pubs.acs.

org/journal/ascecg Research Article

Greening Oxidation Catalysis: Water as a Solvent for Efficient Alkene

Epoxidation over a Titanosilicate/H2O2 System
Yunkai Yu, Nan Fang, Zhen Chen, Dongxu Liu, Yueming Liu,* and Mingyuan He
Cite This: ACS Sustainable Chem. Eng. 2022, 10, 11641−11654 Read Online

ACCESS Metrics & More Article Recommendations *

sı Supporting Information
Downloaded via MANIPAL ACADEMY OF HIGHER EDUCATION on October 27, 2024 at 14:49:36 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: Herein, we disclose the application of water (H2O) solvent in

titanosilicate-catalyzed epoxidation of alkenes. A highly efficient H2O-solvent
propylene epoxidation system (Ti-MWW/H2O/H2O2/propylene/K+) was
achieved, possessing high activity, selectivity, and stability. Our investigation
shows that this excellent catalytic performance is ascribed to the structure variation
of H2O induced by K+. The superior H2O structure is the one that donates two
and accepts one hydrogen bond (HB), which is known as the “DDA” H2O
structure. The current report benefits sustainable applications in titanosilicate-
catalyzed alkene epoxidation and conforms to the principles of green chemistry.
KEYWORDS: H2O solvent, titanosilicates, alkene epoxidation, H2O2 activation, H2O structures

1. INTRODUCTION The use of H2O as a solvent could avoid these side reactions
The extensive use of solvents has long been an essential part of derived from the introduction of organic solvents in Scheme
conventional organic synthesis; however, it is one of the major S1,20,21 which greatly reduce the subsequent energy con-
contributors to industrial waste streams, and escaping of sumption for the separation of epoxides. Besides, H2O could
volatile organic solvents into the environment also causes directly participate in the whole reaction process by the
environmental issues.1−3 To minimize environmental impacts interaction with Ti active sites or Ti−OOH species (Scheme
caused by organic solvents, the use of green solvents (e.g., S2). Generally, the alkene epoxidation included the activation
water, supercritical carbon dioxide, and ionic liquids) has been of H2O2 to form Ti−OOH intermediates and the further
intensively encouraged for green chemistry and sustainability transfer of active oxygen to the reactant alkenes.22−27 When Ti
over the past 2 decades.4,5 Water (H2O), the most abundant active sites in titanosilicates were exposed to H2O, the Ti−O−
and important solvent on earth, is cheap, green, readily Si bridge would be broken through the hydrolysis to form Ti−
available, and nonflammable. It possesses unique physical and OH and Si−OH groups.13,28 Corma et al. proposed that the
chemical properties such as a strong polarity, large heat H2O molecule could stabilize Ti−OOH species by forming
capacity, and hydrogen-bonding effect, endowing it with species II with a five-membered ring structure.16 Through in
practical advantages as the greenest solvent in many important situ XANES characterization, Bordiga et al. suggested that H2O
fields of research.6−8 could transform the hydroperoxo Ti−OOH species into the
Epoxides are vital chemical intermediates to produce peroxo Ti−OO− species.29 By mechanistic insights into alkene
solvents, materials, and biologically active ingredients within epoxidation with H2O2 over Ti-polyoxometalates (Ti-POMs),
the food, construction, cosmetic, automobile, and pharma- it was demonstrated that the formation of epoxides through
ceutical industries.9 Titanosilicate-catalyzed epoxidation of the Ti−OOH intermediate is energetically more favorable than
alkenes involving hydrogen peroxide (H2O2) as the oxidant through the Ti−OO− intermediate.22,23 Besides, the presence
is a vital route for the production of epoxides,10 following the of confined H2O molecule clusters within hydrophilic pores
principles of green carbon science.11,12 It is an environmentally promoted the stability of epoxidation transition states by the
benign chemical process that avoids equipment corrosion, hydrogen bonds (HBs) between Ti−OOH species, neighbor-
environmental pollution, and low atom economy in the ing H2O molecules, and (Si−OH)4 groups.30,31 Many reports
conventional chlorohydrin process. However, a limitation of
this epoxidation reaction in terms of sustainability is the use of
volatile and toxic solvents, such as methanol (MeOH)13 or Received: June 15, 2022
acetonitrile (MeCN).14−16 One of the problems related to Revised: August 9, 2022
organic solvents (e.g., MeOH) is that it generates a series of Published: August 23, 2022
trace impurities from the side reactions (Scheme S1),17
resulting in the deactivation of the catalyst and the reduced
selectivity of epoxides.18,19

© 2022 American Chemical Society https://doi.org/10.1021/acssuschemeng.2c03524

11641 ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Scheme 1. Proposed Activation Mechanism Diagram of H2O2 without (Pathway I) and with (Pathway II) the Introduction of
H2O Based on the Role of H2O in Facilitating the Proton Transfer on H2O2 Activation32−35

indicated that the H2O molecule could act as a mediator to Our recent work37 suggested that by increasing the acid-treated time,
facilitate the proton transfer on H2O2 activation to form a Ti− the content of Ti(OSi)3OH sites increased from the transformation of
OOH intermediate and lower the H2O2 activation barrier the Ti(OSi)4 sites. To investigate the influence of hydrophobicity on
through the HB interaction among the H2O molecule, the propylene epoxidation in H2O solvent, the hydrophobic Re-Ti-MWW
zeolite (Si/Ti = 52, Si/B = 49) was prepared by a reversible structural
H2O2 molecule, and Ti−OH active site according to density rearrangement of Ti-MWW according to the literature.38 It was
functional theory (DFT) calculations.32−35 More specifically, carried out at a piperidine/SiO2 molar ratio of 1 and a H2O/SiO2
the H2O molecule accelerated the proton transfer through the molar ratio of 10 at 443 K for 1 day. The resulting solid was further
process of accepting a proton of H2O2 followed by donating a recovered by filtration, drying, and calcination to obtain Re-Ti-MWW.
proton of H2O itself, as shown in Scheme 1.32−35 Based on the The TS-1 zeolite (Si/Ti = 76) was prepared in a two-step hydrolysis
practical advantages of the H2O-solvent system and the method.39
important role of H2O molecules on the epoxidation reaction, 2.2. Catalyst Characterization. X-ray diffraction (XRD) analysis
it is reasonable that the H2O solvent is a preferred alternative was achieved on a Rigaku Ultima IV diffractometer with Cu Kα
to conventional organic solvents for the titanosilicate-catalyzed radiation (1.5406 Å). The UV−visible spectra (UV−vis) were
measured under ambient conditions with a spectrophotometer (UV-
alkene epoxidation system. 2400PC) using BaSO4 as a background. ICP spectrometry was
The epoxidation of propylene is regarded as a typical performed with an IRIS Intrepid II XSP spectrometer. Scanning
epoxidation reaction. Here, we proposed a highly efficient electron microscopy (SEM) was performed with a Hitachi S-4800
propylene epoxidation system in H2O solvent (Ti-MWW/ microscope. Nitrogen adsorption was measured with a BEL-MAX
H2O/H2O2/propylene/K+) for the first time, which possesses instrument at 77 K, and all samples were treated at 573 K for 6 h in a
high activity, selectivity, and stability. To understand this vacuum before adsorption−desorption. The H2O physisorption
excellent catalytic performance, we investigated the influence measurements were performed using a BEL SORP instrument at
of diffusion, adsorption/desorption, H2O structures, and Ti 298 K. Before measurements, all catalysts were degassed at 623 K for
active sites on propylene epoxidation in H2O solvent. It is 12 h under vacuum. 29Si MAS NMR spectra were performed on a
spectrometer (Varian VNMRS-400WB). Q8M8 ([(CH3)3SiO]8-
revealed that the catalytic performance of propylene
SiO12) was applied for the pulse calibration. The Raman spectra of
epoxidation in H2O solvent depends on the structures of H2O were recorded with a SmartRaman spectrometer (DXR2) at
H2O molecules, i.e., the “DDA” structure (that donates two room temperature. The band resolution was 1 cm−1, and the
and accepts one HB). The proportion of DDA H2O can be excitation wavelength was 532 nm. The OH stretching Raman spectra
increased by introducing a certain amount of K+ to break the of H2O were recorded in the region between 2500 and 4000 cm−1.
HB of H2O molecules, resulting in an excellent catalytic The infrared spectra (IR) of titanosilicates and the liquid were
performance for propylene epoxidation catalyzed by Ti-MWW obtained on a Nicolet Nexus 670 spectrometer with 2 cm−1
in H2O solvent. resolution. The spectra in the hydroxyl stretching region were
collected at ambient temperature after the treatment at 723 K for 2 h
in a vacuum. The IR spectra in the framework vibration region of
2. EXPERIMENTAL SECTION titanosilicates were recorded using the KBr technique at 298 K.
2.1. Catalyst Preparation. A Ti-MWW lamellar precursor was 2.3. Catalytic Reactions. 2.3.1. Propylene Epoxidation. The
synthesized at 443 K for 7 days with the introduction of boric acid.36 reagents with the catalyst (0.1 g), solvent (20 mL), and H2O2 (30 wt
The molar ratio of the sol was 1.0 SiO2: 0.04 TiO2: 1.4 PI: 0.67 B2O3: % aqueous solution, 30 mmol) were loaded into a bath. Propylene
19 H2O. The obtained precursor was acid-treated with 2 M nitric acid was charged at constant pressure (0.4 MPa), and the mixture was held
under reflux conditions for 20 h and was recovered by filtration, under vigorous stirring at 323 K for 50 min. The products were
drying, and calcination to obtain Ti-MWW (Si/Ti = 52, Si/B = 46). If measured by gas chromatography (Agilent GC-7890A, DB-WAX
the acid-treated time was altered, the Ti-MWW-xh zeolites were capillary column) with MeCN for H2O solvent and toluene for
obtained, where x (x = 2, 8, 12, and 20 h) represented the hours of organic solvents as the internal standard. The residual free H2O2 was
the acid-treated time. If not specified, the acid-treated time was 20 h. titrated with Ce(SO4)2 solution.

11642 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

2.3.2. 1-Pentene/1-Hexene Epoxidation. Briefly, 50 mg of the

catalyst, 10 mL of the solvent, 10 mmol of 1-pentene/1-hexene, and
10 mmol of H2O2 (30 wt % aqueous solution) were mixed under
vigorous stirring at 333 K for 2 h. The products were measured with
an Agilent GC-7890A (DB-WAX capillary column) with cyclo-
hexanone as the internal standard. The residual H2O2 was titrated
with Ce(SO4)2 solution.
2.3.3. H2O2 Activation Reaction. The reagents with the catalyst
(0.1 g), H2O (10 mL), and H2O2 (30 wt % aqueous solution, 10
mmol) were mixed under vigorous stirring at 333 K for 2 h. The
residual free H2O2 was titrated by Ce(SO4)2 solution.
The additives (if added) were chlorine salts (LiCl, NaCl, KCl, and
CsCl), simplified as the corresponding alkali metal ions (Li+, Na+, K+,
and Cs+). The concentrations of the additive were 0.13−0.54 mol/L.
If not specified, the concentration of the additive was 0.27 mol/L.
Note: In our experiment, acetone was used as the solvent or added
to the reaction to increase the solubility of propylene. However, in the
presence of H2O2, acetone can be oxidized to very hazardous acetone Figure 1. Stability in propylene epoxidation over Ti-MWW in H2O
cyclic peroxides.40 In industrial application, acetone should be solvent with the additive of K+. Propylene epoxidation conditions: K+
avoided. content 0.27 mol/L, others were the same as in Table 1.

3. RESULTS AND DISCUSSION (Table 1, entry 8). The lower PO selectivity in H2O solvent is
3.1. Highly Efficient H2O-Solvent Propylene Epox- due to hydrolysis of the product epoxide over weak acid sites
idation System Catalyzed by Ti-MWW with the (e.g., the Si−OH(Ti) group and Ti−OOH species) to produce
side products.15,20,41 The hydrolysis can be suppressed via
Table 1. Propylene Epoxidation Catalyzed by Ti-MWW in poisoning the acid sites by MeCN16 or K+.20,42−45 Thus, using
Various Solvents H2O as the solvent with the introduction of K+ offers a
a a a competitive system relative to the MeCN system for efficient
PO H2O2 PO
yield conv. sel. a
H2O2 b
TON propylene epoxidation. This system also shows excellent
entry solvent (%) (%) (%) eff. (%) value stability with little to no decrease in PO yield and H2O2
1 pure H2O 38.2 49.1 95.3 77.8 358 conversion after five catalytic cycles (Figure 1). It implies that
2 c
MeOH + H2O 31.3 43.7 80.3 71.6 293 the deactivation of Ti-MWW in propylene epoxidation in H2O
3 c
EtOH + H2O 21.9 32.7 100 62.3 205 solvent is not remarkable.
4 c
1-PrOH + H2O 13.1 25.0 100 52.4 123 Therefore, through the introduction of a certain amount of
5 c
i-PrOH + H2O 20.5 32.8 100 57.9 192 K+, Ti-MWW presents considerable activity, selectivity, and
6 c
MeCN + H2O 88.7 99.5 100 89.1 830 stability in H2O-solvent propylene epoxidation, implying that
7 c
acetone + H2O 60.4 79.7 95.0 75.9 565 the H2O solvent system (Ti-MWW/H2O/H2O2/propylene/
8 d
pure H2O + K+ 89.5 98.6 100 90.8 838 K+) is expected to be applied for the industrial application.
9 e
pure H2O 38.6 48.3 95.5 79.9 361 3.2. Key to Efficient Propylene Epoxidation in H2O
10 f
pure H2O 39.2 48.6 90.6 80.7 367 Solvent. To explore the underlying reason for the highly
a
Propylene epoxidation conditions: Ti-MWW 100 mg, solvent 20 mL, efficient propylene epoxidation in H2O solvent with the
propylene pressure 0.4 MPa, H2O2 (30 wt %) 30 mmol, temp 323 K, assistance of K+, the influences of diffusion, adsorption/
time 50 min. bTurnover number per Ti site for PO yield. cVolume desorption, H2O structures, and Ti active sites on the
ratio of the organic solvent to H2O is calculated to be 89:11. dKCl performance of propylene epoxidation in H2O solvent were
content of 0.27 mol/L. eA small amount of organic solvent acetone (1 investigated.
mL) was introduced for increasing the solubility of propylene. fUsed 3.2.1. Influence of Diffusion and Adsorption/Desorption
catalyst is Re-Ti-MWW. on Propylene Epoxidation. When the solvent is pure H2O, the
epoxidation occurred in a three-phase (organic−aqueous−
Introduction of K+. To achieve the superior catalytic solid) reaction system because the reactant propylene is
performance of propylene epoxidation in H2O solvent, the immiscible with the aqueous solution. A small amount of
hydrophilic Ti-MWW was used as the catalyst. As a organic solvent acetone (1 mL) was introduced into the
comparison, other solvents (protic solvents: MeOH, EtOH, epoxidation reaction in H2O solvent for increasing the
1-PrOH, and i-PrOH; aprotic solvents: MeCN and acetone) solubility of propylene, which, however, could not improve
were also applied (with H2O as the cosolvent from H2O2 the catalytic performance of propylene epoxidation (Table 1,
solution). Table 1, entries 1−7 show that the yield of PO and entry 9). This indicates that the diffusion limitation caused by
TON values decreases in the order of MeCN > acetone > pure the immiscibility can be avoided under vigorous stirring
H2O > MeOH > EtOH > i-PrOH > 1-PrOH. Ti-MWW conditions. It is consistent with the previous report that the
presents a lower PO yield in H2O solvent than in MeCN epoxidation of small alkenes was diffusion unlimited when the
solvent (Table 1, entries 1 and 6), which was reported as the crystal size of titanosilicates was lower than 38 μm based on
best solvent.14,15 Interestingly, the PO yield in H2O solvent is the analysis of the Weisz modulus.46
greatly improved and even exceeds that in MeCN solvent (89.5 In general, the hydrophobic surface of titanosilicates could
vs 88.7% in Table 1, entry 8), by introducing a certain amount adsorb and enrich organic reactants to enhance the transfer of
of K+ (0.27 mol/L). The selectivity of PO is lower in H2O active oxygen to bring about a high catalytic activity.38,47,48
solvent than in MeCN solvent (95.3 vs 100%, entries 1 and 6 However, the enhanced hydrophobicity is likely to suppress
in Table 1) and it is improved to 100% by introducing K+ the adsorption of the oxidant, H2O2. To investigate the
11643 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 2. Raman spectra (A) and the deconvolution bands (B−H) at 3051 cm−1 (DAA), 3233 cm−1 (DDAA), 3393 cm−1 (DA), 3511 cm−1
(DDA), and 3628 cm−1 (free) in the O−H stretching of H2O in pure H2O and cosolvents (MeOH + H2O, EtOH + H2O, 1-PrOH + H2O, i-PrOH
+ H2O, MeCN + H2O, and acetone + H2O). The corresponding proportion of various H2O structures (DDAA, DA, DDA, and free) in Raman
spectra (I) is plotted against the PO yield in propylene epoxidation catalyzed by Ti-MWW in various solvents (Table 1). The volume ratios of the
organic solvent to H2O are 89:11.

Table 2. Corresponding Proportions of Various H2O Structures with the Introduction of Organic Solvents from Raman (and
IR) Spectra
a a a a a
entry solvent DAA (%) DDAA (%) DA (%) DDA (%) free (%)
1 pure H2O 2.6 (5.5) 35.9 (56.2) 27.8 (31.9) 31.7 (5.5) 1.9 (0.9)
b
2 MeOH + H2O 0 (0) 26.1 (44.0) 43.7 (50.3) 27.7 (4.7) 2.6 (1.0)
b
3 EtOH + H2O 0 (0) 27.6 (41.9) 44.2 (52.7) 25.1 (4.3) 3.1 (1.1)
b
4 1-PrOH + H2O 0 (0) 26.7 (36.7) 42.9 (59.4) 25.4 (2.2) 5.0 (1.7)
b
5 i-PrOH + H2O 0 (0) 29.9 (38.1) 42.0 (57.1) 24.7 (4.1) 3.5 (0.7)
b
6 MeCN + H2O 0 (0) 14.1 (18.1) 37.3 (53.5) 46.0 (26.9) 2.6 (1.5)
b
7 acetone + H2O 0 (0) 15.5 (19.3) 43.0 (60.5) 39.5 (18.6) 2.0 (1.6)
a
Proportions of various H2O structures with the introduction of organic solvents were obtained from the deconvolution results in Raman (Figure
2) and IR (Figure 4) spectra. The IR results are enclosed in parentheses. bVolume ratios of the organic solvent to H2O are 89:11.

influence of hydrophobicity of titanosilicates on propylene MWW (the detailed characterization data and analyses are
epoxidation in H2O solvent, a hydrophobic Re-Ti-MWW38 presented in the SI, Section S1). Table 1, entry 10 shows that
was prepared by a reversible structural rearrangement of Ti- Re-Ti-MWW exhibits a propylene epoxidation activity similar
11644 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 3. Raman vibration modes ((A): 2750−3000 cm−1 for γC−H; (B): 1000−2200 cm−1 for γC−O, γC�N, and γC=O) of organic−H2O cosolvents
(volume ratios of the organic solvent to H2O are 89:11).

Table 3. Raman Vibration Modes of Pure Solvents and the Organic−H2O Cosolvents
entry solvent γC−H (cm−1) γC−O (cm−1) γC�N (cm−1) γC=O (cm−1)
a
1 pure MeOH 2946 2836 1037
b
2 MeOH + H2O 2952 2843 1022
a
3 pure EtOH 2973 2927 2876 1063
b
4 EtOH + H2O 2979 2933 2884 1047
a
5 pure 1-PrOH 2961 2938 2913 2878 1068
b
6 1-PrOH + H2O 2968 2940 2918 2883 1061
a
7 pure i-PrOH 2972 2938 2919 2881 1132
b
8 i-PrOH + H2O 2980 2939 2920 2887 1128
a
9 pure MeCN 2943 2263
b
10 MeCN + H2O 2945 2257
a
11 pure acetone 2921 1708
b
12 acetone + H2O 2928 1698
a
Raman vibration modes of pure solvents were obtained from the spectral database for organic compounds at the National Institute of Advanced
Industrial Science and Technology (AIST) (https://sdbs.db.aist.go.jp/sdbs/cgi-bin/direct_frame_top.cgi). bRaman vibration modes of cosolvents
(volume ratios of the organic solvent to H2O are 89:11) were obtained from Figure 3.

to Ti-MWW, indicating that the hydrophobicity of titanosili- introduction of additional H2O molecules did not result in a
cates has no obvious effect on propylene epoxidation in H2O further noticeable change.34 According to a recent study, the
solvent, in line with the early report in the literature.49 reorganization of confined H2O structures could improve
Therefore, diffusion and adsorption/desorption are not the alkene epoxidation rates by up to 400-fold.52 Motivated by
determining factors on the catalytic performance of Ti-MWW these reports, we surmise that the activity in propylene
in the H2O-solvent propylene epoxidation. epoxidation in H2O solvent depends on the hydrogen-bond
3.2.2. Influence of H2O Structures on Propylene Epox- structures of H2O molecules and there is probably a proper
idation. Many theoretical results showed that the H2O2 H2O structure for propylene epoxidation.
activation process was regarded as the rate-determining step The structures of H2O molecules are diverse in the complex
of the global epoxidation reaction. 27,32,50,51 On H 2 O 2 network of HBs. Experimental infrared52,53 and Raman53,54
activation to form Ti−OOH species, the proton transfer spectroscopies and theoretical calculation55,56 demonstrated
process occurred inevitably but with high activation that H2O molecules have five kinds of hydrogen-bonding
energy.32,50,51 A protic molecule (ROH or H2O), which structures (Scheme S3), which include “free” H2O that lacks
could donate and accept a proton, could be used as the HBs (3628 cm−1 for Raman and 3636 cm−1 for IR), “DDA”
mediator to facilitate proton transfer on H2O2 activation, while H2O that donates two and accepts one HB (3511 cm−1 for
an aprotic molecule could not because it could only accept a Raman and 3570 cm−1 for IR), “DA” H2O that donates one
proton.32,33 When the solvent is pure H2O, the possible and accepts one HB (3393 cm−1 for Raman and 3430 cm−1 for
proton-transfer mediator is only protic H2O molecules, and its IR), “DDAA” H2O that both donates and accepts two HBs
effect is shown in Scheme 1. Many reports indicated that the (3233 cm−1 for Raman and 3220 cm−1 for IR), and “DAA”
structure of H2O molecules had an important impact on H2O2 H2O that donates one and accepts two HBs (3051 cm−1 for
activation.33,34,52 Sever and root proposed that the presence of Raman and 3040 cm−1 for IR). To investigate the influence of
a single H2O molecule lowered the activation barrier for the H2O structures on the activity in propylene epoxidation, we
Ti−OOH formation by 4.1 kcal/mol, while hydrogen-bonded modified H2O structures by introducing organic solvents and
H2O molecules could lower it by 5.0 kcal/mol.33 Eckert et al. ions.
suggested that the introduction of H2O molecules reduced the 3.2.2.1. Influence of Organic Solvents. The introduction of
reaction barrier to 25.7 kcal/mol in the case of a single an organic solvent could break the HB network of H2O
molecule with almost unchanged reaction energy, whereas the molecules through intermolecular interactions to change H2O
11645 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Scheme 2. Proposed Mechanism of Breaking the Hydrogen Bond of H2O with the Introduction of Protic ROH

Scheme 3. Proposed Mechanism of Breaking the Hydrogen Bond of H2O with the Introduction of Aprotic MeCN or Acetone

structures.57−59 A certain amount of the organic solvent used of H2O molecules is broken.59 The Raman spectra were
as mentioned in Section 3.1 was introduced into the H2O deconvoluted via a Gaussian peak fitting method into five
solvent (the volume ratios of the organic solvent to H2O are peaks,53,54 and the corresponding deconvolution bands at
89:11) for simulating the H2O structures in the reaction 3051, 3233, 3393, 3511, and 3628 cm−1 were attributed to
system. The H2O structures were studied by Raman spectros- DAA, DDAA, DA, DDA, and free H2O, respectively (Figure
copy. The band intensity of the O−H group of protic alcohols 2B−H). Within pure H2O, three kinds of H2O structures
with a strong polarity is so low that it can be ignored. In (35.9% DDAA at 3233 cm−1, 27.8% DA at 3393 cm−1, and
addition, aprotic nitriles and ketones have no characteristic 31.7% DDA at 3511 cm−1) dominated with a small percentage
peak in the O−H region. Thus, the Raman spectra in the O−H of DAA (2.6%) at 3051 cm−1 and free H2O (1.9%) at 3628
region at 3000−4000 cm−1 can be applied to identify the H2O cm−1 (Table 2, entry 1).
structures. As shown in Figure 2A, the Raman band in the O− After introducing the organic solvent, the proportions of
H region of H2O gives a pronounced blue shift after the DAA and DDAA decrease along with the increase of DA and
introduction of various organic solvents, implying that the HB DDA. However, the proportion of free H2O remains almost
11646 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 4. IR spectra (A) and the deconvolution band (B−H) at 3040 cm−1 (DAA), 3220 cm−1 (DDAA), 3430 cm−1 (DA), 3570 cm−1 (DDA), and
3636 cm−1 (free) in the O−H stretching of H2O in the cosolvents (pure H2O, MeOH + H2O, EtOH + H2O, 1-PrOH + H2O, i-PrOH + H2O,
MeCN + H2O, and acetone + H2O). The corresponding proportion of various H2O structures (DDAA, DA, DDA, and free) in IR spectra (I) is
plotted against the PO yield in propylene epoxidation catalyzed by Ti-MWW in various solvents (Table 1). The volume ratios of the organic
solvent to H2O are 89:11.

unchanged (Table 2, entries 2−7), implying that DAA and propylene epoxidation (Table S2). It implies that the
DDAA were transformed into DA and DDA. Especially, the generated peroxide intermediate is not responsible for the
aprotic MeCN decreases the proportions of DAA and DDAA fast reaction with MeCN.
from 2.6 to 0% and from 35.9 to 14.1%, respectively, and To understand the interaction of organic solvents with H2O
increases the proportions of DA and DDA from 27.8 to 37.3% molecules, the Raman vibration modes (γC−O, γC−H, γC�N, and
and from 31.7 to 46.0%, respectively (Table 2, entry 6). The γC=O) of organic solvents were studied (Figure 3), and the
corresponding proportions of various H2O structures versus corresponding characteristic peak positions are shown in Table
the epoxidation rate were plotted (Figure 2I). Interestingly, 3. Compared to pure protic alcohol solvents, γC−O in alcohol−
PO yields have a positive relationship with the proportions of H2O cosolvents gives a relatively pronounced red shift, and
DDA but not with the proportion of the other H2O structures. γC−H gives a blue shift (Table 3, entries 1−8). The shifts of
Therefore, DDA is considered the superior H2O structure for γC−O and γC−H are derived from the decreased electron density
propylene epoxidation. The aprotic MeCN presents the around the C−O group and an increased electron density
highest PO yield with the highest DDA proportion (46.0% around the C−H group, respectively, which indicates that the
in Table 2, entry 6), which rationalizes that MeCN is the best HB occurs between the O atom of ROH and the O−H group
solvent for Ti-MWW in alkene epoxidations.14,15,60−64 It is well of the H2O molecule, namely, the H2O molecule acts as a
known that MeCN could react with H2O2 to form the peroxide proton donor and the ROH molecule acts as a proton acceptor
intermediate.65 This intermediate could be more effective for (like the blue ROH-H2O structure in Scheme 2). Combining
olefin epoxidations. However, as the solvent is H2O, Raman and infrared spectroscopic study, Nedić et al. suggested
introducing a small amount of MeCN into the epoxidation that H2O played the role of the HB donor, not because it is a
reaction cannot improve the catalytic performance of better HB donor than ROH, but because it is a much inferior
11647 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 5. IR spectra in the hydroxyl stretching of H2O in the different MeCN concentrations (A) and the corresponding deconvolution bands at
3040 cm−1 (DAA), 3220 cm−1 (DDAA), 3430 cm−1 (DA), 3570 cm−1 (DDA), and 3636 cm−1 (free). MeCN concentrations are 0 (B), 25 (C), 50
(D), and 89% (E), respectively.

Table 4. Proportions of Various H2O Structures in DIfferent MeCN Concentrations from IR Spectra and the Corresponding
Propylene Epoxidation Catalyzed by Ti-MWW
a a a a b c c c c c
PO yield H2O2 conv. PO sel. H2O2 eff. TON DAA DDAA DA DDA free
entry solvent (%) (%) (%) (%) value (%) (%) (%) (%) (%)
1 pure H2O 38.2 49.1 95.3 77.8 358 5.5 56.2 31.9 5.5 0.9
2 25% MeCN + 75% H2O 46.9 58.2 100 80.6 439 0 46.1 44.0 9.4 0.5
3 50% MeCN + 50% H2O 73.8 86.5 100 85.3 691 0 31.8 52.6 14.8 0.7
4 89% MeCN + 11% H2O 88.7 99.5 100 89.1 830 0 18.1 53.5 26.9 1.5
a
Propylene epoxidation conditions: solvent (pure H2O or MeCN−H2O cosolvent) 20 mL, others were the same as Table 1. bTurnover number per
Ti site for the PO yield. cProportions of various H2O structures were obtained from the deconvolution results in IR spectra (Figure 5).

Figure 6. Optimized structures of free H2O, DA, and DDA. The color codes are small white spheres, H atoms and red spheres, O atoms.

HB acceptor.66 Compared to the pure aprotic MeCN solvent, and the O−H group of the H2O molecule (like the blue
γC�N in MeCN−H2O cosolvent gives a relatively pronounced acetone−H2O structure in Scheme 3) leads to the blue shift of
red shift, and γC−H gives a blue shift (Table 3, entries 9 and γC−H and the red shift of γC=O (Table 3, entries 11 and 12).
10), which indicates that a HB occurs between the N atom of Based on the above results and analyses, we proposed the
MeCN and the O−H group of the H2O molecule, namely, the mechanisms of breaking the HB of H2O with the introduction
H2O molecule acts as a proton donor and MeCN act as a of organic solvents (protic solvents: ROH; aprotic solvents:
proton acceptor (like the blue MeCN−H2O structure in MeCN and acetone). Introducing protic ROH (Scheme 2),
Scheme 3). Similarly, the HB between the O atom of acetone which could donate and accept a proton, would promote the
11648 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Table 5. Charge Distribution of Free H2O, DA, and DDA Table 6. Propylene Epoxidation and H2O2 Activation over
a Ti-MWW in H2O Solvents with the Additive of Alkali Metal
atomic charge
Ions
entry structure of H2O H1 H2 O
a a a c
PO H2O2 PO H2O2
1 free 0.342 0.342 −0.684 additive yield conv. sel. a
H2O2 b
TON conv.
2 DA 0.346 0.381 −0.721 entry (mol/L) (%) (%) (%) eff. (%) value (%)
3 DDA 0.356 0.357 −0.782 1 38.2 49.1 95.3 77.8 358 3.9
a
Optimized structural parameters of the models of the H2O structures 2 Li+-0.27 80.2 94.6 100 84.8 751 10.1
calculated at the B3LYP. 3 Na+- 83.8 98.6 100 85.0 784 13.5
0.27
4 K+-0.27 89.5 98.6 100 90.8 838 14.2
transformation of DAA and DDAA into DA and DDA, 5 Cs+-0.27 92.3 99.5 100 92.8 864 19.5
respectively, and DDA can be further transformed to DA, 6 K+-0.13 83.6 94.1 100 88.8 782 6.9
evidenced by the increased proportion of DA (Table 2, entries 7 K+-0.40 59.8 75.9 100 78.8 560 26.5
2−5). Introducing aprotic solvents (MeCN or acetone) 8 K+-0.54 40.7 56.4 100 72.2 381 41.5
(Scheme 3), which could only accept a proton, would a
Propylene epoxidation conditions: additive (Li+, Na+, K+, and Cs+)
transform DAA and DDAA into DA and DDA, respectively. 0.13−0.54 mol/L, others were the same as in Table 1. bTurnover
However, it is difficult to further transform DDA into DA, number per Ti site for the PO yield. cH2O2 activation conditions: Ti-
evidenced by the increased proportion of DDA (Table 2, MWW 100 mg, H2O 10 mL, H2O2 10 mmol, temp 333 K, time 120
entries 6 and 7). Therefore, it is reasonable that introducing min, additive (Li+, Na+, K+, and Cs+) 0.13−0.54 mol/L.
aprotic nitriles and ketones mainly generates DDA. So Ti-
MWW presents a superior catalytic performance in aprotic
solvents due to the high proportion of DDA for the proton Besides, the influences of MeCN concentrations on H2O
transfer on H2O2 activation. structures and the rate of propylene epoxidation of Ti-MWW
Likewise, the result of IR spectra also indicates that were investigated. Figure 5 shows that increasing the MeCN
introducing an organic solvent causes the transformation concentration (from 0 to 89%) reinforces the blue shift of IR
from DAA and DDAA to DA and DDA by changing the HB of spectra in the O−H region of H2O and the proportion of DDA
H2O molecules (Figure 4 and Table 2), respectively. The increases gradually from 5.5 to 26.9%, accompanied by the
positive correlation (Figure 4I) between the proportion of increase of PO yield from 38.2 to 88.7% and the TON value
DDA and the catalytic performance of propylene epoxidation from 358 to 830 (Table 4). This further verifies the decisive
is in line with the results of Raman spectra (Figure 2I). role of DDA H2O on propylene epoxidation.

Figure 7. Raman spectra (A), difference in Raman spectra after subtraction of pure H2O (B) and the deconvolution bands (C−F) at 3051 cm−1
(DAA), 3233 cm−1 (DDAA), 3393 cm−1 (DA), and 3511 cm−1 (DDA) in the O−H stretching of H2O with the addition of different alkali metal
ions.

11649 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 8. Raman spectra (A), the difference in Raman spectra after subtraction of pure H2O (B) and the deconvolution band (C−F) at 3051 cm−1
(DAA), 3233 cm−1 (DDAA), 3393 cm−1 (DA), and 3511 cm−1 (DDA) in the O−H stretching of H2O with the addition of different K+ contents.

3.2.2.2. Influence of Ions. Ions with high charge densities

could cause the rearrangement of H2O molecules through the
strong polarization effect and change H2O structures.67 We
investigated the influence of K+ on H2O structures through the
characterization of Raman spectra of H2O in the O−H
stretching region. Figure 7A shows the Raman spectra of H2O
with the introduction of K+, and the difference in Raman
spectra after subtraction of pure H2O is shown in Figure 7B.
Obviously, after introducing K+, the band at 3000−3300 cm−1
is negative, and the band at 3300−3600 cm−1 is positive
(Figure 7B), indicating the transformation of the O−H
stretching region68 from 3000−3300 to 3300−3600 cm−1.
This transformation implies the breaking of the HB of H2O
Figure 9. Influence of the additive K+ on alkene epoxidation.
molecules through the strong polarization effect of K+, so Ti-
Propylene epoxidation conditions: Ti-MWW 100 mg, solvent 20 mL,
propylene pressure 0.4 MPa, H2O2 (30 wt %) 30 mmol, temp 323 K, MWW presents the improved rates of H2O2 activation and
time 50 min, K+ content (if added) 0.27 mol/L; other alkene propylene epoxidation in H2O solvent (Table 6, entry 4). To
epoxidation conditions: catalyst, 50 mg, solvent 10 mL, 1-hexene/1- study the correlation between the catalytic performance and
pentene 10 mmol, H2O2 10 mmol, temp 333 K, time 2 h, K+ content H2O structures, the spectra in Figure 7B were deconvoluted via
(if added) 0.27 mol/L. Detailed data are shown in Table S3. a Gaussian peak fitting method. The deconvolution results
(Figure 7E) point out that the negative bands at 3051 cm−1
(DAA) and 3233 cm−1 (DDAA) together with the positive
As shown in Scheme 1, the mediator H2O facilitates proton bands at 3393 cm−1 (DA) and 3511 cm−1 (DDA), indicate the
transfer by accepting a proton in H2O2 and then donating a transformation of the H2O structure from DAA and DDAA to
proton in H2O itself. It implies that DDAA and DAA cannot DA and DDA, respectively. Especially, the transformation from
effectively facilitate the proton transfer because they have no DDAA to DDA dominates. Based on the conclusion in Section
ability to accept an additional proton. DFT calculation was 3.2.2.1, the increased proportion of DDA contributes to H2O2
applied to investigate the ability of free, DA, and DDA H2O to activation and propylene epoxidation. As a result, the rates of
accept a proton (Figure 6). The charge distributions of these H2O2 activation and propylene epoxidation were improved
three H2O structures are listed in Table 5. It is convincing that with the introduction of K+ (Table 6, entry 4).
DDA possesses the highest charge density of the O atom in Other alkali metal ions (Li+, Na+, and Cs+) were introduced
favor of accepting a proton, which is, therefore, identified as to change H2O structures. As the size of the alkali metal ions
the superior H2O structure for proton transfer on H2O2 increases, the transformation of the O−H stretching region
activation. from 3000−3300 to 3300−3600 cm−1 in Raman spectra is
11650 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Table 7. Epoxidation of Propylene over Ti-MWW-xh (x Represents the Acid Treatment Time of the Ti-MWW Lamellar
Precursor37) and TS-1 Zeolites in H2O Solvent with or without the Additive of K+
a a b b b b c
entry catalyst Si/Ti Si/B additive PO yield (%) H2O2 conv. (%) PO sel. (%) H2O2 eff. (%) TON value
1 Ti-MWW-2 h 52 36 23.4 41.3 100 56.7 219
2 K+-0.27 mmol/L 63.0 80.1 100 78.7 590
3 Ti-MWW-8 h 52 41 31.0 45.2 100 68.6 290
4 K+-0.27 mmol/L 74.1 89.6 100 82.7 694
5 Ti-MWW-12 h 52 43 34.2 47.3 100 72.3 320
6 K+-0.27 mmol/L 83.6 94.7 100 88.3 782
7 Ti-MWW-20 h 52 46 38.2 49.1 95.3 77.8 358
8 K+-0.27 mmol/L 89.5 98.6 100 90.8 838
9 TS-1 76 11.8 14.1 100 83.0 161
10 K+-0.27 mmol/L 27.2 32.6 100 83.4 372
a b
Determined by ICP. Propylene epoxidation conditions: catalyst (Ti-MWW-xh or TS-1) 100 mg, K+ content (if added) 0.27 mol/L, others were
the same as in Table 1. cTurnover number per Ti site for the PO yield.

promoted (Figure 7B), which indicates the intensified breaking designed a series of Ti-MWW-xh (x = 2, 8, 12, and 20)
of the HB of H2O molecules. The deconvolution results catalysts that contained various amounts of Ti(OSi)3OH
(Figure 7C−F) point out the increased DDA peak area in the species along with a constant MWW topology and total Ti
order of Li+ < Na+ < K+ < Cs+, leading to the gradual contents, based on our recent work.37 This series of Ti-MWW-
improvement of H2O2 activation and propylene epoxidation xh catalysts were applied to propylene epoxidation in H2O
(Table 6, entries 2−5). solvent. Our recent work suggested that more Ti(OSi)4 species
The influences of the K+ content on the H2O structures and were transformed into the Ti(OSi)3OH species by increasing
the catalytic performance of Ti-MWW were further inves- the acid-treated time of the Ti-MWW lamellar precursor.37
tigated. As shown in Figure 8B, as the K+ content increases, the The increase of Ti(OSi)3OH species leads to the incremental
transformation of the O−H stretching region from 3000−3300 PO yield and TON value (Table 7, entries 1, 3, 5, and 7).
to 3300−3600 cm−1 in Raman spectra is promoted, indicating Upon addition of K+, better catalytic performances in
the intensified breaking of the HB of H2O molecules. The propylene epoxidation are observed for all Ti-MWW samples
deconvolution results show that the DDA peak area increases (Table 7, entries 1−8). In the pure H2O solvent, the possible
(Figure 8C−F), promoting the H2O2 activation ability from proton-transfer mediator is only H2O molecules whether the
3.9 to 41.5% (Table 6, entries 1, 4, 6−8). However, the PO Ti active site is Ti(OSi)4 or Ti(OSi)3OH species. Thus, the
yield and TON value in propylene epoxidation first increase transformation of the H2O structure especially into DDA
and then decrease (Table 6, entries 4, 6−8). Many studies enhances H2O 2 activation and propylene epoxidation.
reported that the excessive K+ could lead to the poisoning of Especially, Ti-MWW-20h with the biggest densities of
titanosilicates through the exchange of Ti−OH to Ti−OK,45,69 Ti(OSi)3OH species presents the highest PO yield of 89.5%
and the latter had no ability in H2O2 activation. It makes clear because of the strong HB interaction between the H2O
that the poisoning of titanosilicates when introducing excessive molecule and Ti(OSi)3OH site in Scheme 1, Pathway II.
K+ is not responsible for the decreased PO yield. Therefore, Therefore, a proper Ti active site (e.g., the Ti(OSi)3OH site) is
the decreased PO yield with the introduction of excessive K+ is necessary to achieve a highly efficient propylene epoxidation.
due to the fact that the transfer of active oxygen to propylene Besides, TS-1 with the MFI topology structure was used as the
cannot match the accelerated H2O2 activation,70 decreasing catalyst for H2O-solvent propylene epoxidation. When the
H2O2 utilization efficiency (Table 6, entries 7 and 8). solvent is pure H2O, TS-1 presents a low PO yield of 11.8%
Furthermore, different alkenes were also used as reactants and a TON value of 161 (Table 7, entry 9). By introducing a
for the epoxidation reaction. With the introduction of K+, the certain amount of K+ for increasing the DDA proportion, the
catalytic activity of Ti-MWW for all alkene epoxidations in PO yield and TON value of propylene epoxidation over TS-1
H2O is greatly improved, exceeding that in MeCN (Figure 9 increase to 27.2% and 372, respectively (Table 7, entry 10),
and Table S3). With the growth of carbon chains from indicating the universal role of K+ in improving the catalytic
propylene to 1-hexene, the selectivity of the epoxide slightly performance of propylene epoxidation.
increases from 95.3 to 98.9% (Table S3). It is because the
stronger electron-donating ability of the alkyl group in alkenes 4. CONCLUSIONS
decreases the C−O polarization of epoxide. After the In this work, we proposed a highly efficient H2O-solvent alkene
introduction of K+, the acid sites were suppressed and PO epoxidation system (e.g., Ti-MWW/H2O/H2O2/propylene/
selectivity increased to 100% for various alkenes. These results K+) for the first time, which possesses high activity, selectivity,
showcase the suitability of this highly efficient H2O-solvent and stability. A positive correlation between the proportion of
system for short-chain alkene epoxidations. For longer and the DDA H2O structure and the catalytic performance of
more complex alkenes (e.g., cyclopentene,63 cyclohexene,71 2, propylene epoxidation was observed. It is revealed that the
5-dihydrofuran,72 and 1-octene),73 it may lead to the formation catalytic performance of propylene epoxidation in H2O solvent
of a broader series of side products (e.g., allylic oxidation depends on the structures of H2O molecules, and the DDA
byproducts) and so a low selectivity of epoxides. H2O structure with a strong ability to accept an additional
3.2.5. Influence of Ti Active Sites on Propylene proton is the superior proton-transfer mediator on H2O2
Epoxidation. To explore the influence of Ti active sites (e.g., activation. Introducing organic solvents and alkali metal ions
Ti(OSi)4 vs Ti(OSi)3OH sites) on propylene epoxidation, we can change the H2O structures by breaking the HBs of H2O
11651 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

molecules and thus affects the rate of propylene epoxidation.

The aprotic MeCN exhibits the highest rate of propylene
■ ACKNOWLEDGMENTS
This work was supported by the National Natural Science
epoxidation due to the highest DDA proportion in the reaction Foundation of China (22072044 and 21673076) and the
system, so it is the best organic solvent for Ti-MWW. By Research Funds of Happiness Flower ECNU (2020ST2203).
introducing alkali metal ions (e.g., K+) to increase the DDA
proportion, the catalytic activity for alkene epoxidations in
H2O solvent was greatly improved, even exceeding that in the
best solvent MeCN. The excellent catalytic performance for
■ REFERENCES
(1) Curzons, A. D.; Constable, D. C.; Cunningham, V. L. Solvent
H2O-solvent propylene epoxidation catalyzed by Ti-MWW selection guide: a guide to the integration of environmental, health
was thus achieved. Overall, the current report highlights the and safety criteria into the selection of solvents. Clean Prod. Processes
1999, 1, 82−90.
suitability of this highly efficient H2O-solvent system for alkene (2) Henderson, R. K.; Jiménez-González, C.; Constable, D. J.;
epoxidations. Alston, S. R.; Inglis, G. G.; Fisher, G.; Sherwood, J.; Binks, S. P.;

■ ASSOCIATED CONTENT
* Supporting Information
sı
Curzons, A. D. Expanding GSK’s solvent selection guide−embedding
sustainability into solvent selection starting at medicinal chemistry.
Green Chem. 2011, 13, 854−862.
(3) Prat, D.; Pardigon, O.; Flemming, H.-W.; Letestu, S.; Ducandas,
The Supporting Information is available free of charge at
V.; Isnard, P.; Guntrum, E.; Senac, T.; Ruisseau, S.; Cruciani, P.;
https://pubs.acs.org/doi/10.1021/acssuschemeng.2c03524. Hosek, P. Sanofi’s solvent selection guide: A step toward more
XRD patterns; UV−visible spectra; IR spectra; 29Si MAS sustainable processes. Org. Process Res. Dev. 2013, 17, 1517−1525.
NMR spectra; SEM images; BET results; and ICP (4) Capello, C.; Fischer, U.; Hungerbühler, K. What is a green
results (PDF) solvent? A comprehensive framework for the environmental assess-
ment of solvents. Green Chem. 2007, 9, 927−934.

■ AUTHOR INFORMATION
Corresponding Author
(5) Sheldon, R. A. Green solvents for sustainable organic synthesis:
state of the art. Green Chem. 2005, 7, 267−278.
(6) Li, C.-J.; Chen, L. Organic chemistry in water. Chem. Soc. Rev.
2006, 35, 68−82.
Yueming Liu − Shanghai Key Laboratory of Green Chemistry (7) Zhou, F.; Hearne, Z.; Li, C.-J. Water�the greenest solvent
and Chemical Processes, School of Chemistry and Molecular overall. Curr. Opin. Green Sustainable Chem. 2019, 18, 118−123.
Engineering, East China Normal University, Shanghai (8) Lubineau, A.; Augé, J. Water as Solvent in Organic Synthesis. In
200062, China; Institute of Eco-Chongming, Shanghai Topics in Current Chemistry; Knochel, P., Ed.; Springer: Heidelberg,
202162, China; orcid.org/0000-0001-9640-6301; 1999 DOI: 10.1007/3-540-48664-X_1.
Email: ymliu@chem.ecnu.edu.cn (9) Oyama, S. T. Rates, Kinetics, and Mechanisms of Epoxidation:
Homogeneous, Heterogeneous, and Biological Routes. In Mechanisms
Authors in Homogeneous and Heterogeneous Epoxidation Catalysis; Elsevier,
Yunkai Yu − Shanghai Key Laboratory of Green Chemistry 2008; pp 3−99 DOI: 10.1016/B978-0-444-53188-9.00001-8.
and Chemical Processes, School of Chemistry and Molecular (10) Ratnasamy, P.; Srinivas, D.; Knozinger, H. Active sites and
reactive intermediates in titanium silicate molecular sieves. Adv. Catal.
Engineering, East China Normal University, Shanghai
2004, 48, 1−169.
200062, China (11) He, M.; Sun, Y.; Han, B. Green carbon science: scientific basis
Nan Fang − Shanghai Key Laboratory of Green Chemistry for integrating carbon resource processing, utilization, and recycling.
and Chemical Processes, School of Chemistry and Molecular Angew. Chem., Int. Ed. 2013, 52, 9620−9633.
Engineering, East China Normal University, Shanghai (12) He, M.; Sun, Y.; Han, B. Green Carbon Science: Efficient
200062, China Carbon Resource Processing, Utilization, and Recycling towards
Zhen Chen − Shanghai Key Laboratory of Green Chemistry Carbon Neutrality. Angew. Chem., Int. Ed. 2022, 61, No. e202112835.
and Chemical Processes, School of Chemistry and Molecular (13) Bellussi, G.; Carati, A.; Clerici, M. G.; Maddinelli, G.; Millini, R.
Engineering, East China Normal University, Shanghai Reactions of titanium silicalite with protic molecules and hydrogen
200062, China peroxide. J. Catal. 1992, 133, 220−230.
(14) Wu, P.; Tatsumi, T.; Komatsu, T.; Yashima, T. A novel
Dongxu Liu − Shanghai Key Laboratory of Green Chemistry
titanosilicate with MWW structure: II. Catalytic properties in the
and Chemical Processes, School of Chemistry and Molecular selective oxidation of alkenes. J. Catal. 2001, 202, 245−255.
Engineering, East China Normal University, Shanghai (15) Fan, W.; Wu, P.; Tatsumi, T. Unique solvent effect of
200062, China microporous crystalline titanosilicates in the oxidation of 1-hexene
Mingyuan He − Shanghai Key Laboratory of Green Chemistry and cyclohexene. J. Catal. 2008, 256, 62−73.
and Chemical Processes, School of Chemistry and Molecular (16) Corma, A.; Esteve, P.; Martinez, A. Solvent effects during the
Engineering, East China Normal University, Shanghai oxidation of olefins and alcohols with hydrogen peroxide on Ti-beta
200062, China; Institute of Eco-Chongming, Shanghai catalyst: The influence of the hydrophilicity-hydrophobicity of the
202162, China zeolite. J. Catal. 1996, 161, 11−19.
(17) Wang, G.; Li, Y.; Zhu, Q.; Li, G.; Zhang, C.; Guo, H. Influence
Complete contact information is available at: of impurities in a methanol solvent on the epoxidation of propylene
https://pubs.acs.org/10.1021/acssuschemeng.2c03524 with hydrogen peroxide over titanium silicalite-1. Catalysts 2020, 10,
15.
Author Contributions (18) Corma, A.; García, H. Lewis acids as catalysts in oxidation
The manuscript was written through contributions of all reactions: from homogeneous to heterogeneous systems. Chem. Rev.
authors. All authors have given approval to the final version of 2002, 102, 3837−3892.
the manuscript. (19) Thiele, G. Process for the production of epoxides from olefins.
US5,675,026, 1997.
Notes (20) Wu, L.; Zhao, S.; Lin, L.; Fang, X.; Liu, Y.; He, M. In-depth
The authors declare no competing financial interest. understanding of acid catalysis of solvolysis of propene oxide over

11652 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

titanosilicates and titanosilicate/H2O2 systems. J. Catal. 2016, 337, (38) Wang, L.; Liu, Y.; Xie, W.; Wu, H.; Li, X.; He, M.; Wu, P.
248−259. Improving the hydrophobicity and oxidation activity of Ti−MWW by
(21) Wang, G.; Li, Y.; Lin, L.; Guo, Y.; et al. Liquid-phase reversible structural rearrangement. J. Phys. Chem. C 2008, 112,
epoxidation of propylene with H2O2 over TS-1 zeolite: Impurity 6132−6138.
formation and inhibition study. Ind. Eng. Chem. Res. 2021, 60, 12109− (39) Deng, X.; Wang, Y.; Shen, L.; Wu, H.; Liu, Y.; He, M. Low-cost
12122. synthesis of titanium silicalite-1 (TS-1) with highly catalytic oxidation
(22) Kholdeeva, O. A. Hydrogen peroxide activation over Ti(IV): performance through a controlled hydrolysis process. Ind. Eng. Chem.
what have we learned from studies on Ti-containing polyoxometa- Res. 2013, 52, 1190−1196.
lates? Eur. J. Inorg. Chem. 2013, 2013, 1595−1605. (40) Payne, G. B.; Smith, C. W. Reactions of Hydrogen Peroxide.
(23) Antonova, N. S.; Carbo, J. J.; Kortz, U.; Kholdeeva, O. A.; III. Tungstic Acid Catalyzed Hydroxylation of Cyclohexene in
Poblet, J. M. Mechanistic insights into alkene epoxidation with H2O2 Nonaqueous Media. J. Org. Chem. 1957, 22, 1682−1685.
by Ti-and other TM-containing polyoxometalates: Role of the metal (41) Serrano, D.; Sanz, R.; Pizarro, P.; Moreno, I. Turning TS-1
nature and coordination environment. J. Am. Chem. Soc. 2010, 132, zeolite into a highly active catalyst for olefin epoxidation with organic
7488−7497. hydroperoxides. Chem. Commun. 2009, 1407−1409.
(24) Russo, V.; Tesser, R.; Santacesaria, E.; Di Serio, M. Kinetics of (42) Khouw, C. B.; Davis, M. E. Catalytic activity of titanium
propene oxide production via hydrogen peroxide with TS-1. Ind. Eng. silicates synthesized in the presence of alkali-metal and alkaline-earth
Chem. Res. 2014, 53, 6274−6287. ions. J. Catal. 1995, 151, 77−86.
(25) Clerici, M. G.; Ingallina, P. Epoxidation of lower olefins with (43) Shetti, V. N.; Srinivas, D.; Ratnasamy, P. Enhancement of
hydrogen-peroxide and titanium silicalite. J. Catal. 1993, 140, 71−83. chemoselectivity in epoxidation reactions over TS-1 catalysts by alkali
(26) Nie, X.; Ji, X.; Chen, Y.; Guo, X.; Song, C. Mechanistic and alkaline metal ions. J. Mol. Catal. A: Chem. 2004, 210, 171−178.
investigation of propylene epoxidation with H2O2 over TS-1: Active (44) Wang, Y.; Liu, Y.; Li, X.; Wu, H.; He, M.; Wu, P.
site formation, intermediate identification, and oxygen transfer Intermolecular condensation of ethylenediamine to 1,4-diazabicyclo-
pathway. Mol. Catal. 2017, 441, 150−167. [2,2,2]octane over TS-1 catalysts. J. Catal. 2009, 266, 258−267.
(27) Wells, D. H.; Delgass, W. N.; Thomson, K. T. Evidence of (45) Yu, Y.; Tang, Z.; Liu, W.; Wang, J.; Chen, Z.; Shen, K.; Wang,
defect-promoted reactivity for epoxidation of propylene in titanosi- R.; Liu, H.; Huang, X.; Liu, Y. Enhanced catalytic oxidation
licate (TS-1) catalysts: A DFT study. J. Am. Chem. Soc. 2004, 126, performance of K+-modified Ti-MWW through selective breaking of
2956−2962. interfacial hydrogen-bonding interactions of H2O2. Appl. Catal., A
(28) Panyaburapa, W.; Nanok, T.; Limtrakul, J. Epoxidation reaction 2019, 587, No. 117270.
of unsaturated hydrocarbons with H2O2 over defect TS-1 Investigated (46) Ramachandran, C. E.; Zhao, Q.; Zikanova, A.; Kocirik, M.;
by ONIOM method: Formation of active sites and reaction Broadbelt, L. J.; Snurr, R. Q. Is 1-hexene epoxidation in TS-1 diffusion
mechanisms. J. Phys. Chem. C 2007, 111, 3433−3441. limited in different solvents? Catal. Commun. 2006, 7, 936−940.
(29) Prestipino, C.; Bonino, F.; Usseglio, S.; Damin, A.; Tasso, A.; (47) Drago, R. S.; Dias, S. C.; McGilvray, J. M.; Mateus, A. L. M. L.
Clerici, M. G.; Bordiga, S.; D′Acapito, F.; Zecchina, A.; Lamberti, C. Acidity and hydrophobicity of TS-1. J. Phys. Chem. B 1998, 102,
Equilibria between peroxo and hydroperoxo species in the 1508−1514.
titanosilicates: An in situ high-resolution XANES investigation. (48) Na, K.; Jo, C.; Kim, J.; Ahn, W.-S.; Ryoo, R. MFI titanosilicate
ChemPhysChem 2004, 5, 1799−1804. nanosheets with single-unit-cell thickness as an oxidation catalyst
(30) Bregante, D. T.; Johnson, A. M.; Patel, A. Y.; Ayla, E. Z.; using peroxides. ACS Catal. 2011, 1, 901−907.
Cordon, M. J.; Bukowski, B. C.; Greeley, J.; Gounder, R.; Flaherty, D. (49) Ramachandran, C. E.; Du, H.; Kim, Y. J.; Kung, M. C.; Snurr,
W. Cooperative effects between hydrophilic pores and solvents: R. Q.; Broadbelt, L. J. Solvent effects in the epoxidation reaction of 1-
catalytic consequences of hydrogen bonding on alkene epoxidation in hexene with titanium silicalite-1 catalyst. J. Catal. 2008, 253, 148−
zeolites. J. Am. Chem. Soc. 2019, 141, 7302−7319. 158.
(31) Sever, R. R.; Root, T. W. DFT study of solvent coordination (50) Wells, D. H., Jr.; Joshi, A. M.; Delgass, W. N.; Thomson, K. T.
effects on titanium-based epoxidation catalysts. Part Two: Reactivity A quantum chemical study of comparison of various propylene
of titanium hydroperoxo complexes in ethylene epoxidation. J. Phys. epoxidation mechanisms using H2O2 and TS-1 Catalyst. J. Phys. Chem.
Chem. B 2003, 107, 4090−4099. B 2006, 110, 14627−14639.
(32) Nie, X.; Ren, X.; Ji, X.; Chen, Y.; Janik, M. J.; Guo, X.; Song, C. (51) Sinclair, P. E.; Catlow, C. R. A. Quantum chemical study of the
Mechanistic insight into propylene epoxidation with H2O2 over mechanism of partial oxidation reactivity in titanosilicate catalysts:
titanium silicalite-1: Effects of zeolite confinement and solvent. J. Phys. active site formation, oxygen transfer, and catalyst deactivation. J.
Chem. B 2019, 123, 7410−7423. Phys. Chem. B 1999, 103, 1084−1095.
(33) Sever, R. R.; Root, T. W. DFT study of solvent coordination (52) Bregante, D. T.; Chan, M. C.; Tan, J. Z.; Ayla, E. Z.; Nicholas,
effects on titanium-based epoxidation catalysts. Part one: Formation C. P.; Shukla, D.; Flaherty, D. W. The shape of water in zeolites and
of the titanium hydroperoxo intermediate. J. Phys. Chem. B 2003, 107, its impact on epoxidation catalysis. Nat. Catal. 2021, 4, 797−808.
4080−4089. (53) Bakker, H. J.; Skinner, J. Vibrational spectroscopy as a probe of
(34) Stare, J.; Henson, N. J.; Eckert, J. Mechanistic aspects of structure and dynamics in liquid water. Chem. Rev. 2010, 110, 1498−
propene epoxidation by hydrogen peroxide. Catalytic role of water 1517.
molecules, external electric field, and zeolite framework of TS-1. J. (54) Sun, Q. The Raman OH stretching bands of liquid water. Vib.
Chem. Inf. Model. 2009, 49, 833−846. Spectrosc. 2009, 51, 213−217.
(35) Huang, J.; Akita, T.; Faye, J.; Fujitani, T.; Takei, T.; Haruta, M. (55) Sadlej, J. Theoretical study of structure and spectra of cage
Propene epoxidation with dioxygen catalyzed by gold clusters. Angew. clusters (H2O) n, n= 11, 12. Chem. Phys. Lett. 2001, 333, 485−492.
Chem., Int. Ed. 2009, 48, 7862−7866. (56) Auer, B.; Kumar, R.; Schmidt, J.; Skinner, J. Hydrogen bonding
(36) Wu, P.; Tatsumi, T.; Komatsu, T.; Yashima, T. A novel and Raman, IR, and 2D-IR spectroscopy of dilute HOD in liquid
titanosilicate with MWW structure. I. Hydrothermal synthesis, D2O. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 14215−14220.
elimination of extraframework titanium, and characterizations. J. (57) Li, F.; Men, Z.; Li, S.; Wang, S.; Li, Z.; Sun, C. Study of
Phys. Chem. B 2001, 105, 2897−2905. hydrogen bonding in ethanol-water binary solutions by Raman
(37) Tang, Z.; Yu, Y.; Liu, W.; Chen, Z.; Wang, R.; Liu, H.; Wu, H.; spectroscopy. Spectrochim. Acta, Part A 2018, 189, 621−624.
Liu, Y.; He, M. Deboronation-assisted construction of defective (58) Lam, R. K.; Smith, J. W.; Saykally, R. J. Communication:
Ti(OSi)3OH species in MWW-type titanosilicate and their enhanced Hydrogen bonding interactions in water-alcohol mixtures from X-ray
catalytic performance. Catal. Sci. Technol. 2020, 10, 2905−2915. absorption spectroscopy. J. Chem. Phys. 2016, 144, No. 191103.

11653 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

(59) Cao, X.; Gong, N.; Zhao, H.; Li, Z.; Sun, C.; Men, Z. Raman
spectroscopic study of nonlinear modulation on Fermi resonance of
acetonitrile by hydrogen-bonding network. J. Mol. Liq. 2019, 279,
625−631.
(60) Wu, P.; Tatsumi, T. Unique trans-selectivity of Ti-MWW in
epoxidation of cis/trans-alkenes with hydrogen peroxide. J. Phys.
Chem. B 2002, 106, 748−753.
(61) Wang, L.; Liu, Y.; Xie, W.; Zhang, H.; Wu, H.; Jiang, Y.; He,
M.; Wu, P. Highly efficient and selective production of epichlorohy-
drin through epoxidation of allyl chloride with hydrogen peroxide
over Ti-MWW catalysts. J. Catal. 2007, 246, 205−214.
(62) Wu, P.; Tatsumi, T. A novel titanosilicate with MWW
structure: III. Highly efficient and selective production of glycidol
through epoxidation of allyl alcohol with H2O2. J. Catal. 2003, 214,
317−326.
(63) Wu, P.; Nuntasri, D.; Liu, Y.; Wu, H.; Jiang, Y.; Fan, W.; He,
M.; Tatsumi, T. Selective liquid-phase oxidation of cyclopentene over
MWW type titanosilicate. Catal. Today 2006, 117, 199−205.
(64) Wu, P.; Liu, Y.; He, M.; Tatsumi, T. A novel titanosilicate with
MWW structure: catalytic properties in selective epoxidation of diallyl
ether with hydrogen peroxide. J. Catal. 2004, 228, 183−191.
(65) Kirm, I.; Medina, F.; Rodríguez, X.; Cesteros, Y.; Salagre, P.;
Sueiras, J. Epoxidation of styrene with hydrogen peroxide using
hydrotalcites as heterogeneous catalysts. Appl. Catal., A 2004, 272,
175−185.
(66) Nedić, M.; Wassermann, T. N.; Larsen, R. W.; Suhm, M. A. A
combined Raman-and infrared jet study of mixed methanol−water
and ethanol−water clusters. Phys. Chem. Chem. Phys. 2011, 13,
14050−14063.
(67) Hribar, B.; Southall, N. T.; Vlachy, V.; Dill, K. A. How ions
affect the structure of water. J. Am. Chem. Soc. 2002, 124, 12302−
12311.
(68) Walrafen, G. E. Raman spectral studies of the effects of
temperature on water and electrolyte solutions. J. Chem. Phys. 1966,
44, 1546−1558.
(69) Capel-Sanchez, M. C.; Campos-Martin, J. M.; Fierro, J. L. G.
Impregnation treatments of TS-1 catalysts and their relevance in Recommended by ACS
alkene epoxidation with hydrogen peroxide. App. Catal., A 2003, 246,
69−77.
(70) Yu, Y.; Tang, Z.; Wang, J.; Wang, R.; Chen, Z.; Liu, H.; Shen, Highly Selective Production of Renewable 1,3-Pentadiene
K.; Huang, X.; Liu, Y.; He, M. Insights into the efficiency of hydrogen from 1,4-Pentanediol over an Acid–Base (K–Ce/ZrSi)
peroxide utilization over titanosilicate/H2O2 systems. J. Catal. 2020, Catalyst by Adjusting the Parallel-Reaction Pathway
381, 96−107. Dongying Yan, Chenxi Bai, et al.
(71) Jorda, E.; Tuel, A.; Teissier, R.; Kervennal, J. Synthesis, FEBRUARY 23, 2023
Characterization, and Activity in the Epoxidation of Cyclohexene with INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH READ
Aqueous H2O2 of Catalysts Prepared by Reaction of TiF4with Silica.
J. Catal. 1998, 175, 93−107. Selective Ammoximation of Ketones via In Situ H2O2
(72) Wu, H.; Liu, Y.; Wang, L.; Zhang, H.; He, M.; Wu, P.
Synthesis
Epoxidation of 2, 5-dihydrofuran to 3, 4-epoxytetrahydrofuran over
Ti-MWW catalysts. App. Catal., A 2007, 320, 173−180. Richard J. Lewis, Graham J. Hutchings, et al.
(73) Palumbo, C.; Tiozzo, C.; Ravasio, N.; Psaro, R.; Carniato, F.; JANUARY 19, 2023
Bisio, C.; Guidotti, M. An efficient epoxidation of terminal aliphatic ACS CATALYSIS READ
alkenes over heterogeneous catalysts: when solvent matters. Catal. Sci.
Technol. 2016, 6, 3832−3839. Catalytic Transfer Hydrogenation and Ethanolysis of
Furfural to Ethyl Levulinate Using Sulfonated Hf- or Ni-
Catalysts Prepared with Mixed Solvents
Hengyu Hao, Richard Lee Smith Jr, et al.
NOVEMBER 30, 2022
ACS SUSTAINABLE CHEMISTRY & ENGINEERING READ

Low-Temperature Glycolysis of Polyethylene Terephthalate

Ngan Hong Le, Joungmo Cho, et al.
NOVEMBER 23, 2022
ACS SUSTAINABLE CHEMISTRY & ENGINEERING READ

Get More Suggestions >

11654 https://doi.org/10.1021/acssuschemeng.2c03524
ACS Sustainable Chem. Eng. 2022, 10, 11641−11654