order to evaluate the efficiency of
HPMo-MR(2) in peroxidative conversion of glycerol, the reaction
16]
Figure 1. SEM images illustrating fabrication of HPMo-MRs through layer-by-
layer assembly. (a) bare capillary, (b) capillary-PDDA, (c) HPMo-MR(1), (d)
HPMo-MR(2), and (e) HPMo-MR(3). The coated capillary was cut at an
inclined angle to expose the inner wall of the capillary under the SEM lens.
was conducted in batch mode under the H
2
O
2
-to-glycerol molar
o
ratio (4:1) and same temperature (60 C) using 50 cm of ground
HPMo-MR(2) (Table S1, entries 1-5). In batch mode, the highest
yield to LA obtained as 50 % at 75 % conversion of glycerol after
480 min.
Electroosmotic flow (EOF) was also used to determine the
successful fabrication of HPMo-MRs. The bare capillary being
coated by negatively charged silanol groups showed a positive
EOF (from anode to cathode). After application of the PDDA
coating, the EOF was reversed due to the residues of the
positively charged PDDA units. After treatment of the capillary-
PDDA with HPMo solution, the direction of EOF changed back
to positive. As a result, the alternative direction of the EOFs
supports the successful formation of the HPMo-modified
capillary. Clearly, the existence of the positively-charged PDDA
and the negatively-charged HPMo leading to multiple strong
anion-cation interactions kept the POM molecules tightly bound
on the microfluidic reactor surfaces.
The product distribution was determined to be LA,
glyceraldehyde (GCA) and dihydroxyacetone (DHA), and other
uncertain oxidative products. According to a previous report,[
13]
glycerol can be oxidized by H
(Scheme S1) in batch mode. H PMo11VO40@SiO
to show 68 % selectivity to formic acid and 27 % selectivity to
2
O
2
through several pathways
4
2
was reported
glycolic acid at 61 % conversion using a H
2
O
2
to glycerol molar
ratio = 5:1 at 70 oC for 12 h with 0.06 mg of catalyst.
[13c]
The
reaction proceeded via pathway I for peroxidation of glycerol
with a TOF =195 h-1 (TOF = [Gly]×conversion/usage of POM
×reaction time). Under our batch reaction conditions, glycerol
was oxidized to LA via pathway II with a TOF =200 h-1 in the
presence of ground HPMo-MR(2). This noteworthy rate is
attributed to the molecular environment in HPMo-MR, in which
HPMo is embedded in the hydrophobic PDDA molecules which
favors LA generation.[17] In our previous report, homogeneous
HPMo (no PDDA present) had a TOF of 32 h-1 in batch mode [18]
PDDA clearly alters the microenvironment of HPMo-MR and
thus influences the efficiency of glycerol peroxidation. In addition,
the residence time also plays an important role in glycerol
conversion and LA yield (Table 1, entries 8).These increase to
98 and 93 %, respectively, as the residence time is increased to
16 min.
The FTIR spectra of HPMo-MRs with varying layers are
illustrated in Figure S1. The four characteristic POM peaks from
-1
1
056 to 720 cm , are similar to and in registry with the
corresponding bands in pure H 40 (1060 (νas P-O), 965
νas Mo=O), 882 (νas Mo-O-Mo) and 748 cm (νas Mo-O-Mo))
3
PMo12O
-1
(
with some red shifts. The broader bands may because of the
grinding before recording on the KBr pellets. This indicates that
after coating the capillary PDDA with the POM catalyst, HPMo,
this Keggin POM structure remains intact. Further, these
vibrational bands are consistent with some interaction between
PDDA and a terminal oxygen of Mo=O which results in red
[
13]
shifting of the asymmetric stretch, νas, of Mo=O. The shifts
-1
increase from 4, 10 and to 18 cm , for HPMo-MR(1), HPMo-
MR(2) and HPMo-MR(3), respectively, which is attributed to the
increasing interaction between the PDDA and the HPMo units.
The amount of HPMo in the MRs was determined by Uv-Vis
spectroscopy from the intensity of peaks at 220 and 260 cm-1
corresponding to the Keggin structure of HPMo [14] (Figure S2).
A certain length (20 cm) of HPMo-MR(n) were immersed in HF
The performance (yield and rate) of HPMo producing LA
increases dramatically ongoing from batch mode to microreactor
mode: (TOFLA = molar yield of LA/molar of catalyst × reaction
time) increases from 200 h-1 to 20000 h-1.
A further attempt to increase the concentration of glycerol
from 0.2 to 1.0 M (Table S1, entries 7, 11-14) was conducted.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201901057
COMMUNICATION
The conversion initially increased together with the selectivity for
LA, and the maximum value was obtained at 0.6 M glycerol. The
decrease in conversion and selectivity was observed as more
DHA formed with increasing glycerol concentration; the DHA
requires more time to transform to LA than GCA does.
Increasing the temperature leads to an expected decrease in
selectivity (Table S1, entries 6-10): LA is easily over-oxidized at
the higher temperature in the presence of H O . The most
2 2
important parameters for microfluidic reactor performance are
the length and diameter of the HPMo layer. Increasing the layers
The HPMo-MR(2) was reused several times with very little
loss inactivity (Figure 2). Conversion and selectivity decreased
from 98 % and 95 % to 93 % and 86 %, respectively. After each
catalytic run, the catalytic MR was not treated before reuse. We
attempted to clarify the loss of the activity by examination of the
solution from the HPMo-MR(2) reactions. No HPMo was
detected in most cases (assessed by UV-Vis spectra). The
strong electrostatic attraction of the multicharged PDDA
surfaces and the multicharged (3-) HPMo polyanion appears to
eliminate POM leaching. The loss in activity is likely attributable
to the adsorption of products during the peroxidation reaction.
Conversions of glycerol as high as 93 % are maintained after ten
cycles. The FTIR and SEM of HPMo-MR(2) after reaction also
confirmed its stability (Figure S5-6). In normal peroxidatic
in the MR from one to three doesn’t increases conversion and
Table 1. Optimization of the peroxidation of glycerol under the catalysis of
a
HPMo-MR
H
2 2
O :
Yields %
3 2 2
reaction in the presence of H PMo12O40 and H O , metal
ent
ry
Gly
(mol/
mol)
catalyst
Con%b
peroxide is usually formed. While in the FTIR spectrum after
reaction, no peaks of molybdenum peroxide (around 578 and
DHA+GCA
LA
others
-1
[19]
5
32 cm ) appeared. It means the HPMo-MR(n) catalysts were
1
2
3
4
5
6
none
4:1
4:1
2:1
4:1
6:1
8:1
4:1
4:1
2
0
5
0
2
2
stable under the reaction condition, which was attributed to the
capillary-PDDA microfluidic structure.
HPMo-MR(1)
HPMo-MR(2)
HPMo-MR(2)
HPMo-MR(2)
HPMo-MR(2)
HPMo-MR(3)
HPMo-MR(2)
31
23
77
88
94
51
98
24
18
54
42
32
28
93
1
3
Conclusions
20
31
38
19
4
3
In summary, a POM-containing microfluidic reactor leads to a
scalable, selective and high yield production of lactic acid (LA).
A facile protocol for fabrication of these POM-MR catalysts is
given. Their use eliminates the catalyst-product separation
problem, minimizes mass transfer difficulties, and facilitates a
very short residence time. The use of HPMo-MR(2) produces LA
in 95 % selectivity at 98 % conversion of glycerol under
optimized reaction conditions, with a TOF 100 times higher than
batch mode.
13
20
4
7
8c
1
[
a] The length and the diameter of HPMo-MR are 50 cm and 100 μm,
o
respectively. Conditions: 0.6 M glycerol, 60 C, residence time 8 min. [b]
Conversion (percent). [c] Residence time 16 min.
selectivity necessarily (Table 1, entries 2, 4, 7), and the highest
yield of LA is obtained by HPMo-MR(2). It is likely that there are Experimental Section
too few active sites in HPMo-MR(1), while too many layers might
lead to the stacking inside the capillary which decreases the
fluidity. The performance of HPMo-MR(2) with different
diameters is consistent with this hypothesis (Figure S4a). The
length of microfluidic reactor also influences the efficiency:
longer MR give rise to higher yields of LA (Figure S4b).
The materials and methods are described in the supporting
information (SI). The main components of the HPMo-MR
microreactors are illustrated in Scheme 2. Details of fabrication
and the catalytic oxidation procedures are given in the SI.
Scheme 2. Key components in glycerol peroxidation to LA catalyzed by
HPMo-MR.
Acknowledgements
Figure 2. Reusability of the HPMo-MR(2) in catalytic peroxidation of glycerol.
Reaction conditions: 0.6 M glycerol, molar ratio of H O to glycerol as 4:1, 60
2 2
ºC, residence time 16 min, 100 μm diameter and 50 cm length. Code: glycerol
conversion, black; LA selectivity, red.
We thank the National Natural Science Foundation of China
(51578119) and Jilin Provincial Science and Technology
Department (20180414069GH) for support of this research.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201901057
COMMUNICATION
ChemCatchem. 2013, 5, 2091-2099; d) C.P. Park and D. P, Kim, J. Am.
Chem. Soc. 2010, 132, 10102-10106.
Keywords: Microfluidic reactors • Electrostatic interaction •
Glycerol oxidation • Lactic acid
[
11] a) F. Cavani and J. H. Teles, ChemSusChem. 2009, 2, 508-534; b) H. P.
L. Gemoets, Y. Su, M. Shang, V. Hessel, R. Luque and T. Noël, Chem.
Soc. Rev. 2016, 45, 83-117; c) B. Gutmann, D. Cantillo and C. O.
Kappe, Angew. Chem. Int. Ed. 2015, 54, 6688-6728.
[
1]
2]
a) W. J. Liu, W. W. Li, H. Jiang and H. Q. Yu, Chem. Rev. 2017, 117,
6367-6398; b) G. W. Huber, S. Iborra and A. Corma, Chem. Rev. 2006,
106, 4044-4098.
[
12] a) S. M. Lauinger, J. M. Sumliner, Q. S. Yin, Z. H. Xu, G. J. Liang, E. N.
Glass, T. Q. Lian and C. L. Hill, Chem Mater. 2015, 27(17), 5886-5891;
b) S. M. Lauinger, B. D. Piercy, W. Li, Q. S.Yin, D. L.Collins-Wildman,
E. N.Glass, M. D.Losego, D. W.Wang, Y. V.Geletii and C. L.Hill, ACS
Appl Mater Interfaces. 2017, 9(40), 35048-35056.
[
a) T. J. Morgan and R. Kandiyoti, Chem. Rev. 2014, 114, 1547-1607; b)
P. Mäki-Arvela, I. L. Simakova, T. Salmi and D. Y. Murzin, Chem. Rev.
2014, 114, 1909-1971.
[
3]
4]
S. S. Wang and G. Y. Yang, Chem. Rev. 2015, 115, 4893-4962.
[
T. Punniyamurthy, S. Velusamy and I. Iqbal, Chem. Rev. 2005, 105,
[
13] a) J. Zhu, P. Wang and M. Lu, Catal. Sci. Technol. 2015, 5, 3383-3393;
b) C. Gao, Z. Li, L. Zhang, C. Wang, K. Li, C. Ma and P. Xu, Green
Chem. 2015,17, 804-807; c) M. Yuan, D. Li, X. Zhao, W. Ma, K. Kong,
W. Ni, Q, Gu and Z. Hou, Acta Physico-Chimica Sinica. 2018, 34, 886-
2
329-2364.
N. Mizuno, K. Yamaguchi and K. Kamata, Coord. Chem. Rev. 2005, 249,
944-1956.
[5]
[6]
[7]
1
K. Kastner, J. Forster, H. Ida, G. N. Newton, H. Oshio and C. Streb,
Chem. Eur. J. 2015, 21, 7686-7689.
895; d) A. M. Pemberea and Z. Luo, Phys. Chem. Chem. Phys. 2017,
19, 6620-6625.
a) Z. Shi, C. Zhang, C. Tang and N. Jiao, Chem. Soc. Rev. 2012, 41,
[
14] A. Ayati, A. Ahmadpour, F. F. Bamoharram, M. M. Heravi and M.
Sillanpää, Gold Bull. 2012 45, 145-151.
3381-3430; b) S. D. McCann and S. S. Stahl, Acc. Chem. Res. 2015,
48, 1756-1766; c) S. E.Allen, R. R. Walvoord, R. Padilla-Salinas and M.
[
15] F. J. Méndez, A. Llanos, M. Echeverría, R. Jáuregui, Y. Villasana, Y.
Díaz, G. Liendo-Polanco, M. A. Ramos-García, T. Zoltan and J. L. Brito,
Fuel. 2013, 110, 249-258.
C. Kozlowski, Chem. Rev. 2013, 113, 6234-6458.
[
8]
9]
T. Noël and S. L. Buchwald, Chem. Soc. Rev. 2011, 40, 5010-5029.
M. Lechner, R. Güttel and C. Streb, Dalton Trans. 2016, 45, 16716-
[
[
[
[
16] S. Poloju, K. Cholkar, G. K. Kouassi, J. D. V. Horn, K. S. Rangappa, N.
16726.
M. M. Gowda, Am. J. Org. Chem. 2012, 2, 58-62.
[
10] a) S. M. Kim, Y. H. Park, S. W. Seo, C. P. Park, S. Y. Park and I. In,
Adv. Mater. Interfaces. 2015, 2, 1500174; b) J. H. Park, C. Y. Park, H.
S. Song, Y. S. Huh, G. H. Kim and C. P. Park, Org. Lett. 2013,15,752-
17] M. Tao, N. Sun, Y. Li, T. Tong, M. Wielicako, S. Wang and X. Wang, J.
Mater. Chem. A. 2017, 5, 8325–8333.
18] M. Tao, X. Yi, I. Delidovich, R. Palkovits, J. Shi and X. Wang,
ChemSusChem. 2015, 8, 4195-4201.
755; c) B. B. Xu, Y. L. Zhang, S. Wei, H. Ding and H. B. Sun,
[
19] M. Wang, Q. Wang, R. Feng, S. Zhang, Fine Chem. 2016, 33, 557-563.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201901057
COMMUNICATION
COMMUNICATION
This article is protected by copyright. All rights reserved.