Simple process for preparing mesoporous sol-gel silica adsorbents with high water adsorption capacities

Accepted Manuscript Simple process for preparing mesoporous sol-gel silica adsorbents with high water adsorption capacities Thays Lorrane Rodrigues Mota, Ana Paula Marques de Oliveira, Eduardo Henrique Martins Nunes, Manuel Houmard PII:

S1387-1811(17)30482-1

DOI:

10.1016/j.micromeso.2017.07.010

Reference:

MICMAT 8439

To appear in:

Microporous and Mesoporous Materials

Received Date: 4 April 2017 Revised Date:

23 May 2017

Accepted Date: 5 July 2017

Please cite this article as: T.L. Rodrigues Mota, A.P. Marques de Oliveira, E.H.M. Nunes, M. Houmard, Simple process for preparing mesoporous sol-gel silica adsorbents with high water adsorption capacities, Microporous and Mesoporous Materials (2017), doi: 10.1016/j.micromeso.2017.07.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

1

ACCEPTED MANUSCRIPT SIMPLE PROCESS FOR PREPARING MESOPOROUS SOL-GEL SILICA ADSORBENTS WITH HIGH WATER ADSORPTION CAPACITIES

Thays Lorrane Rodrigues Mota(1), Ana Paula Marques de Oliveira(1), Eduardo Henrique

RI PT

Martins Nunes(2), Manuel Houmard(3)*

(1) Department of Chemical Engineering, Federal University of Minas Gerais -

SC

UFMG, Avenida Presidente Antônio Carlos, 6627, Campus UFMG, Belo

Horizonte, MG, CEP: 31270-901, Escola de Engenharia, bloco 2, 5° andar –

M AN U

Brasil.

(2) Department of Metallurgical and Materials Engineering, Federal University of Minas Gerais - UFMG, Avenida Presidente Antônio Carlos, 6627, Campus

TE D

UFMG, Belo Horizonte, MG, CEP: 31270-901, Escola de Engenharia, bloco 2, sala 2233, Brasil.

EP

(3) Department of Materials Engineering and Civil Construction, Federal University of Minas Gerais - UFMG, Avenida Presidente Antônio Carlos, 6627, Campus

AC C

UFMG, Belo Horizonte, MG, CEP: 31270-901, Escola de Engenharia, bloco 1,

* [email protected]

sala 3304, Brasil.

2

ACCEPTED MANUSCRIPT ABSTRACT

In this work we present a simple route to prepare sol-gel silica samples for water adsorption applications. These materials were not chemically modified neither heat-

RI PT

treated before performing the water adsorption tests at 30 °C and atmospheric pressure. The materials studied in this work were examined by N2 adsorption and Fourier transform infrared spectroscopy (FTIR). The samples prepared herein showed higher

SC

adsorption capacities than a commercial silica commonly used as water adsorbent. It was shown that it is possible to produce materials with high water adsorption capacities

M AN U

controlling the sol pH. Samples obtained from sols with very low pH presented high number of small pores, whereas those prepared from sols without acid displayed larger pore; both cases increase the volume of pores and, consequently, the water uptake of the material. It is an important finding since several works available in the literature deals

TE D

with complex and multi-step procedures, commonly using pore-formers or poreexpanders agents, which makes them cost ineffective and time consuming. The absence of acidic reactants in the preparation step of the adsorbent with the highest adsorption

EP

capacity represents an advantage in terms of safety and environmental sustainability, besides reinforcing the simplicity of the processing route employed herein. The pseudo-

AC C

second order kinetic model was the best model to fit the water adsorption kinetics of the investigated samples, indicating a high interaction between the silica surface and water molecules. Besides, the water desorption was successfully carried out in relatively short times (< 2 h) and at temperatures as low as 90 °C.

KEYWORDS: Silica; Sol-gel synthesis; Mesoporous structures; Water sorption; Adsorption kinetics.

3

ACCEPTED MANUSCRIPT

INTRODUCTION

It is well established that the preparation of porous materials with an expressive water

RI PT

adsorption capacity is crucial for several industrial applications, including gas and food storage, sensing, catalysis, and delivery of drinking water in remote areas [1–3]. According to Roque-Malherbe [4], the water adsorption on porous solids is a promising

SC

technology for dehydration applications since it shows both a low energy consumption and high efficiency. In this context, silica has already been recognized as a powerful

M AN U

water adsorbent [5–7]. In addition to its remarkable water adsorption capacity, silica is non-toxic, non-corrosive and widely available material [8]. The water adsorption on the silica surface is generally due to the interaction between water molecules and silanol groups present in the porous structure of silica [9–11]. The mechanisms ascribed to

TE D

water adsorption on silica have been deeply investigated in previous works [12–17]. It is also well known that the specific pore volume, specific surface area and concentration of silanol groups on the silica surface can be tailored by using the sol-gel process [18].

EP

Thus, a fine control of the sol-gel parameters should allow improving the performance

AC C

of silica adsorbents.

Indeed, the size and arrangement of pores in silica is a key factor for the development of molecular sieves with a high adsorption capacity. Several studies published in the literature over the past years are mostly focused on tailoring the pore structure of silica by either incorporating different sacrificial additives or using expensive hybrid precursors [19–23]. These approaches commonly deal with the use of pore-formers or pore-expanders agents for obtaining silica samples with tailored structures. Thus, they

4

ACCEPTED MANUSCRIPT

are usually complex and may include multi-step procedures, such as multiple washing or heat treatment steps for removing previously incorporated additives or functional groups, which makes them cost ineffective and time consuming.

RI PT

In this work, we present a simple route to prepare sol-gel silica samples for water adsorption applications. These materials were not chemically modified neither heat treated before performing the water adsorption tests. It is shown that a simple control of

SC

the sol pH allowed preparing silica adsorbents with tailored pore structures and high adsorption capacities. The materials used in this work were examined by Nitrogen

M AN U

adsorption and Fourier transform infrared spectroscopy (FTIR). The adsorption kinetics of water on these materials was investigated on the basis of the pseudo first- and second order kinetic models. An intraparticle diffusion model was also used for elucidating the water adsorption mechanism observed in the samples examined herein. A commercial

TE D

silica sample commonly used as water adsorbent was used as the reference in this study. It were compared the structural properties and water uptake capacity of this sample with

EP

those of the materials prepared in this work.

AC C

MATERIALS AND METHODS

Silica sols were prepared using methodology similar to that reported by Houmard et al. [24].

Briefly,

a

solution

of

deionized

MilliQ

water,

Hydrochloric

acid

(HCl/Aldrich/37%) and ethanol (EtOH/Synth/≥99.5%) was initially prepared at room temperature. Tetraethyl orthosilicate (TEOS/Aldrich/98%) was then added to this solution. The as-prepared solution was kept under stirring at room temperature for 30 min. The molar ratio of TEOS: EtOH: H2O: HCl was kept at 1: 4: 4: x, where x could be

5

ACCEPTED MANUSCRIPT

0, 0.001 or 1. These molar ratios were selected on the basis of a previous work where we showed that they could lead to silica samples with different pore structures [24]. In order to speed up the synthesis of the silica structures, the prepared solutions were then kept at 90 °C for 24 h until the total gelation of the sol-gel solution. Finally, after air-

RI PT

drying at 90 °C, the samples were washed and filtered with plenty of milliQ water and air-dried again at 90 °C before the adsorption tests. As discussed previously, these materials were not chemically modified neither heat-treated. The simplicity and novelty

SC

of the processing route used in this work consist in tailoring the pore structure of the silica adsorbent by only varying the sol pH. A commercial silica sample (high purity

M AN U

grade / Cromoline Química Fina - Brazil) commonly used in dehydration applications was also examined for comparison purposes. According to the provider, this material shows a particle size ranging from 1 to 4 mm and density of about 0.74-0.87 g.cm-3.

TE D

As aforementioned, the samples used in this study were examined by Nitrogen adsorption and FTIR. Nitrogen adsorption was carried out in a Quantachrome Nova 1200e apparatus within an experimental error of 5%. The samples used in these tests

EP

were previously degassed under vacuum at 150 °C for up to 12 h. The specific surface area was assessed by the multipoint BET (Brunauer-Emmett-Teller) method. The mean

AC C

pore diameter (dp) was estimated by Equation (1):

dp =

4Vmes Sp

(1)

where Vmes and Sp are the mesoporous volume and specific surface area, respectively. Vmes was assessed from the collected N2 adsorption isotherms. FTIR was performed using air-dried samples (90 °C) in a Bruker Alpha spectrometer with an attenuated total

6

ACCEPTED MANUSCRIPT

reflectance (ATR) module. A diamond crystal was used as the reflection element in these tests. The spectra were taken in the spectral range from 4000 to 400 cm-1 at a resolution of 4 cm-1 and 128 scans. Next they were normalized using the band at 460 cm-1 as the reference. The curve fittings shown in this work were obtained using the

RI PT

Peakfit software.

The water adsorption capacity of the materials used herein was assessed as follows.

SC

Samples were initially kept in a stove at 90 °C in air until their masses reached a constant value. This step was performed in order to dry these samples before the water

M AN U

adsorption tests. They were subsequently transferred to a humidity-controlled chamber (Fanem 345) and kept at 30 °C in an atmosphere of 60% relative humidity. The mass gain of these materials during the water adsorption step was monitored as a function of time. This procedure was performed until a constant mass was reached. The water

TE D

desorption step was then carried out by keeping the samples in the aforementioned stove in air at 90 °C. Their mass loss was also monitored as a function of time until a constant

AC C

Equation (2):

EP

mass was reached. The adsorption capacity of the tested samples was evaluated by

qt =

m t − m dry m dry

( 2)

where qt represents the adsorbed amount of water, mt the mass assessed at a time t, and mdry the mass of the dried sample. It is worth mentioning that both the water adsorption and desorption steps were performed at atmospheric pressure (about 1 bar).

7

ACCEPTED MANUSCRIPT RESULTS AND DISCUSSION

Figure 1 exhibits a typical FTIR spectrum of the samples obtained in this work. The absorption band at 460 cm-1 is related to the rocking mode of Si-O-Si bonds [25]. The

RI PT

bands at about 806 and 1090 cm-1 are associated, respectively, with symmetric and asymmetric stretching modes of these bonds [26,27]. The presence of these bands in the collected FTIR spectra confirms that sol-gel silica samples were successfully obtained

SC

in this study. The band at 560 cm-1 has been assigned to the skeletal vibration of 4-fold siloxane rings [28]. This feature is not commonly observed in silica samples because

M AN U

they are usually examined in the heat-treated condition. It is worth recalling that the samples prepared herein were not heat-treated but only air-dried. The features at 960 and 1630 cm-1 are ascribed to Si-OH groups and physisorbed water, respectively [29,30]. The broad band centered at about 3470 cm-1 is assigned to OH groups present

TE D

in the silica framework [31]. The high intensity of the band at 960 cm-1 reveals a significant presence of surface silanol groups on these samples. As will be better discussed further on, this behavior could explain the expressive affinity between the

EP

samples prepared herein and water. The commercial silica used as the reference adsorbent in this work showed a FTIR spectrum similar to that exhibited in Figure 1,

AC C

which confirms that it is a high purity silica sample.

Figure 2 depicts the water sorption cycles obtained in this study. It can be observed that two sorption cycles are shown for each sample. No significant variations in the samples masses was observed after the second desorption step, which could indicate that no additional modification took place in their structures after the first desorption cycle. On the basis on this finding, no additional water sorption cycles were collected. As it will

8

ACCEPTED MANUSCRIPT

be better described further on, the mass loss observed for the sol-gel samples after the first desorption cycle could be also related to the removal of residual ethoxy groups. From Figure 2, one notices that the sample obtained with no catalyst addition showed the highest adsorption capacity, followed by the samples prepared using TEOS: HCl

RI PT

molar ratios of 1: 1 or 1: 0.001. It can also be observed that the commercial silica exhibited an intermediary adsorption capacity, only larger than the capacity of the

SC

sample prepared using a TEOS: HCl molar ratio of 1: 0.001.

Figure 3 shows the N2 adsorption isotherms obtained for the samples studied in this

M AN U

work. Hysteresis loops ascribed to the capillary condensation of nitrogen in mesopores are clearly observed in these isotherms [32]. One notices that this hysteresis loop is wider for the material prepared with no catalyst addition, which reveals a significant presence of mesoporous in its pore framework. A hysteresis loop is also clearly

TE D

observed for the sample obtained with a 1 TEOS: 4 EtOH: 4 H2O: 1 HCl molar ratio. On the other hand, narrow hysteresis loops were observed for both the commercial silica and sample prepared using a TEOS: HCl molar ratio of 1: 0.001.Table 1 gives the

EP

specific surface area, mesoporous volume and mean pore diameter assessed for the tested samples. From Figure 2 and Table 1, one notices that the water adsorption

AC C

behavior of the samples examined herein is closely related to their pore structure. The sample prepared with no catalyst addition showed the highest adsorption capacity due to its expressive surface area (597 ± 30 m2.g-1), mesoporous volume (0.94 ± 0.05 cm3.g-1) and mean pore diameter (6.3 ± 0.4 nm). Thus, one may suggest that this material exhibits a significant number of sites available for water adsorption. Both the commercial silica and sample obtained with a 1 TEOS: 4 EtOH: 4 H2O: 1 HCl molar ratio displayed less intense water adsorption capacities because of their smaller

9

ACCEPTED MANUSCRIPT

mesoporous volumes and mean pore diameters. The silica prepared with a TEOS: HCl molar ratio of 1: 0.001 exhibited the lowest water adsorption capacity, which is probably related to its small mesoporous volume and mean pore diameter. The sol-gel solution that gave rise to this sample showed a pH around 2.7, which is close to the

RI PT

isoelectric point of silica [33,34]. It has been reported that for pH between 2 and 3 the interaction between adjacent –Si–O–Si– polymeric chains is less likely, which leads to a decrease in the kinetics of the condensation reactions [35]. Because of this slower

SC

condensation rate, adjacent –Si–O–Si– polymeric chains are able to react in a more compacted fashion, which gives rise to less porous structures. Therefore, the small

M AN U

mesoporous volume and mean pore diameter displayed by the sample prepared using a TEOS: HCl molar ratio of 1: 0.001 could be related to the pH of the sol-gel solution used to obtain it. The low water adsorption capacity of this sample could be ascribed to

TE D

a smaller volume of pores accessible for water adsorption.

From Figure 2, one notices that the samples prepared in this study showed a significant mass loss when the masses observed before and after the water sorption tests are taken

EP

into consideration (after these tests their masses were below 100%). This behavior could be related to the removal of residual ethoxy groups from samples during the water

AC C

sorption tests. As these materials were not heat-treated before these tests, a significant amount of organic residues could be present in their structures at the beginning of the water sorption tests, which is in agreement with our previous work [24]. This expressive amount of organic residues could be also related to a fast gelation rate of the sols at 90 °C, which inhibits the completion of the sol-gel reactions. As the commercial silica was probably heat-treated during its processing, this sample did not present such mass loss after the sorption test. The larger mass loss was observed for the sample prepared

10

ACCEPTED MANUSCRIPT

with a TEOS: HCl molar ratio of 1: 0.001, which is also the sample with the larger amount of organic residues because of the slow kinetics of the sol-gel reactions [24]. Figure 4 depicts the deconvolution of the FTIR spectra ascribed to the sample obtained using the aforementioned molar ratio, before and after the water sorption tests. The

RI PT

absorption bands at about 1050, 1150 and 1215 cm-1 are ascribed to silica, whereas those at 1075, 1100 and 1180 cm-1 are associated with TEOS [36]. Table 2 gives the deconvolution parameters used for fitting the spectra shown in Figure 4. One notices

SC

that after the water sorption tests the bands related to SiO2 became more noticeable in terms of both full-width half maximum (FWHM) and area, whereas those ascribed to

M AN U

TEOS became less apparent. This behavior may confirm that the removal of TEOS residues from the sample structures took place during the water sorption tests. Again, the difference observed in the samples masses before and after the water sorption tests could be related to the removal of organic residues from the samples structures (Figure

TE D

2).

The pseudo first- and second order kinetic models and an intraparticle diffusion model

EP

were used to elucidate the water adsorption mechanism on the samples examined in this study. This procedure was carried out by applying these models to the collected

AC C

adsorption data. As reported by Huang et al. [37], the knowledge of the adsorption kinetic is an important step for investigating the possibility of using an adsorbent in a particular application. The Lagergren kinetic rate equation can be written as [38]:

dq t n = k n (q e − q t ) dt

(3)

11

ACCEPTED MANUSCRIPT

where qe and qt represent, respectively, the amount of water adsorbed per unit mass of silica (mg.g-1) at equilibrium and at a time t. kn is the rate constant for the nth-order adsorption. The linear pseudo-first order model can be expressed as [39]:

(4)

RI PT

ln(q e − q t ) = lnq e − k 1t

where k1 is the rate constant of the first-order adsorption (min-1). k1 can be assessed by

SC

plotting ln(qe−qt) as a function of time. The linear pseudo-second order model can be

M AN U

written as [40]:

t 1 t = + 2 q t k 2q e q e

(5)

TE D

where k2 is the rate constant of the second-order adsorption (g.mg-1.min-1). k2 can be estimated by assessing t/qt as a function of time. The intraparticle diffusion kinetic

qt = kp t + c

(6)

AC C

EP

model is expressed as [41]:

where kp is the intraparticle diffusion rate constant (g.mg-1.min-1) and c is a constant (mg.g-1). kp can be estimated by plotting qt against t1/2. Table 3 gives the values assessed by plotting equations (4) to (6), as well as the correlation coefficients (R2) obtained from the linear fitting of these curves. One observes that the pseudo-second order kinetic model showed a better fit when compared to the pseudo-first order model since smaller values of R2 were obtained for the former. As a consequence, the water

12

ACCEPTED MANUSCRIPT

adsorption on the silica samples investigated herein is better described by the linear pseudo-second order kinetic model. According to Oliveira et al. [42], this behavior may reveal that there is a strong interaction between the tested samples and water, which is associated with either valence or covalent forces. The valence forces could be ascribed

RI PT

to the sharing of electrons between adsorbent and adsorbate, whereas the covalent ones are related to the exchange of electrons between them. From Table 3, one observes that the samples with highest water adsorption capacities (1: 4: 4: 0 and 1: 4: 4: 1) exhibited

SC

the slowest adsorption rates according to the pseudo-second order kinetic model. This behavior could be related to the large specific surface areas and mesoporous volumes

M AN U

shown by these samples. Thus, one may suggest that it takes longer times filling the pore network of these materials with water.

It can also be inferred from Table 3 that the amount of water adsorbed at equilibrium in

TE D

the sorption tests (qe,expI, qe,expII) and that estimated from the pseudo-second order kinetic model (q2) are very close. This behavior reinforces that this kinetic model is the bestfitting model for the samples investigated in this work. Although Lagergren [38] first

EP

suggested Equation (3) for the adsorption of solutes present in a liquid solution on solids, the pseudo-second order model obtained from his equation fitted well with the

AC C

sorption tests performed herein. Therefore, we believe that the Lagergren kinetic rate equation can be used for investigating the water adsorption on the samples studied herein. Indeed, similar models were already used by other authors for examining the water adsorption on solids [43].

Figure 5 displays the intraparticle diffusion model assessed using Equation (6). It can be clearly observed that the tested materials display a linear portion in their water uptake

13

ACCEPTED MANUSCRIPT

curves, followed by a plateau. From Table 3 it can also be noted that the constant c ascribed to the intraparticle diffusion model shows values different from zero. These findings reveal that the intraparticle diffusion is present in the adsorption process but it is not the only rate-limiting mechanism [44,45]. As the curves shown in Figure 5 are not

RI PT

linear over the whole time range, more than one mechanism affected the water adsorption on the tested materials [46]. Further studies are needed for investigating the

SC

adsorption mechanisms of water on the silicas studied in this work.

It is worth emphasizing that we used in this work a simple processing route for

M AN U

preparing silica-based water adsorbents. No chemical modification neither heattreatment was used for obtaining these materials, which could represent an important economic advantage in the large-scale production of these adsorbents. In addition, the sample of highest water adsorption capacity was prepared with no catalyst addition. The

TE D

absence of acidic reactants in the preparation step of these materials represents a safety advantage of this method when compared to others commonly used for obtaining solgel silica samples, besides reinforcing its simplicity and novelty. The non-use of acidic

EP

reactants also demonstrates the environmentally friendly character of this methodology. From Figure 3, one also notices that the water desorption from the tested samples was

AC C

carried out in relatively short times (below 2 h) and at temperatures as low as 90 °C. Many materials used in water adsorption applications show a slow desorption kinetics, so that the pressure swing adsorption (PSA) approach is commonly used for promoting the water desorption [47,48]. PSA can be an energy intensive and cost-ineffective technology depending on the operating conditions. Thus, the short desorption times (< 2 h) and low desorption temperatures (90 °C) shown by the materials prepared herein can be considered a strategic advantage in water sorption applications.

14

ACCEPTED MANUSCRIPT

CONCLUSIONS

In this work we successfully obtained silica samples with promising applications as

RI PT

water adsorbents. These materials were prepared by a simple processing route where no heat-treatment neither chemical modification was used before performing the water sorption tests. It was shown that a simple control of the sol pH allowed preparing silica

SC

adsorbents with adsorption capacities higher than a commercial silica commonly used as water adsorbent. Thus, this work brings an alternative approach to those reported in

M AN U

several works reported in the literature where complex and multi-step procedures are used for obtaining adsorbents with high adsorption capacities. In addition, the sample with the highest water adsorption capacity was prepared with no addition of acidic catalyst. The absence of acidic reactant in the preparation step of this material represents

TE D

an important advantage in terms of safety and environmental sustainability, besides reinforcing the simplicity and novelty of the processing route employed herein. The pseudo-second order kinetic model was the best-fitting model for the water adsorption

EP

on the samples investigated in this work. This behavior may reveal that there is a strong interaction between the prepared samples and water. It was also observed that the

AC C

intraparticle diffusion mechanism was present in the adsorption process but it was not the only rate-limiting step. Finally, the short desorption times (< 2 h) and low desorption temperatures (about 90 °C) shown by the materials prepared in this study can be considered a strategic advantage in water sorption applications.

15

ACCEPTED MANUSCRIPT ACKNOWLEDGEMENTS

The authors thank the financial support from CNPq, FAPEMIG, and PRPq-UFMG. INCT-Acqua Institute, Ilda Batista, Isabel Batista, prof. Paulo Brandão and Sérgio

RI PT

Carneiro are kindly acknowledged for the support provided in FTIR and N2 adsorption.

[1]

D. Farrusseng, Metal-Organic Frameworks: Applications from Catalysis to Gas

J. Canivet, A. Fateeva, Y. Guo, B. Coasne, D. Farrusseng, Chem. Soc. Rev. 43 (2014) 5594–5617.

H. Furukawa, F. Gándara, Y.-B. Zhang, J. Jiang, W.L. Queen, M.R. Hudson,

TE D

[3]

M AN U

Storage, Wiley-VCH, Weinheim, 2011. [2]

SC

REFERENCES

O.M. Yaghi, J. Am. Chem. Soc. 136 (2014) 4369–81. [4]

Rolando M. A Roque-Malherbe, Adsorption and Diffusion in Nanoporous

F. Ohashi, M. Maeda, K. Inukai, M. Suzuki, S. Tomura, J. Mater. Sci. 34 (1999)

AC C

[5]

EP

Materials, CRC Press, Boca Raton, 2007.

1341–1346.

[6]

K. Wörmeyer, I. Smirnova, Chem. Eng. J. 225 (2013) 350–357.

[7]

H. Chen, W. Wang, X. Wei, J. Ding, J. Yang, Adsorption 21 (2015) 67–75.

[8]

M.A. Tahat, Appl. Energy 69 (2001) 19–27.

[9]

K. Klier, J.H. Shen, A.C. Zettlemoyer, J. Phys. Chem. 77 (1973) 1458–1465.

16

ACCEPTED MANUSCRIPT [10]

K. Yamashita, H. Daiguji, J. Phys. Chem. C 117 (2013) 2084–2095.

[11]

T. Takei, A. Yamazaki, T. Watanabe, M. Chikazawa, J. Colloid Interface Sci. 188 (1997) 409–414. A. Feng, B.J. McCoy, Z.A. Munir, D.E. Cagliostro, J. Colloid Interface Sci. 180

RI PT

[12]

(1996) 276–284. [13]

S. Bocian, G. Rychlicki, M. Matyska, J. Pesek, B. Buszewski, J. Colloid Interface

M. Delle Piane, S. Vaccari, M. Corno, P. Ugliengo, J. Phys. Chem. A 118 (2014)

M AN U

[14]

SC

Sci. 416 (2014) 161–166.

5801–5807. [15]

B. Sun, A. Chakraborty, Energy 84 (2015) 296–302.

[16]

S. Saliba, P. Ruch, W. Volksen, T.P. Magbitang, G. Dubois, B. Michel,

TE D

Microporous Mesoporous Mater. 226 (2016) 221–228. S. Mitra, N. Aswin, P. Dutta, Int. J. Heat Mass Transf. 95 (2016) 853–864.

[18]

C. Brinker, G. Scherer, Sol-Gel Science: The Physics and Chemistry of Sol-Gel

EP

[17]

AC C

Processing, Academic Press, New York, 1990. [19]

T.-W. Kim, R. Ryoo, M. Kruk, K.P. Gierszal, M. Jaroniec, S. Kamiya, O.

Terasaki, J. Phys. Chem. B 108 (2004) 11480–11489.

[20]

A. Sayari, S. Hamoudi, Y. Yang, Chem. Mater. 17 (2005) 212–216.

[21]

S. Kim, J. Ida, V. V Guliants, Y.S. Lin, J. Phys. Chem. B 109 (2005) 6287–6293.

[22]

M. Kruk, L. Cao, Langmuir 23 (2007) 7247–7254.

[23]

T. Prenzel, M. Wilhelm, K. Rezwan, Chem. Eng. J. 235 (2014) 198–206.

17

ACCEPTED MANUSCRIPT [24]

M. Houmard, E.H.M. Nunes, D.C.L. Vasconcelos, G. Berthomé, J.C. Joud, M. Langlet, W.L. Vasconcelos, Appl. Surf. Sci. 289 (2014) 218–223. P. Innocenzi, J. Non. Cryst. Solids 316 (2003) 309–319.

[26]

J. Román, S. Padilla, M. Vallet-Regí, Chem. Mater. 15 (2003) 798–806.

[27]

H.Y. Jung, R.K. Gupta, E.O. Oh, Y.H. Kim, C.M. Whang, J. Non. Cryst. Solids

RI PT

[25]

SC

351 (2005) 372–379.

H. Yoshino, K. Kamiya, H. Nasu, J. Non. Cryst. Solids 126 (1990) 68–78.

[29]

D.J. Taylor, B.D. Fabes, J. Non. Cryst. Solids 147 (1992) 457–462.

[30]

M. Yu, S. Jambhrunkar, P. Thorn, J. Chen, W. Gu, C. Yu, Nanoscale 5 (2012)

M AN U

[28]

178–183.

R.M. Almeida, C.G. Pantano, J. Appl. Phys. 68 (1990) 4225–4232.

[32]

K.S.W. Sing, Pure Appl. Chem. 54 (1982) 2201–2218.

[33]

H.L. Castricum, G.G. Paradis, M.C. Mittelmeijer-Hazeleger, W. Bras, G.

TE D

[31]

EP

Eeckhaut, J.F. Vente, G. Rothenberg, J.E. Ten Elshof, Microporous Mesoporous

AC C

Mater. 185 (2014) 224–234. [34]

H.L. Castricum, H.F. Qureshi, A. Nijmeijer, L. Winnubst, J. Memb. Sci. 488

(2015) 121–128.

[35]

T. Tomita, S. Kawasaki, K. Okada, J. Porous Mater. 11 (2004) 107–115.

[36]

N. Primeau, C. Vautey, M. Langlet, Thin Solid Films 310 (1997) 47–56.

[37]

H. Huang, Y. Fan, J. Wang, H. Gao, S. Tao, Macromol. Res. 21 (2013) 726–731.

18

ACCEPTED MANUSCRIPT S. Lagergren, K. Sven. Vetenskapsakademiens, Handl. 24 (1898) 1–39.

[39]

H. Yuh-Shan, Scientometrics 59 (2004) 171–177.

[40]

Y.. Ho, G. McKay, Process Biochem. 34 (1999) 451–465.

[41]

F.-C. Wu, R.-L. Tseng, R.-S. Juang, Chem. Eng. J. 153 (2009) 1–8.

[42]

L.S. Oliveira, A.S. Franca, T.M. Alves, S.D.F. Rocha, J. Hazard. Mater. 155

RI PT

[38]

SC

(2008) 507–512.

Q. Zeng, D. Zhang, K. Li, Transp. Porous Media 109 (2015) 469–493.

[44]

N. Li, X.-M. Lei, J. Incl. Phenom. Macrocycl. Chem. 74 (2012) 167–176.

[45]

I.A.W. Tan, J.C. Chan, B.H. Hameed, L.L.P. Lim, J. Water Process Eng. 14

M AN U

[43]

(2016) 60–70.

B.H. Hameed, A.A. Rahman, J. Hazard. Mater. 160 (2008) 576–581.

[47]

P. Pruksathorn, T. Vitidsant, Korean J. Chem. Eng. 26 (2009) 1106–1111.

[48]

Y. Wang, M.D. LeVan, J. Chem. Eng. Data 54 (2009) 2839–2844.

AC C

EP

TE D

[46]

19

ACCEPTED MANUSCRIPT FIGURE CAPTIONS

- Figure 1: Typical FTIR spectrum of the samples obtained in this work.

RI PT

- Figure 2: Water sorption cycles performed in this study. Samples were identified according to the TEOS: EtOH: H2O: HCl molar ratio used in the synthesis step. The

SC

solid lines are used as a guide to the eyes only.

- Figure 3: N2 adsorption isotherms of the samples used in this work. Samples were

M AN U

identified according to the TEOS: EtOH: H2O: HCl molar ratio used in the synthesis step. The solid lines are used as a guide to the eyes only.

- Figure 4: Deconvolution of the FTIR spectra ascribed to the sample obtained using a

TE D

1 TEOS: 4 EtOH: 4 H2O: 0.001 HCl molar ratio, before and after the water sorption. The area ascribed to SiO2 and TEOS bands do not reach 100% because there is also a

EP

contribution of the band associated with silanol groups (960 cm-1).

- Figure 5: Intraparticle diffusion model assessed for the samples examined in this

AC C

study. The solid lines are used as a guide to the eyes only.

20

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 1

21

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 2

22

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 3

23

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 4

24

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 5

25

ACCEPTED MANUSCRIPT Table 1: Morphological parameters assessed from N2 adsorption. Specific surface area (m2.g-1)

Mesoporous volume (cm3.g-1)

Mean pore diameter (nm)

1: 4: 4: 0

597 ± 30

0.94 ± 0.05

6.3 ± 0.4

1: 4: 4: 0.001

587 ± 29

0.33 ± 0.02

2.2 ± 0.2

1: 4: 4: 1

771 ± 39

0.50 ± 0.03

Commercial silica

645 ± 32

0.37 ± 0.02

RI PT

Sample (TEOS: EtOH: H2O: HCl)

2.6 ± 0.2

AC C

EP

TE D

M AN U

SC

2.3 ± 0.2

26

ACCEPTED MANUSCRIPT

Table 2: Deconvolution parameters used for fitting the FTIR spectra shown in Figure 4. Assignment

1040

SiO2

1080

Sample condition

FWHM*

Area (%)

As-prepared

63

36

After water adsorption

70

43

As-prepared

53

10

TEOS After water adsorption

40

07

78

18

75

15

61

12

After water adsorption

67

13

As-prepared

47

4

After water adsorption

47

3

As-prepared

60

8

After water adsorption

60

7

As-prepared 1100

TEOS

1183

TEOS

1215

SiO2

M AN U

As-prepared SiO2

EP

TE D

* Full width at half-maximum

AC C

SC

After water adsorption 1150

RI PT

Band (cm-1)

ACCEPTED MANUSCRIPT

27

Table 3: Kinetic parameters for water adsorption on the silica samples studied in this work. qe,expI and qe,expII were obtained in the water sorption

RI PT

tests at the end of the first and second adsorption cycles, respectively. k1, q1, k2 and q2 were estimated from the pseudo-first and second order kinetic models. kp and c were assessed using the intraparticle diffusion model. Pseudo-second order

1: 4: 4: 0

577 / 567

4 × 10-3

461

0.994

1: 4: 4: 0.001

190 / 193

11 × 10-3

50

0.785

1: 4: 4: 1

473 / 468

6 × 10-3

166

0.666

Commercial silica

329 / 332

4 × 10-3

55

R2

k2 (g.mg-1.min-1) q2 (mg.g-1)

AC C

EP

0.5

R2

Intraparticle diffusion kp (g.mg-1.min-1) c (mg.g-1)

R2

1.7 × 10-5

614

0.997

31.5

6.1

0.995

6.8 × 10-4

192

1.000

23.0

4.8

0.952

5.6 × 10-5

490

0.998

33.4

7.1

0.965

1.7 × 10-4

333

0.999

28.2

5.4

0.951

TE D

k1 (min-1) q1 (mg.g-1)

M AN U

qe,expI / qe,expII (mg.g-1)

SC

Pseudo-first order

Sample (TEOS: EtOH: H2O: HCl)

ACCEPTED MANUSCRIPT

Highlights

RI PT

- Mesoporous silica adsorbents synthetized by sol-gel technology

- Size and volume of the mesopores controlled by the sol-gel parameters

SC

- Higher water adsorption for samples presenting higher pore volume

M AN U

- Water adsorption capacity about two times superior than comercial silica absorbent

AC C

EP

TE D

- Kinetics of water adsorption follow a second order kinetic model

DEMC | DEPTO. DE ENGENHARIA DE MATERIAIS E CONSTRUÇÃO Avenida Presidente Antônio Carlos . 6627 . Belo Horizonte . MG Bloco I . 3º andar . Sala 3408 [email protected] +55 (31) 3409.1964