Effects of Calcination Condition on the Network Structure of

Subscriber access provided by WEBSTER UNIV

Separations

Effects of Calcination Condition on the Network Structure of Triethoxysilane (TRIES) and How Si-H Groups Influence Hydrophobicity Under Hydrothermal Conditions Tsukasa Tanaka, Masakoto Kanezashi, Hiroki Nagasawa, and Toshinori Tsuru Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.8b06390 • Publication Date (Web): 19 Feb 2019 Downloaded from http://pubs.acs.org on February 21, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Effects of Calcination Condition on the Network Structure of Triethoxysilane (TRIES) and How Si-H Groups Influence Hydrophobicity Under Hydrothermal Conditions

Tsukasa Tanaka, Masakoto Kanezashi*, Hiroki Nagasawa, and Toshinori Tsuru Department of Chemical Engineering, Graduate School of Engineering, Hiroshima University Higashi-Hiroshima, 739-8527, Japan *[email protected]

ABSTRACT Network size control was evaluated for microporous membranes derived from triethoxysilane (TRIES) that contains highly reactive Si-H groups.

It was possible to

control the concentration of the Si-H groups via the conditions of calcination (temperature, atmosphere). Si-H groups remained within their network structure when the TRIES membrane was calcined at 350 ˚C under a N2 atmosphere, and had a loose network structure (H2 permeance: 5.40 × 10-7 mol m-2 s-1 Pa-1, H2/CH4 selectivity: 36). When calcination at high temperatures converted the Si-H groups to Si-O-Si groups, the TRIES membrane showed a high level of separation performance (H2 permeance: 2.34 ×10-7 mol m-2 s-1 Pa-1, H2/CH4 selectivity: 590) due to a densification of the network structure.

Compared with conventional microporous silica membranes, a TRIES

membrane with Si-H groups showed hydrophobic properties, but water vapor was adsorbed and/or capillary-condensed in the microporous structure, and permeation blocking for He molecules was observed at temperatures below 150 ˚C in the presence 1 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of saturated water vapor at 25 ˚C.

Page 2 of 36

Hydrophobic Si-H groups improved the

hydrothermal stability at 300 ˚C, but depending on the partial pressure of the steam, the reaction between Si-H groups and water vapor degraded the hydrothermal stability of the TRIES membranes.

KEYWORDS: Pore size control Gas separation Hydrophobic membrane Hydrothermal stability

2 ACS Paragon Plus Environment

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. INTRODUCTION Membrane-based gas separation is attractive, because there is no phase change, less energy is consumed, and operation is relatively easy, all of which makes this a promising technology, particularly in the petrochemical field.1,2

For example, the

processes of separating hydrocarbons3,4 in crude oil purification and the methane reforming of membrane reactors5, 6 can both be accomplished with membrane separation, which reduces much of the cost.

In order to realize pragmatic application, however,

many problems must be solved. Reports of the utility of gas separation membranes continue to show promise, but stability (thermal, chemical, hydrothermal) and membrane performance (permeance, selectivity) remain low, which means that pore size control for improved separation and stability remains quite important. Amorphous silica membranes that use the sol-gel method could be candidates for application to gas separation because they feature a thin separation layer, thermal resistance,7 and controllable pore size.8-10 One of the methods used to control a network involves selecting a Si precursor such as a pendant type11-15 or bridged type8,9,16,17 alkoxysilane.

Pendant type precursors have 3 functional groups so that the degree of

condensation is smaller than that for precursors with 4 functional groups that form looser silica networks.

Bridged-type precursors such as bis(triethoxysilyl)methane

(BTESM) and bis(triethoxysilyl)ethane (BTESE), which consist of Si-C-Si and Si-C-C-Si units, can increase gas permeance by adjusting the silica network. On the other hand, the performance of sol-gel derived silica membranes under hydrothermal conditions must be improved.

Some studies have reported that silica

membranes are unstable against water vapor.18,19 This is because siloxane bonding is attacked by water vapor and changes to Si-OH groups, which forms another form of 3 ACS Paragon Plus Environment


Page 4 of 36

siloxane bonding. This could lead to the densification of a silica network and a greatly decreased level of gas permeance.20

Methods to improve stability during hydrothermal

testing have included enhancement of hydrophobicity, and controlling the density of Si-OH groups that cause densification. hydrothermal

Researchers have attempted to enhance

stability by increasing hydrophobicity,21-24 enhancing structural

flexibility,22,23,25

and

increasing

the

degree

of

hydrolysis/condensation.23,25,26

Hydrophobic properties can decrease water adsorption ability and structural flexibility imparts freedom to silica networks, thereby reducing their reactivity with water vapor. A higher degree of hydrolysis/condensation is known to enhance hydrothermal stability. These properties reduce the density of Si-OH groups in a silica network and play an important role by preventing siloxane bonding from re-arranging under a hydrothermal atmosphere.27,28

Organosilica-derived membranes with either carbon groups or carbon

chains in their silica network structure are known to enhance both hydrophobicity and hydrothermal stability.21,29-32 Triethoxysilane (TRIES, HSi(OEt)3) is a pendant-type precursor with highly reactive Si-H groups.

Si-H groups are used to introduce new functional groups,14

reduce metal ions,33,34 and to promote a hydrosilylation reaction with vinyl groups (Si-CH=CH2).35,36

Si-H groups are also known as hydrophobic groups.37,38

Previous

papers regarding membrane fabrication have reported that the reaction between Si-H groups and NH3 formed Si-NH or Si-NH2 groups in an amorphous structure, enhanced H2 selectivity,14 and promoted thermally induced hydrosilylation with vinyl groups (Si-CH=CH2)35, 36 to effectively enhance thermal stability of the SiOC structure. The effect that the water molar ratio in a sol can exert on the network size of TRIES membranes15 has also been reported.

The degree of the dehydrogenation of Si-H 4

ACS Paragon Plus Environment

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


groups in an aqueous solution is independent of the H2O/Si molar ratio in a sol, but the degree of the hydrolysis and polymerization of ethoxy groups (-OEt) depends on the H2O molar ratio.

The pyrolysis of ethoxy groups can serve as a template39 in a

network following calcination to enlarge the network size in the case of a low H2O molar ratio. TRIES-derived membranes are expected to control pore size by utilizing Si-H reactivity and to show hydrothermal stability due to the presence of hydrophobic Si-H groups in a network structure.

It remains unclear, however, how the conditions of

calcination (temperature, atmosphere) exert their effects on Si-H groups and on hydrothermal stability under different forms of steam atmosphere.

In the present study,

TRIES-derived membranes were fabricated under different calcination conditions, and the effects exerted on network pore size and Si-H groups were evaluated according to gas permeation properties, as verified by FT-IR measurement.

The microstructure and

hydrophobic properties of TRIES powder prepared under different calcination conditions were evaluated via measurement of the N2 adsorption and H2O adsorption isotherms, respectively.

The gas permeation properties of the hydrophobic TRIES

membrane were evaluated under a moist atmosphere.

Hydrothermal stability was also

evaluated by measuring the time course for gas permeance under different forms of steam atmosphere (300 ˚C, H2O partial pressure: 3-90 kPa).

2. EXPERIMENTAL SECTION 2.1 Sol Preparation and Membrane Fabrication TRIES sols were prepared by hydrolysis and polymerization for 2 h at 50 ˚C in isopropyl alcohol with water and HCl (TRIES:H2O:HCl = 1:240:0.1 in a molar 5 ACS Paragon Plus Environment


ratio).14,15

Page 6 of 36

For membrane fabrication, porous silica glass tubes (average pore size of

500 nm, a porosity of 64%; Sumitomo Electrical Industries, Ltd., Japan) were used as substrates.27

A membrane was fabricated on the outside of the tube, and a picture of

the porous silica glass tube is shown in the Supporting Information (Figure S1).

The

TRIES membrane was composed of 3 layers: particle, intermediate, and separation layer. First, silica glass particles in a SiO2 sol (particle diameter: 200-300 nm) were deposited to flatten the surface, then SiO2-ZrO2 sols (20 nm) were coated onto particle layers. Both layers were calcined at 550 ˚C under air for 15 minutes to produce an average pore size of approximately 1 nm.

Finally, TRIES sols were coated to form gas-separation

layers, which was followed by calcination at 350-550 ˚C under either a N2 or an air atmosphere.

2.2 Characterization of TRIES-Derived Sols and Gels The FT-IR spectra were measured using a FT-IR spectrometer (FT/IR-4100, Jasco, Japan). Films were prepared to coat TRIES sols onto KBr plates for calcination under different conditions.

TRIES gels intended for use in the adsorption experiment and for

application to the diffuse reflectance FT-IR method were prepared by drying a TRIES sol at 40 ˚C. N2 and H2O adsorption isotherms for TRIES and TEOS gels were measured at 77 K and 298 K, respectively, to evaluate the micro structures (BELMAX, BEL JAPAN INC.). All samples were evacuated at 200 ˚C for 12 h before measurement. Brubauer-Emmett-Teller (BET) analysis was used to determine the surface area to compare levels of hydrophobicity, which was established via the measurement of water contact angles on Si-wafers for TRIES and TEOS films calcined under different conditions. 6 ACS Paragon Plus Environment

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2.3 Gas Permeation Experiment and Hydrothermal Stability Figure 1 shows a schematic diagram of single-gas permeation and hydrothermal stability testing.

Gas was fed on the outside of the membrane, and a picture of the

permeation module is shown in Figure S2.

Gas permeation properties were evaluated

by feeding gases (He, H2, CO2, N2, CH4) at temperatures ranging from 100 to 500 ˚C. Gas permeance was recorded for a feed stream at atmospheric pressure, and the permeate stream was evacuated via a vacuum pump.

In the case of moist gas

permeation (H2O partial pressure = 3 kPa), He was fed to the permeation cell via an H2O evaporator at room temperature.

The H2O partial pressure was maintained as

saturation vapor at room temperature.

When the H2O partial pressure was larger than

10 kPa, partial pressure was maintained by controlling the flow rates of N2 and H2O.27 Water from a plunger pump was heated to evaporation and introduced to a membrane module via N2. The reaction between TRIES and water vapor was evaluated by diffuse reflectance FT-IR spectra.

Water vapor pressure was controlled using the same procedure as

membrane treatment.

TRIES gel was set in the gas permeation module, heated to 300

˚C, and exposed to a hydrothermal atmosphere with H2O partial pressure that was controlled from 0 to 70 kPa.

After one hour of exposure to water vapor, single N2 gas

was fed for an hour to dry the sample at 300 ˚C. For the diffuse reflectance FT-IR spectra, KBr powder was added to dilute the TRIES powder (TRIES:KBr = 1:9 in a volume fraction).

Each of the peak ratios for Si-O-Si (1080 cm-1), Si-H (2250 cm-1),

and Si-OH (3740 cm-1) was calculated to evaluate the reactivity of TRIES with water vapor. 7 ACS Paragon Plus Environment


1. 2. 3. 4. 5. 6.

Gas cylinder Pressure controller Pressure gauge Mass flow controller Flow meter Electric furnace

Operation conditions Feed : 0.1 MPa Temp. : 25 ~ 300oC

7. Membrane 8. Temperature controller 9. Orifice 10. Vacuum pump 11. Water bubbler 12. Plunger pump

Probe gases He, H2, N2, CO2, CH4 Hydrothermal test H2O partial pressure: 0-90 kPa

3

2

2

1

Page 8 of 36

1

4 5 11 12

6

9

7

10

3

8

Figure 1 A schematic diagram of single-gas permeation and hydrothermal stability testing.

3. RESULTS and DISCUSSION 3.1 Effect of Calcination Conditions on the Physicochemical Properties of TRIES Figure 2 shows the FT-IR spectra of TRIES films ranging from 400-4000 cm-1 (a) and 600-900 cm-1 (b) calcined under different conditions.

The peaks for Si-H groups

were detected at around 830 and 2250 cm-1, respectively. The peaks for Si-OH groups were detected at the shoulder of peaks between 800 and 900 cm-1.40

A TRIES film

before calcination showed a peak at 1100 cm-1, which can be assigned to Si-O-Si 8 ACS Paragon Plus Environment

Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


asymmetric stretching vibrations,36 which suggests the formation of silica networks. It should be noted that CH3 peaks in -OR groups (2974 cm-1),15 corresponding to the ethoxy groups and/or the interchange reaction between ethoxy groups of the TRIES and isopropanol, were not detected in the sample before calcination, indicating the complete hydrolysis of Si-OR groups.

A TRIES film (350˚C, N2) showed approximately the

same peak height of Si-H groups (2250 cm-1) of a TRIES film before calcination, and a drastic decrease in Si-H groups (2250 cm-1) was observed when the TRIES was calcined at 550 oC under N2. When a TRIES film was calcined at 550 ˚C under air, Si-H peaks were not detected, because the Si-H groups were oxidized completely (Si-H + 1/2O2 → Si-OH).38 Table 1 summarizes the peak height ratios of Si-H (2250 cm-1)/Si-O-Si (1080 cm-1) for TRIES films calcined under different conditions.

A calcined TRIES film (350 ˚C,

N2) showed approximately the same peak height ratio of Si-H/Si-O-Si as a TRIES film before calcination, indicating that Si-H groups didn’t react at 350 ˚C under a N2 atmosphere.

The peak height ratios of Si-H/ Si-O-Si for two TRIES films calcined at

350 ˚C and 550 ˚C, both under N2, were 0.319 and 0.12, respectively, because dehydrogenation between both the Si-H and the Si-OH groups occurs at around 400 ˚C (Si-H + Si-OH → Si-O-Si + H215, 37 ). The Si-H groups were not reacted completely because the Si-H groups were fixed in the silica networks, which meant that some of the Si-H groups were far from Si-OH groups. Thus, the reaction of Si-H groups was controlled by the calcination conditions (temperature, atmosphere), and a low calcination temperature in an inert atmosphere was preferable to preserve the presence of Si-H groups in TRIES-derived networks.



Si-OH

(a)

Si-O-Si

Si-H

(b)

Si-H

550°C, air 550°C, N2 350°C, N2

550°C, N2 350°C, N2 before calcination

before calcination 4000

550°C, air

Si-OH Si-H

Absorbance [a. u.]

Absorbance [a. u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 36

3000 2000 Wavenumber [cm -1]

1000

900

800 700 -1 Wavenumber [cm ]

600

Figure 2 FT-IR spectra of a TRIES film calcined under different conditions (temperature, atmosphere) ((a) 400-4000 cm-1, (b) 600-900 cm-1).

Table 1 Peak height ratios of TRIES films calcined under different conditions (temperature, atmosphere).

Si-H (2250 cm -1) / Si-O-Si (1080 cm -1)

TRIES Before calcination

0.391

TRIES 350 ˚C, N2

0.319

TRIES 550 ˚C, N2

0.120

TRIES 550 ˚C, air

N. A.

Figure 3 shows the N2 adsorption isotherm at 77 K for TRIES and TEOS powders


Page 11 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


calcined under different conditions.

All samples showed type I adsorption isotherms,

indicating microporous structures, irrespective of either the Si precursor or the calcination conditions.

The adsorption amounts had the following order: TEOS (350

˚C, air) > TRIES (350 ˚C, N2) > TRIES (550 ˚C, N2) > TRIES (550 ˚C, air).

The

adsorption amount of TRIES (350 ˚C, N2) was smaller than that of TEOS (350 ˚C, air). This was likely caused by the number of functional groups being different, and precursors with 4 functional groups (TEOS) formed a more porous structure than those with 3 functional groups (TRIES).

It should be noted that the N2 adsorption properties

for TEOS-derived powders were independent of calcination atmosphere (N2, air) (Figure S3). The adsorption amount for TRIES powder was decreased as calcination temperature increased.

Since Si-H groups reacted with Si-OH groups as temperature

was increased under a N2 atmosphere, after this reaction, new Si-O-Si bonds were formed and the network structures became dense.

When TRIES powder was calcined

at high temperatures under an air atmosphere, Si-H groups were oxidized completely so that a much denser network was formed. Thus, the adsorption amount for TRIES powder (550 ˚C, air) was smaller than that for TRIES powder that was calcined under different conditions (550 ˚C, N2).

It should be noted that the pore size distribution

showed a maximum of 2.1 nm for TRIES (350 oC, N2), a maximum of 1.9 nm for TRIES (550 oC, N2), and a maximum of 1.8 nm for TRIES (550 oC, air), indicating that the average pore size became smaller as the Si-H groups decreased, as shown in Figure S4.



400 TEOS 350°C air

300

TRIES 350°C N 2 550°C N 2 550°C air

3

-1

Va [cm g ]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 36

200

100

0 0

0.1

0.2

0.3

0.4

0.5

P/P0 [-]

Figure 3 N2 adsorption isotherm at 77 K for TEOS- and TRIES-derived powders calcined under different conditions (temperature, atmosphere).

Figure 4 shows the H2O adsorption isotherms at 298 K for TRIES and TEOS powders calcined under different conditions. The H2O adsorption amount of TRIES powder was much lower than that of TEOS, and the amount of adsorption increased as the calcination temperature increased.

The amount of H2O adsorption by TRIES

powder calcined at 550 ˚C under an air atmosphere was higher than that calcined under N2. Table 2 summarizes the data for water contact angle, BET surface area, and α (SBET(H2O)/SBET(N2)) of TRIES and TEOS films and powder calcined under different conditions.

TRIES film (350 ˚C, N2) showed a water-contact angle of 100˚, which

approximated the value of hydrophobic material such as methyltriethoxysilane (MTES).21

The water contact angle of TRIES film decreased as the calcination

temperature increased.

TRIES film (550˚C, air) with no Si-H groups showed a water 12 ACS Paragon Plus Environment

Page 13 of 36

contact angle of less than 40˚, which indicated a hydrophilic nature that approximated that of TEOS (550 ˚C, air).

In Table 2, α represents the ratio of the BET surface area

of H2O adsorption to that of N2 adsorption, which was used to evaluate hydrophobicity. The α of TRIES (350 ˚C, N2) had the smallest value due to its hydrophobic property. We concluded that Si-H groups are hydrophobic.

400

Va [cm3 g-1]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TEOS 350°C, TRIES 550°C, 550°C, 350°C,

300

200

air air N2 N2

100

0 0

0.1

0.2

0.3

0.4

0.5

P/P0 [-]

Figure 4 H2O adsorption isotherm at 298 K for TEOS- and TRIES-derived powders calcined under different conditions (temperature, atmosphere).

Table 2 Water contact angle, BET surface area, and α (SBET(H2O)/SBET(N2)) of TRIES and TEOS films and powders calcined under different conditions (temperature, atmosphere).



Page 14 of 36

Contact angle [˚]

SBET(H2O) [m2 g-1]

SBET(N2) [m2 g-1]

α (SH2O/SN2) [-]

TRIES 350 ˚C, N2

100

60

280

0.21

TRIES 550 ˚C, N2

50

77

130

0.59

TRIES 550 ˚C, air

40

80

104

0.77

TEOS 350 ˚C, air

30

410

610

0.64

Figure 5 shows an SEM image of a cross section of the TRIES-derived membranes calcined at 350 oC (a) and 550 oC (b). intermediate/silica-glass deposited layer.

A continuous thin layer can be seen on the The thickness and morphology of the

TRIES-derived membrane was independent of calcination conditions.

(a) Separation layer Intermediate/particle layer

Porous silica glass tube support

(b) Separation layer Intermediate/particle layer

Porous silica glass tube support


Page 15 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5 SEM image of the cross section of a TRIES-derived membrane calcined at 350 o

C (a) and 550 oC (b) (both membranes were calcined under a N2 atmosphere).

Figure 6 shows the molecular size dependence of gas permeance at 300 ˚C for TRIES membranes fabricated under different conditions. Gas permeance decreased approximately to the extent of molecular size increases, irrespective of membrane fabrication conditions, which indicated that molecular sieving dominated the permeation properties.

Table S1 shows gas selectivity at 300 oC for a TRIES

membrane calcined at 350 oC prepared with different types of acid (HCl, HNO3). He and H2 selectivity, corresponding to the network pore size, was approximately the same, so that the network pore size of the TRIES membrane was independent of the types of acid (HCl, HNO3), probably because both had the same level of acidity, which can affect the hydrolysis degree. TRIES membranes calcined at 350 ˚C under N2 showed higher levels of gas permeance (H2 permeance = 5.40 × 10-7 mol m-2 s-1 Pa-1) and a smaller degree of selectivity (He/N2 = 18) than those calcined at 550 ˚C under N2 (H2 permeance = 4.41 × 10-7 mol m-2 s-1 Pa-1, He/N2 = 190).

TRIES membranes calcined at 550˚C under air

showed low levels of permeance (H2 permeance = 2.34 × 10-7 mol m-2 s-1 Pa-1) with a high degree of selectivity (He/N2 = 260) by comparison with TRIES calcined at 550 ˚C under N2.

Based on selectivity, the order of network pore size was TRIES (350 ˚C, N2)

> TRIES (550 ˚C, N2) > TRIES (550 ˚C, air).

These results are consistent with N2

adsorption experiments showing that the network structure becomes denser as the



number of Si-H groups decreases.

It should be noted that the templating effect of

Si-OR groups would not affect the network pore size since the Si-OR groups in the TRIES were completely hydrolyzed, as shown in Figure 2.

10-5 Permeance [mol m-2 s-1 Pa-1 ]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 36

10

-6

He

350°C, N2 550°C, N2 550°C, air

H2 CO2

10-7

N2 CH4

10-8 10-9 10-10 0.25

0.3

0.35

0.4

Molecular size [nm]

Figure 6 Molecular size dependence of gas permeance at 300 ˚C for TRIES membranes calcined under different conditions (temperature, atmosphere).

Figure 7 displays a schematic image of TRIES-derived membrane structures fabricated under different conditions.

Under an oxygen atmosphere at high


Page 17 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


temperatures, the oxidation of Si-H groups occurred simultaneously with their reaction, which formed a much denser network structure than when calcined under N2 atmosphere at high temperature. These results led us to conclude that silica network structures could be controlled via manipulation of the Si-H groups.

350˚C, N2

550˚C, N2

H O O Si

Si

550˚C, Air OH

H O

O Si O H OH Si Si H O O Si H Si O Si H O OH O O Si Si H O OH H Si

O O Si

Si

O

O Si O H Si O Si O O O Si Si H O Si O O O Si Si H O OH H Si

Si-H + Si-OH → Si-O-Si + H2

O O Si

O

Si

O

Si

Si

O Si O

O

O

Si O

O

Si

O

O

Si O

Si Si O O

Si

OH

Si-H + 1/2O2 → Si-O-Si

Figure 7 Schematic image of TRIES-derived membrane structures calcined under different conditions (temperature, atmosphere).

3.2 Effect of Steam on the Hydrothermal Stability of a TRIES Membrane TRIES was confirmed as hydrophobic, and this led to an evaluation of the hydrothermal stability of TRIES membranes.

First, we evaluated the gas-permeation

properties of hydrophobic TRIES membranes under a moist atmosphere.

Figure 8

shows the time course of He permeance at 25 ˚C for a TRIES membrane calcined at 350 ˚C under either a dry or a moist N2 (H2O partial pressure: 3 kPa) atmosphere. beginning, He permeance was steady under dry conditions at 25 ˚C.

In the

After switching

from dry to wet gas, however, the He permeance was drastically decreased (dry: 4.1 × 17 ACS Paragon Plus Environment


Page 18 of 36

10-7 mol m-2 s-1 Pa-1, moist atmosphere: less than detection limit (1.0 × 10-10 mol m-2 s-1 Pa-1)).

This was because the water vapor that was adsorbed by the membrane surface,

and/or the water vapor that had condensed on the membrane surface, prevented He permeation, because the relative pressure equalled 1 (bubbling temperature and gas permeation temperature: 25 oC). Next, the time course of He permeance was measured under a dry atmosphere. Adsorbed water molecules on the surface were removed by feeding single dry gas, so that He permeance was increased largely in the initial stage of the drying process and approached a steady value of 2.0 × 10-7 mol m-2 s-1 Pa-1, which was approximately half the initial value.

It should be noted that after the measurement of He permeance at 25

˚C, He permeance was fully recovered after conditioning at 300 ˚C for several hours, which suggests that the decrease in He permeance by adsorbed water is a reversible phenomenon. Although He permeance was expected to be unchanged in the presence of saturated water vapor at 25 ˚C, due to the hydrophobic nature of TRIES, water vapor was capillary condensed into the microporous structure where it exerted a permeation blocking effect for He molecules. The reason for condensed water vapor in a hydrophobic TRIES membranes is unclear at this stage, but one possibility might be the size of water vapor. Since the kinetic diameter of water vapor is the same as that of He (0.26 nm),41 there is the possibility that water vapor can capillary-condense in network pores through which He can permeate due to the small molecular size. Another possibility is the formation of a SiO2-rich phase with hydrophilic properties in the amorphous structure.

Since partial

dehydrogenation of Si-H groups (-Si-H + H2O → -Si-OH + H2) occurred during the TRIES sol preparation stage,15 water vapor can adsorb and/or capillary-condense into 18 ACS Paragon Plus Environment

Page 19 of 36

SiO2-rich networks, which would be very difficult to remove completely via the feeding of a single dry gas at 25 oC.

-2

-1

-1

He Permeance [mol m s Pa ]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10-6 10-7 10-8 10-9 detection limit

10-10 dry 10

0

dry

moist

-11

10

20

30

Time [h]

Figure 8 Time course of He permeance at 25 ˚C for TRIES membranes calcined under different conditions.

Figure 9 shows temperature dependence of He permeance for TRIES membrane calcined at 350 ˚C in N2 under dry and moist atmosphere.

He permeance in dry

atmosphere was decreased in a linear fashion as temperature decreased, which indicated that He had permeated the TRIES structure via activated diffusion.

On the contrary,

the temperature dependence of He permeance under a moist atmosphere was different from that under a dry atmosphere, and the slope of permeance was different.

The

slope of temperature dependence for He permeance followed the approximate trajectory 19 ACS Paragon Plus Environment


of permeation at temperatures higher than 150 ˚C, but a blocking effect was clearly observed below 100 ˚C.

A similar dependence on temperature was confirmed in

Nickel-doped silica membranes.42

-6

10

Permeance [mol m -2 s-1 Pa-1 ]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 36

300

Temperature [°C] 200 100 50

25

10-7

10-8

10-9

10-10 1.5

He He + water vapor

2

2.5 3 -1 1000/T [K ]

3.5

Figure 9 Temperature dependence of He permeance for TRIES membrane calcined at 350 oC in N2 (closed symbols: measurement under dry atmosphere, open symbols: measurement in a moist atmosphere, PH2O = 3kPa).

Figure 10 shows the time courses for gas permeance (He, N2) and for He/N2 selectivity at 300 ˚C for a TRIES membrane calcined at 350 ˚C under N2 and both dry and steam (H2O partial pressure: 3 kPa) atmospheres.

Before exposure to a steam 20

ACS Paragon Plus Environment

Page 21 of 36

atmosphere, steady states were confirmed for He and N2 permeance under a dry atmosphere at 300 ˚C. The values for gas permeance did not change, even under a steam atmosphere.

Although the blocking effect was obvious at low temperatures, gas

permeance was stable under a steam atmosphere (H2O partial pressure: 3 kPa, 300 ˚C). We concluded that even though microporous membranes show a hydrophobic nature, their separation performance is similar at temperatures higher than 150 ˚C under the

50 40 30 20 10-6

-1

-1

He/N2 [-]

presence of saturated water vapor at 25 ˚C.

Permeance [mol m s Pa ]

He

-2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dry

10-7

moist

dry

N2 10-8 0

5

10 15 20 25 Time [h]



Page 22 of 36

Figure 10 Time course for gas permeance and for He/N2 selectivity at 300 ̊C for a TRIES membrane calcined at 350 ̊C under N2 in a dry or moist atmosphere (H2O partial pressure: 3 kPa).

The effect that the partial pressure of steam exerts on the hydrothermal stability of a TRIES membrane was evaluated by measuring the molecular size dependence of gas permeance before/after steam treatment (H2O partial pressure: 30-90 kPa, 300 ˚C). Figure 11 shows the molecular size dependence of gas permeance at 300 ˚C for a TEOS-derived membrane (350 ˚C, N2) and for a TRIES-derived membrane (350 ˚C, N2) before/after steam treatment (300 ˚C, H2O partial pressure: 30-90 kPa).

It should be

noted that molecular size dependence of gas permeance was measured after confirming the steady state under each type of steam treatment.

The gas permeance of

TEOS-derived membranes was greatly decreased after steam treatment.

H2 and N2

permeance decreased by 72 and 90%, respectively, resulting in an increase in the H2/N2 permeance ratio of from 15 to 41 after steam treatment (300 oC, H2O partial pressure: 30 kPa). In the TEOS membrane, H2 was more permeable than He, which indicated that, for small molecules, Knudsen diffusion was dominant in the silica network.

After

hydrothermal treatment, however, molecular sieving dominated the permeation properties of the silica network for all molecules.

The H2/He permeance ratio changed

from 1.51 to 0.94, which indicated that the network structure became denser after hydrothermal treatment.

This densification can be explained by the generation of

silanol groups and recombination, and by the rearrangement of silanol groups in the siloxane network.20 22 ACS Paragon Plus Environment

Page 23 of 36

The TRIES membrane showed small changes in permeance after steam treatment (300 oC, H2O partial pressure: 30 kPa), so that there was no large change in H2/N2 permeance ratio before/after steam treatment in comparison with TEOS membrane. This could have been caused by enhanced hydrophobicity of the membrane. Hydrophobic Si-H groups decreased water adsorption ability,21 and prevented water vapor from attacking siloxane bonding.25 for hydrothermal stability.

Hydrophobic Si-H groups proved effective

However, gas permeance was greatly decreased following

steam treatment at 90 kPa.

H2 and N2 permeance decreased by 54 and 97%,

respectively, showing an increased H2/N2 permeance ratio of from 21 to 49.

The

H2/He permeance ratio for TRIES membranes decreased as water vapor pressure increased, 1.24 (fresh), 1.05 (30 kPa), and 0.79 (90 kPa).

Thus, it can be concluded

that TRIES-derived network membranes became denser as the partial pressure of steam increased

10-5

10-5

-6

10

He

Fresh 30 kPa

H2



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CO2 N2 CH4

10-7

10-8

(a) 10-9 0.25

0.3 0.35 Molecular size [nm]

0.4

Fresh 30 kPa 90 kPa

-6

10

He

H2 CO2

10-7

N2

CH4

10-8

(b) 10-9 0.25

0.3 0.35 Molecular size [nm]

0.4



Page 24 of 36

Figure 11 Molecular size dependence of gas permeance at 300 ˚C for a TEOS-derived membrane (350 ˚C, N2) (a) and for a TRIES-derived membrane (350 ˚C, N2) (b) before/after steam treatment (300 ˚C, H2O partial pressure: 30-90 kPa).

Figure 12 shows the FT-IR spectra for TRIES powder after each steam treatment. Before steam treatment, strong Si-H peaks were confirmed for TRIES. As partial water vapor increased, the isolated Si-OH peaks also increased, but the Si-H peaks decreased gradually.

Figure 13 shows the peak area ratios for TRIES powder (350 ˚C, N2) after

each steam treatment as a function of H2O partial pressure (300 ˚C, H2O partial pressure: < 70 kPa): Si-H (2250 cm-1) / Si-OH (3740 cm-1) and Si-OH (3740 cm-1) / Si-O-Si (1080 cm-1). The number of Si-H groups decreased as partial water vapor pressure increased.37

During TRIES sol preparation, the Si-H groups did not react

with H2O in the liquid phase.15

On the other hand, a reaction between the Si-H groups

and water vapor occurred, and this reaction was reported as dependent on the partial pressure of water vapor (Si-H + H2O → Si-OH + H2 37), which was consistent with FT-IR analysis in the present study, which showed that the hydrophobicity of TRIES was changed to hydrophilic37,38 following hydrothermal treatment, and hydrothermal stability was degraded at 90 kPa.

We concluded that the hydrophobic Si-H groups

improved the hydrothermal stability of the TRIES membrane under conditions that were mild and moist.


Page 25 of 36

70 kPa

K/M [a. u.]

30 kPa 10 kPa 3 kPa PH2O = 0 kPa Si-OH

4000

Si-H 3000

Si-O-Si 2000

1000 600

Wavenumber [cm -1]

Figure 12 FT-IR spectra of TRIES powders calcined at 350 ˚C under N2 following each

15

0.5 0.4

―

10

0.3 0.2

5

―

0.1 0 0

20 40 60 H2O partial pressure [kPa]

0 80

Si-OH (3740 cm-1) Si-O-Si (1080 cm-1) [-]

steam treatment (300 ˚C, H2O partial pressure: < 70 kPa).

Si-H (2250 cm-1)/Si-OH (3740 cm-1) [-]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60




Page 26 of 36

Figure 13 Si-H (2250 cm-1) / Si-OH (3740 cm-1) and Si-OH (3740 cm-1) / Si-O-Si (1080 cm-1) peak area ratios of TRIES powder (350 ˚C, N2) after each steam treatment as a function of H2O partial pressure (300 ˚C, H2O partial pressure: < 70 kPa).

4. CONCLUSIONS Network size control was evaluated for microporous membranes derived from triethoxysilane (TRIES), which possesses highly reactive Si-H groups.

These

TRIES-derived membranes were fabricated under different calcination conditions, and the effect these conditions exerted on network pore size and Si-H groups was evaluated. Si-H concentration was controlled via the conditions of calcination (temperature, atmosphere). The Si-H groups remained within the network structure when TRIES membranes were calcined at 350 ˚C under a N2 atmosphere, and promoted the formation of a loose network structure (H2/CH4 selectivity: 36).

When Si-H groups were

calcined at high temperature under an air atmosphere, they were converted to Si-O-Si groups, which enhanced the separation performance (H2/CH4 selectivity: 590) due to a densification of the network structure. Thus, via changes to the Si-H groups, the network structure of the TRIES membranes could be controlled. Hydrothermal stability was also evaluated by measuring the time courses for gas permeance under different types of steam atmosphere (300 ˚C, partial pressure of steam: 3-90 kPa). By comparison with conventional microporous silica membranes, TRIES membranes showed hydrophobic properties due to the presence of Si-H groups, but water vapor was adsorbed and/or capillary-condensed in the microporous structure, and permeation blocking for He molecules was observed at temperatures below 150 ˚C in 26 ACS Paragon Plus Environment

Page 27 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the presence of saturated water vapor at 25 ˚C.

The hydrophobic Si-H groups

improved the hydrothermal stability at 300 ˚C, but depending on the partial pressure of steam, the reaction between Si-H groups and water vapor tended to degrade the hydrothermal stability of the TRIES membranes.

REFERENCES (1) Sholl, D. S.; Lively, R. P. Seven chemical separations to change the world. Nature. 2016, 532, 435-437. (2) Koros, W. J. Evolving beyond the thermal age of separation processes: membranes can lead the way. AIChE J. 2005, 50, 2326-2334. (3) Tirouni. I.; Sadegh. M.; Pakizeh. M. Separation of C3H8 and C2H6 from CH4 in polyurethane–zeolite 4Å and ZSM-5 mixed matrix membranes. Sep. Purif. Tech. 2015, 141, 394-402. (4) Lin, Y. S.; Kumakiri, I.; Nair, B. N.; Alsyouri, H. Microporous inorganic membranes. Sep. Purif. Methods, 2002, 32(2), 229-379. (5) Tsuru, T.; Tsuge, T.; Kubota, S.; Yoshida, K.; Yoshioka, T.; Asaeda, M. Catalytic membrane reaction for methane steam reforming using porous silica membranes. Sep. Sci. Technol. 2001, 36, 3721-3736. (6) Uemiya, S.; Sato, N.; Ando, H.; Matsuda, T.; Kikuchi, E. Steam reforming of methane in a hydrogen-permeable membrane reactor. Appl. Catal. 1991, 67, 223-230. (7) Agirre, I.; Arias, P. L.; Castricum, H. L.; Creatore, M.; ten Elshof, J. E.; Paradis, G. G.; Ngamou, P. H. T.; van Veen, H. M.; Vente, J. F. Hybrid organosilica membranes and processes: status and outlook. Sep. Purif. Technol. 2014, 121, 2-12. 27 ACS Paragon Plus Environment


Page 28 of 36

(8) Castricum, H. L.; Paradis, G. G.; M.-Hazeleger, M. C.; Kreiter, R.; Vente, J. F.; ten Elshof, J. E. Tailoring the separation behavior of hybrid organosilica membranes by adjusting the structure of the organic bridging group. Adv. Funct. Mater. 2011, 21, 2319-2329. (9) Kanezashi, M.; Yada, K.; Yoshioka, T.; Tsuru, T. Organic-inorganic hybrid silica membranes with controlled silica network size: preparation and gas permeation characteristics. J. Membr. Sci. 2010, 348, 310-318. (10) Cao, G.; Lu, Y.; Delattre, L.; Brinker, C. J.; Lopez, G. P. Amorphous silica molecular sieving membranes by sol-gel processing. Adv. Mater. 1996, 8, 588-591. (11) Li, G.; Kanezashi, M.; Tsuru, T. Preparation of organic-inorganic hybrid silica membranes using organoalkoxysilanes: the effect of pendant groups. J. Membr. Sci. 2011, 379, 287-295. (12) Paradis, G. G.; Shanahan, D. P.; Kreiter, R.; van Veen, H. M.; Castricum, H. L.; Nijmeijer, A.; Vente, J. F. From hydrophilic to hydrophobic HybSi® membranes: a change of affinity and applicability. J. Membr. Sci. 2013, 428, 157-162. (13) Xomeritakis, G.; Tsai, C. Y.; Brinker, C. J. Microporous sol-gel derived aminosilicate membrane for enhanced carbon dioxide separation. Sep. Purif. Technol. 2005, 42, 249-257. (14) Kanezashi, M.; Matsugasako, R.; Tawarayama, H.; Nagasawa, H.; Yoshioka, T.; Tsuru, T. Tuning the pore sizes of novel silica membranes for improved gas permeation properties via an in situ reaction between NH3 and Si-H groups. Chem. Commun. 2015, 51, 2551-2554. (15) Kanezashi, M.; Matsugasako, R.; Tawarayama, H.; Nagasawa, H.; Yoshioka, T.; Tsuru, T. Pore size turning of sol-gel-derived triethoxysilane (TRIES) membranes 28 ACS Paragon Plus Environment

Page 29 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


for gas separation. J. Membr. Sci. 2017, 524, 64-72. (16) Niimi, T.; Nagasawa, H.; Kanezashi, M.; Yoshioka, T.; Ito, K.; Tsuru, T. Preparation of BTESE-derived organosilica membranes for catalytic membrane reactors of methylcyclohexane dehydrogenation. J. Membr. Sci. 2014, 455, 375-383. (17) Moriyama,

N.;

Bis(triethoxy)ethane

Nagasawa,

H.;

Kanezashi,

(BTESE)-derived

silica

M.;

Ito,

membranes:

K.; pore

Tsuru,

T.

formation

mechanism and gas permeation properties. J. Sol-Gel Sci. Technol. 2018, 86, 63-72. (18) Wu, J. C. S.; Sabol, H.; Smith, G. W.; Flowers, D. L.; Liu, P. K. T. Characterization of hydrogen-permselective microporous ceramic membrane. J. Membr. Sci. 1994, 96, 275-287. (19) Sea, B. K.; Watanabe, M.; Kusakabe, K.; Morooka, S.; Kim, S. S. Formation of hydrogen permselective silica membrane for elevated temperature hydrogen recovery from a mixture containing steam. Gas. Sep. Purif. 1996, 10, 187-195. (20) Tsuru, T.; Igi, R.; Kanezashi, M.; Yoshioka, T. Permeation properties of hydrogen and water vapor through porous silica membranes at high temperature. AIChE J. 2011, 57, 618-629. (21) de Vos, R. M.; Maier, W. F.; Verweij, H. Hydrophobic silica membranes for gas separation. J. Membr. Sci. 1999, 158, 277-288. (22) ten Hove, M.; Luiten-Olieman, Mieke W. J.; Huiskes, C.; Nijmeijer, A.; Winnubst, L. Hydrothermal stability of silica, hybrid silica and Zr-doped hybrid silica membranes. Sep. Purif. Technol. 2017, 189, 48-53. (23) Castricum, H. L.; Sah, A.; Kreiter, R.; Blank, D. H. A.; Vente, J. F.; ten Elshof, J. E. Hydrothermally stable molecular separation membranes from organically linked 29 ACS Paragon Plus Environment


Page 30 of 36

silica. J. Mater. Chem. 2008, 18, 2150-2158. (24) Guo, W.; Li, X.; Zhao, X. S. Understanding the hydrothermal stability of large-pore periodic mesoporous organosilica and pure silicas. Microporous Mesoporous Mater. 2006, 93, 285-293. (25) ten Elshof, J. E.; Dral, A. P. Structure-property turning in hydrothermally stable sol-gel-processed hybrid organosilica molecular sieving membranes. J. Sol-Gel Sci. Technol. 2016, 79, 279-294. (26) Castricum, H. L.; Kreiter, R.; van Veen, H. M.; Blank, D. H. A.; Vente, J. F.; ten Elshof, J. E. High-performance hybrid pervaporation membranes with superior hydrothermal and acid stability. J. Membr. Sci. 2008, 324, 111-118. (27) Kanezashi, M.; Matsutani, T.; Nagasawa, H.; Tsuru, T. Fluorine-induced microporous silica membranes: Dramatic improvement in hydrothermal stability and pore size controllability for highly permeable propylene/propane separation. J. Membr. Sci. 2018, 549, 111-119. (28) Kanezashi, M.; Sasaki, T.; Tawarayama, H.; Yoshioka, T.; Tsuru, T. Hydrogen permeation properties and hydrothermal stability of sol-gel-derived amorphous silica membraned fabricated at high temperatures. J. Am. Ceram. Soc. 2013, 96 (9), 2950-2957. (29) Duke, M. C.; da Costa, J. C. D.; Lu, G. Q.; Petch, M.; Gray, P. Carbonised thermally template molecular sieve silica membranes in fuel processing systems: permeation, hydrostability and regeneration. J. Membr. Sci. 2004, 241, 325-333. (30) Duke, M. C.; da Costa, J. C. D.; Do, D. D.; Lu, G. Q.; Petch, M.; Gray, P. Hydrothermally robust molecular sieve silica for wet gas separation. Adv. Funct. Mater. 2004, 16, 1215-1220. 30 ACS Paragon Plus Environment

Page 31 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(31) Castricum, H. L.; Sah, A.; Geenevasen, J. A. J.; Kreiter, R.; Blank, D. H. A.; Vente, J. F.; ten Elshof, J. E. Structure of hybrid organic-inorganic sols for the preparation of hydrothermally stable membranes. J. Sol-Gel Sci. Technol. 2008, 48, 11-17. (32) Qi, H.; Han, J.; Xu, N. P.; Bouwmeester, H. J. M. Hybrid organic-inorganic microporous membranes with high hydrothermal stability for the separation of carbon dioxide. J. Sol-Gel Sci. Technol. 2010, 3 (12), 1375-1378. (33) Ivashchenko, N.; Tertykh, V.; Yanishpolskii, V.; Skubiszewska-Zieba, J.; Leboda, R.; Khainakov, S. Potentialities of silane-modified silicas to regulate palladium nanoparticles sizes. J. Therm. Anal. Calorim. 2012, 108, 1121-1127. (34) Moitra, N.; Kanamori, K.; Shimada, T.; Takeda, K.; Ikuhara, Y. H.; Gao, X.; Nakanishi, K. Synthesis of hierarchically porous hydrogen silsesquioxane monoliths and embedding of metal nanoparticles by on-site reduction. Adv. Funct. Mater. 2013, 23, 2714-2722. (35) Kanezashi, M.; Sazaki, H.; Nagasawa, H.; Yoshioka, T.; Tsuru, T. Evaluating the gas permeation properties and hydrothermal stability of organosilica membranes under different hydrosilylation conditions. J. Membr. Sci. 2015, 493, 664-372. (36) Kanezashi, M.; Sazaki, H.; Nagasawa, H.; Yoshioka, T.; Tsuru, T. Preparation and gas permeation properties of thermally stable organosilica membranes derived by hydrosilylation. J. Mater. Chem. A 2014, 2, 672-680. (37) Miura, T.; Niwano, M.; Shoji, D.; Miyamoto, N. Kinetics of oxidation on hydrogen-terminated Si(100) and (111) surfaces stored in air. J. Appl. Phys. 1996, 79, 4373-4380. (38) Liao, W.; Lee, S. Water-induced room-temperature oxidation of Si-H and -Si-Sibonds in silicon oxide. J. Appl. Phys. 1996, 80, 1171-1176. 31 ACS Paragon Plus Environment


Page 32 of 36

(39) Raman, N. K.; Brinker, C. J. “Organic template” approach to molecular sieving silica membranes. J. Membr. Sci. 1995, 105, 273-279. (40) Socrates, G. Infrared and Raman Characteristic Group Frequencies: Tables and Charts; John Wiley & Sons: New York, 2004. (41) Breck, D. W. Zeolite Molecular Sieves, Structure, Chemistry and Use; John Wiley: New York, 1974. (42) Kanezashi, M.; Fujita, T.; Asaeda, M. Nickel doped silica membranes for separation of helium from organic gas mixtures. Sep. Sci. Technol. 2005, 40, 225-238.


Page 33 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure Captions

Figure 1 A schematic diagram of single-gas permeation and hydrothermal stability testing.

Figure 2 FT-IR spectra of a TRIES film calcined under different conditions (temperature, atmosphere) ((a) 400-4000 cm-1, (b) 600-900 cm-1).

Figure 3 N2 adsorption isotherm at 77 K for TEOS- and TRIES-derived powders calcined under different conditions (temperature, atmosphere).

Figure 4 H2O adsorption isotherm at 298 K for TEOS- and TRIES-derived powders calcined under different conditions (temperature, atmosphere).

Figure 5 SEM image of cross section of TRIES-derived membrane calcined at 350 oC (a) and 550 oC (b) (both membranes were calcined in N2 atmosphere).

Figure 6 Molecular size dependence of gas permeance at 300 ˚C for TRIES membranes calcined under different conditions (temperature, atmosphere).

Figure 7 Schematic image of TRIES-derived membrane structures calcined under different conditions (temperature, atmosphere).



Page 34 of 36

Figure 8 Time course of He permeance at 25 ˚C for TRIES membranes calcined under different conditions.

Figure 9 Temperature dependence of He permeance for TRIES membrane calcined at 350 oC in N2 (closed symbols: measurement under dry atmosphere, open symbols: measurement in a moist atmosphere, PH2O = 3kPa).

Figure 10 Time course for gas permeance and for He/N2 selectivity at 300 ̊C for a TRIES membrane calcined at 350 ̊C under N2 in a dry or moist atmosphere (H2O partial pressure: 3 kPa).

Figure 11 Molecular size dependence of gas permeance at 300 ˚C for a TEOS-derived membrane (350 ˚C, N2) (a) and for a TRIES-derived membrane (350 ˚C, N2) (b) before/after steam treatment (300 ˚C, H2O partial pressure: 30-90 kPa).

Figure 12 FT-IR spectra of TRIES powders calcined at 350 ˚C under N2 following each steam treatment (300 ˚C, H2O partial pressure: < 70 kPa).

Figure 13 Si-H (2250 cm-1) / Si-OH (3740 cm-1) and Si-OH (3740 cm-1) / Si-O-Si (1080 cm-1) peak area ratios of TRIES powder (350 ˚C, N2) after each steam treatment as a function of H2O partial pressure (300 ˚C, H2O partial pressure: < 70 kPa).


Page 35 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table Captions

Table 1 Peak height ratios of TRIES films calcined under different conditions (temperature, atmosphere).

Table 2 Water contact angle, BET surface area, and α (SBET(H2O)/SBET(N2)) of TRIES and TEOS films and powders calcined under different conditions (temperature, atmosphere).



Page 36 of 36

For Table of Contents Only

He

N2 N2 H

H O Si O Si OH H Si Si H O H Si O Si O Si H O OH O O Si HO Si O O Si

Si

O

H

Improved hydrothermal stability by hydrophobic Si-H groups (300 ˚C, PH2O = 30 kPa)


Effects of Calcination Condition on the Network Structure of

Recommend Documents