Desorption on Two Fully Characterized

Article pubs.acs.org/jced

Water Vapor Adsorption/Desorption on Two Fully Characterized Commercial Activated Carbons Frédéric Plantier,† Kamila Marques Fernandes,† Carine Malheiro,† Bernard Delanghe,‡ and Christelle Miqueu*,† †

CNRS, TOTAL - UMR 5150 − LFC-R − Laboratoire des Fluides Complexes et leurs Réservoirs, Université Pau et Pays Adour, BP 1155 PAU, F-64013, France ‡ Université Pau et Pays Adour, UFR Sciences et Techniques Côte Basque, 1 allée du Parc Montaury, 64600, ANGLET, France ABSTRACT: The adsorption/desorption isotherms of water vapor are measured on two commercial activated carbons (ACs) at three temperatures (353, 369, and 386 K) by means of a magnetic suspension balance. These two ACs are fully characterized in terms of textural properties (specific surface area and pore size distribution) and surface active groups concentrations by Boehm titration. The effect of temperature and textural properties of the ACs is analyzed both on the adsorption/ desorption hysteresis and on the total adsorptive capacity. The Do and Do model and two modified versions of this latter are compared for the modeling of all the experimental isotherms.

1. INTRODUCTION Activated carbons (ACs) are widely used adsorbents for removing pollutants from streams in industrial processes because of their powerful and various adsorption properties. The presence of water in ACs is known to have an important negative impact both on the capacity and the selectivity for the removal of organic or inorganic contaminants in the industrial ACs.1−3 The adsorption mechanism of water on ACs is quite different from that of simpler fluids such as nitrogen, alkanes, and so forth.4 While these simple compounds (nitrogen, argon, methane, and so forth) present most usually type I isotherms (IUPAC classification) on ACs, water adsorption isotherm is Sshape, similar to type IV or V depending on the density of active sites. The difference in these behaviors is classically assigned both to the weak water−carbon interactions and to the strong hydrogen bonds that lead to water clusters formation (see the two interesting reviews of water behavior in porous carbons by Brennan et al.5 and Mowla et al.6). Several investigations have been carried out on the measurement and the understanding of water adsorption on graphitic microporous materials (AC, ACF, SWNT, and so forth)7−11 sometimes with a complete surface chemistry and pore structure analysis.12,13 However, the experiments have always been performed at a unique temperature, close to the ambient one, except in the work of Ohba et al.14 and Nastaj and Aleksandrzak.15 In the present study, the adsorption/ desorption of water vapor on two commercial microporous ACs is measured at three temperatures (353, 369, and 386 K) by means of a magnetic suspension balance in order to investigate the influence of temperature on the adsorptive properties. The two ACs are fully characterized in terms of © 2015 American Chemical Society

structural properties and surface activity. The difference in their pore size distribution (PSD), more or less extended, allow for the discussion of the impact of PSD on water adsorption. These two commercial ACs are classically used in industrial processes. Carboxen 1012 (from Supelco) is a highly activated, inert carbon molecular sieve (CMS). It is used effectively for aqueous phase adsorption of organic compounds, or for air sampling of C4−C6 compounds. Ecosorb (from Jacobi) is an AC that is suitable for the removal of a wide range of organic pollutants in the vapor phase. The predominance of micropores in Ecosorb ensures effective removal of low molecular weight contaminants, present in low concentrations. Few years ago, Do and Do16 developed a model (DD) able to explain the role of active sites and microscopic structure of AC in the adsorption of water. This model is based on a twostep mechanism: a water cluster formation on active sites followed by the penetration of the clusters into the micropores. Since its introduction, several modifications concerning the size of clusters, the adsorption on actives sites, and so forth have been proposed for this model. In their critical review of analytic approaches for the modeling of water adsorption on carbons, Furmaniak et al.17 have shown the efficiency of DD derived models on water adsorption on ACs. In the present study, we test the DD model and its two principal modifications18,19 on the experimental isotherms of the two ACs. The paper is organized as follows. First, the textural properties and surface group concentrations are provided for the two ACs. Then, after a presentation of the experimental Received: September 11, 2015 Accepted: November 30, 2015 Published: December 11, 2015 622

DOI: 10.1021/acs.jced.5b00775 J. Chem. Eng. Data 2016, 61, 622−627

Journal of Chemical & Engineering Data

Article

The BET surface areas are estimated to 1800 and 1200 m2· g for the AC Carboxen-1012 and AC Ecosorb, respectively. It can be seen from Figures 1 and 2 that these two ACs are purely microporous as there is no hysteresis in their adsorption/ desorption cycle, characteristics of the mesoporosity. Moreover, the AC Carboxen-1012 has on the one hand a more important adsorptive capacity due to its higher specific area than AC Ecosorb and on the other hand a narrower PSD (in other words its PSD is more monodisperse) centered around 1.3 nm. The identification and quantification of AC surface functions (number of oxygen-containing groups) can be performed by the Boehm’s method.22 Upon activation, the oxidized functions (active sites) are formed on the surface of the AC. These functions can be of three types: acidic, basic, or neutral. Nevertheless, a majority of acid groups, mainly carboxyls, lactones and phenols, predominates. The Boehm’s method is based on an acidimetric titration. NaOH neutralizes acidic groups (carboxyls, lactones, and phenols) and the dosing of the basic functions is carried out by neutralization with hydrogen chloride (HCl). The titration procedure is as follows: (i) an amount of AC is placed under stirring in 2 L of distilled water for 1 h until the pH remains constant. Then the sample is dried for 24 h. (ii) Two hundred milligrams of AC is introduced in different beakers containing NaOH or HCl at 0.02 mol·L−1. The beakers are sealed, stirred, and thermostated at 298 K during 48 h and then filtered. (iii) Each filtrate is titrated with HCl or NaOH (back assay) depending on the original titrant. (iv) Titration is repeated three times. The results are summarized in Table 1.

apparatus and the different models the adsorption and desorption isotherms of water are measured, analyzed, and finally compared with the results from the different models.

−1

2. EXPERIMENTAL SECTION A. Material. Two kinds of activated carbons (ACs) were used as adsorbents in this study. The Ecosorb AC was kindly supplied by JACOBI and the Carboxen-1012 (a carbon molecular sieve) was purchased from SUPELCO. The water used to generate water vapor is so-called superpure. In other words, it has undergone various processes to be gradually cleansed of foreign ions. This treatment can be broken down successively into (i) passage of the fluid through a column of active charcoal followed by reverse osmosis, (ii) passage through another column of active charcoal, followed by filtering through two columns of mixed beds (made up of a set of cationic and anionic resins), and finally (iii) passage through a cartridge containing active charcoal and mixed beds. When water emerges from this chain of purification, its electrical resistivity reaches 18 Mohm·cm and therefore corresponds to a degree of purity such that the concentration of foreign ions does not exceed 10−10 mol/L. B. Adsorbent Characteristics. The main characteristics of the two ACs were determined on the Micromeritics ASAP 2020 apparatus. The specific surface area, pore volume, and PSD were determined by a low-pressure nitrogen adsorption isotherm at 77 K (from 5 × 10−7 to 0.99 P/P0 in relative pressure range). Before analysis, the samples were first degassed at 473 K during 24 h under vacuum. The BET surface areas (SBET) were deduced from the nitrogen adsorption isotherms applying the BET method20 and the PSDs were evaluated by a homemade filling pressure model.21 Nitrogen adsorption isotherms and PSD are plotted for the two ACs in Figures 1 and 2.

Table 1. Surface Groups Concentrations of ACs by Boehm’s Titration

AC Ecosorb Carboxen1012

total acidity

total basicity

total surface groups concentration S0

10−4 molNaOH·g−1

10−4 molHCl· g−1

10−4 mol·g−1

7.9 6.1

3.1 4.4

11.0 10.5

C. Experimental Apparatus. The adsorption measurements were performed by means of a magnetic suspension balance (Rubotherm, Germany). This experimental device and the measuring principle have been described previously.23 The main part of the apparatus is composed of an adsorption chamber where the adsorbate is introduced at the experimental temperature and pressure conditions. During adsorption isotherms measurements, the balance is kept at a constant temperature by two heating jackets, one for the adsorption chamber and another for the suspension coupling. The temperature is measured by a platinum resistance temperature sensor (Pt100) placed directly in the adsorption chamber with an accuracy of ±0.1 K. The overall uncertainty of the amount adsorbed (due to helium calibration procedure, gas density value, stability of balance signal) is determined to be lower than 1% over the entire range investigated in this study. In order to study the adsorption of water vapor, a homemade steam generator has been implemented upstream of the adsorption chamber (Figure 3). This device is made of a pure liquid water tank and a combination of classical and micrometric valves to ensure the introduction of very small amounts of water vapor in the adsorption chamber. To prevent

Figure 1. Nitrogen adsorption isotherms at 77 K for the two ACs: solid line, Carboxen-1012; dashed line, Ecosorb.

Figure 2. Pore size distribution of the two ACS: solid line, Carboxen1012; dashed line, Ecosorb.

623



Article

Neitsch et al.18 generalized the Do and Do model by considering, as shown by several experimental and molecular simulations investigations, that the size of the water clusters is variable according to the density and distribution of primary sites and also the pore widths. Thus, in the model of Neitsch et al. the pentamers of the original DD model are replaced by mmers and the adsorption isotherm equation (DDN) becomes m+1

n = S0

K f ∑i = 1 ihi m+1

1 + K f ∑i = 1 hi

+ nμs

Kμhm + 1 Kμhm + 1 + h

(2)

where m is the size of the water clusters. Finally, following the experimental evidence that for activated carbons, surface active sites are of different kinds and supposing that the strong adsorption of a water molecule on a site is not solely different but also independent of the bonding between the next water molecules, Furmaniak et al.19,28 proposed the heterogeneous Do and Do model (HDD) for water adsorption on carbons: Figure 3. Schematic diagram of the experimental apparatus.

n=

∑ j

(3)

condensation of water in the whole circuit this auxiliary circuit, the pressure lines, and the upper and lower parts of the balance are heated by means of a heating hot wire to a temperature higher than that imposed in the magnetic balance. By use of this apparatus one can obtain water vapor in the tank at saturation pressure imposed by the temperature of the heating hot wire system. A pressure sensor from Wika (with a sensitivity of 0.01% on the full scale) serves first to control the vapor pressure in the water tank and then to control the injection pressure of steam in the adsorption chamber. Several thermocouples are installed along the additional circuit to verify that the temperature remains much higher than that in the measuring cell.

where nmL,j is the surface concentration of the jth surface group type (mol/g), KL,j is the Langmuir constant (as the adsorption on primary sites can be described by the Langmuir adsorption isotherm), N is the maximum number of water molecules adsorbed on the surface sites (N ≥ m + 1), and Kp is the constant of reaction due to hydrogen bonding mechanism between water molecules in a cluster. The other parameters of eq 3 are the same as previously defined.

4. RESULTS AND DISCUSSION Prior to all isotherm adsorption measurements, the ACs samples are placed in the crucible inside the measuring cell. These samples are purified under vacuum ( 0.9. Moreover, one can see in Figure 6b that it is not possible to accurately reproduce the adsorption of water on active sites (P/P0 < 0.4) both with the DD and DDN equations because a unique equilibrium constant is used in these models to describe both the active site/water and water/water interactions. However, thanks to its improvements the HDD equation accurately models all the stages of water adsorption on this AC. The same previous conclusions can be drawn for Carboxen1012 (see Figure 7) with even a stronger effect due to the sharp filling of the micropores. Hence, the slope of this filling, which is controlled by Kμ, can only be shown by a high value for Kμ. This value leads to a filling of micropores at very low relative pressure (P/P0 = 0.1) with the original DD equation. As can be seen in Figure 7, it is not possible to describe the adsorption of water on a microporous adsorbent with such a narrow PSD

Table 3. Values of the Fitted Parameters of the DD, DDN, and HDD Models Ecosorb S0 (mmol/g) nμs (mmol/g) m Kf Kμ S0 (mmol/g) nμs (mmol/g) m Kf Kμ NmL,1 (mmol/g) nμs (mmol/g) m N KL,1 Kp Kμ

Carboxen-1012

DD isotherm (eq 1) 1.10 1.05 0.023 0.0255 5 5 0.5 1 3.5 15000 DDN isotherm (eq 2) 1.10 1.05 0.023 0.0255 9 24 0.2 1 15 15000 HDD isotherm (eq 3) 1.10 1.05 0.023 0.0255 9 24 10 25 10−5 10−5 0.5 0.5 15 15000 626



Article

(11) Furmaniak, S.; Wiśniewski, M.; Werengowska-Ciećwierz, K.; Terzyk, A. P.; Hata, K.; Gauden, P. a.; Kowalczyk, P.; Szybowicz, M. Water at Curved Carbon Surface: Mechanisms of Adsorption Revealed by First Calorimetric Study. J. Phys. Chem. C 2015, 119, 2703−2715. (12) Li, X.; Li, Z. Equilibrium and Do−Do Model Fitting of Water Adsorption on Four Commercial Activated Carbons with Different Surface Chemistry and Pore Structure. J. Chem. Eng. Data 2010, 55, 5729−5732. (13) Li, X.; Li, Z. Adsorption of Water Vapor onto and Its Electrothermal Desorption from Activated Carbons with Different Electric Conductivities. Sep. Purif. Technol. 2012, 85, 77−82. (14) Ohba, T.; Kanoh, H.; Kaneko, K. Cluster-Growth-Induced Water Adsorption in Hydrophobic Carbon Nanopores. J. Phys. Chem. B 2004, 108, 14964−14969. (15) Nastaj, J.; Aleksandrzak, T. Adsorption Isotherms of Water, Propan-2-Ol, and Methylbenzene Vapors on Grade 03 Silica Gel, Sorbonorit 4 Activated Carbon, and HiSiv 3000 Zeolite. J. Chem. Eng. Data 2013, 58, 2629−2641. (16) Do, D. D.; Do, H. D. Model for Water Adsorption in Activated Carbon. Carbon 2000, 38, 767−773. (17) Furmaniak, S.; Gauden, P. a; Terzyk, A. P.; Rychlicki, G. Water Adsorption on Carbons–Critical Review of the Most Popular Analytical Approaches. Adv. Colloid Interface Sci. 2008, 137, 82−143. (18) Neitsch, M.; Heschel, W.; Suckow, M. Water Vapor Adsorption by Activated Carbon: A Modification to the Isotherm Model of Do and Do [6]. Carbon 2001, 39, 1437−1438. (19) Furmaniak, S.; Gauden, P. a; Terzyk, A. P.; Rychlicki, G.; Wesołowski, R. P.; Kowalczyk, P. Heterogeneous Do-Do Model of Water Adsorption on Carbons. J. Colloid Interface Sci. 2005, 290, 1−13. (20) Brunauer, S.; Emmett, P. H.; Teller, E. Adsorption of Gases in Multimolecular Layers. J. Am. Chem. Soc. 1938, 60, 309−319. (21) Malheiro, C.; Mendiboure, B.; Plantier, F.; Guatarbes, B.; Miqueu, C. An Accurate Model for the Filling Pressure of Carbon Slitlike Micropores. Fluid Phase Equilib. 2015, 404, 118−123. (22) Boehm, H. P. Some Aspects of the Surface Chemistry of Carbon Blacks and Other Carbons. Carbon 1994, 32, 759−769. (23) Khaddour, F.; Knorst-Fouran, A.; Plantier, F.; Piñeiro, M. M.; Mendiboure, B.; Miqueu, C. A Fully Consistent Experimental and Molecular Simulation Study of Methane Adsorption on Activated Carbon. Adsorption 2014, 20, 649−656. (24) Kaneko, K.; Katori, T.; Shimizu, K.; Shindo, N.; Maeda, T. Changes in the Molecular Adsorption Properties of Pitch-Based Activated Carbon Fibres by Air Oxidation. J. Chem. Soc., Faraday Trans. 1992, 88, 1305−1309. (25) Iiyama, T.; Kobayashi, Y.; Kaneko, K.; Ozeki, S. In Situ SmallAngle X-Ray Scattering Study of Cluster Formation in Carbon Micropores. Colloids Surf., A 2004, 241, 207−213. (26) Iiyama, T.; Ruike, M.; Kaneko, K. Structural Mechanism of Water Adsorption in Hydrophobic Micropores from in Situ Small Angle X-Ray Scattering. Chem. Phys. Lett. 2000, 331, 359−364. (27) Kaneko, K.; Hanzawa, Y.; Iiyama, T.; Kanda, T.; Suzuki, T. Cluster-Mediated Water Adsorption on Carbon Nanopores. Adsorption 1999, 5, 7−13. (28) Furmaniak, S.; Gauden, P. A.; Terzyk, A. P.; Rychlicki, G. Water Adsorption on Carbons  Critical Review of the Most Popular Analytical Approaches. Adv. Colloid Interface Sci. 2008, 137, 82−143. (29) Lagorsse, S.; Campo, M. C.; Magalhães, F. D.; Mendes, A. Water Adsorption on Carbon Molecular Sieve Membranes: Experimental Data and Isotherm Model. Carbon 2005, 43, 2769−2779. (30) Horikawa, T.; Sakao, N.; Do, D. D. Effects of Temperature on Water Adsorption on Controlled Microporous and Mesoporous Carbonaceous Solids. Carbon 2013, 56, 183−192.

with the original DD model. As for the other AC, solely the HDD equation is able to model the whole water adsorption isotherm on Carboxen-1012 from the lowest pressures to saturation.

5. CONCLUSIONS The adsorption/desorption isotherms of water vapor were measured on two fully characterized ACs at different temperatures. The effect of temperature is clearly shown on the adsorption/desorption hysteresis but is negligible on the total adsorptive capacity of the AC. The different textural properties (specific surface area, pore volume, and pore size distribution) of the two ACs have an evident impact on both the slope of the isotherm corresponding to the adsorption in micropores and the total capacity of adsorption of the AC. The original DD equation and two of its modifications were used to model the measured isotherms. All the isotherms could only be well-fitted by the version of Do and Do model improved by Furmaniak et al. (HDD).

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +33-5-59-5744-15. Fax: +33-5-59-57-44-09. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Gong, R.; Keener, T. C. A Qualitative Analysis of the Effects of Water Vapor on Multi-Component Vapor-Phase Carbon Adsorption. Air Waste 1993, 43, 864−872. (2) Rodríguez-Mirasol, J.; Bedia, J.; Cordero, T.; Rodríguez, J. Influence of Water Vapor on the Adsorption of VOCs on LigninBased Activated Carbons. Sep. Sci. Technol. 2005, 40, 3113−3135. (3) Kaplan, D.; Nir, I.; Shmueli, L. Effects of High Relative Humidity on the Dynamic Adsorption of Dimethyl Methylphosphonate (DMMP) on Activated Carbon. Carbon 2006, 44, 3247−3254. (4) Ohba, T.; Kanoh, H.; Kaneko, K. Affinity Transformation from Hydrophilicity to Hydrophobicity of Water Molecules on the Basis of Adsorption of Water in Graphitic Nanopores. J. Am. Chem. Soc. 2004, 126, 1560−1562. (5) Brennan, J. K.; Bandosz, T. J.; Thomson, K. T.; Gubbins, K. E. Water in Porous Carbons. Colloids Surf., A 2001, 187−188, 539−568. (6) Mowla, D.; Do, D. D.; Kaneko, K. Adsorption of Water Vapor on Activated Carbon: A Brief Overview; Chem. Phys. Carbon; University of Queensland: Brisbane, QLD, Australia, 2003; pp 229−262. (7) Mccallum, C. L.; Bandosz, T. J.; Mcgrother, S. C.; Mu, E. a. A Molecular Model for Adsorption of Water on Activated Carbon: Comparison of Simulation and Experiment. Langmuir 1999, 15, 533− 544. (8) Sullivan, P. D.; Stone, B. R.; Hashisho, Z.; Rood, M. J. Water Adsorption with Hysteresis Effect onto Microporous Activated Carbon Fabrics. Adsorption 2007, 13, 173−189. (9) Kim, P.; Zheng, Y.; Agnihotri, S. Adsorption Equilibrium and Kinetics of Water Vapor in Carbon Nanotubes and Its Comparison with Activated Carbon. Ind. Eng. Chem. Res. 2008, 47, 3170−3178. (10) Wang, H.; Kleinhammes, A.; McNicholas, T. P.; Liu, J.; Wu, Y. Water Adsorption in Nanoporous Carbon Characterized by in Situ NMR: Measurements of Pore Size and Pore Size Distribution. J. Phys. Chem. C 2014, 118, 8474−8480. 627


Desorption on Two Fully Characterized

Recommend Documents