Combining Nitrogen, Argon, and Water Adsorption for Advanced

Article pubs.acs.org/Langmuir

Combining Nitrogen, Argon, and Water Adsorption for Advanced Characterization of Ordered Mesoporous Carbons (CMKs) and Periodic Mesoporous Organosilicas (PMOs) Matthias Thommes,*,† Jürgen Morell,‡ Katie A. Cychosz,† and Michael Fröba*,‡ †

Quantachrome Instruments, 1900 Corporate Drive, Boynton Beach, Florida 33426, United States Institute of Inorganic and Applied Chemistry, University of Hamburg, Martin-Luther-King-Platz 6, 20146 Hamburg, Germany

‡

S Supporting Information *

ABSTRACT: Ordered mesoporous CMK carbons and periodic mesoporous organosilica (PMO) materials have been characterized by combining nitrogen (77.4 K) and argon (87.3 K) adsorption with recently developed quenched solid density functional theory (QSDFT). Systematic, high-resolution water adsorption experiments have been performed in the temperature range from 298 to 318 K in order to ascertain the effect of surface chemistry (using periodic mesoporous organosilicas (PMOs) of given pore size) and pore size/pore geometry (using CMK-3, CMK-8 carbons) on the adsorption, pore filling, condensation and hysteresis behavior. These data reveal how the interplay between confined geometry effects and the strength of the adsorption forces influence the adsorption, wetting, and phase behavior of pore fluids. Further, our results indicate that water adsorption is quite sensitive to both small changes in pore structure and surface chemistry, showing the potential of water adsorption as a powerful complementary tool for the characterization of nanoporous solids.

1. INTRODUCTION During recent years, major progress has been achieved in understanding the adsorption and phase behavior in ordered micro- and mesoporous materials with simple pore geometries.1−3 This has led to major advances in structural characterization by physical adsorption, also because of the development and availability of advanced theoretical approaches based on statistical mechanics (e.g., density functional theory, molecular simulation). The application of methods based on density functional theory (DFT) allows one to describe the adsorption and phase behavior of fluids in pores at a molecular level and to obtain an accurate pore size distribution over the complete micro- and mesopore range.4−7 Recently, various approaches have been suggested to account for the effect of roughness;8−12 e.g., quenched solid density functional theory (QSDFT) quantitatively accounts for the surface geometrical inhomogeneity in terms of a roughness parameter.8,10 However, the effects of geometrical and chemical heterogeneities of the pore walls on the sorption, phase, and wetting behaviors of fluids in porous particles are still under investigation. Within this context, the potential use of water as a probe for surface chemistry and pore structure characterization of nanoporous carbon materials has led to a lot of interest.13−36 The use of water adsorption is attractive because (i) it can be performed at room temperature, (ii) water has a very small kinetic diameter (i.e., 0.28 nm) which allows it to enter pores even smaller than the ones accessible to carbon dioxide or nitrogen,37 and (iii) it is sensitive to surface chemistry. On the other hand, the interpretation of water © 2013 American Chemical Society

adsorption data is not straightforward, mainly because of the competing effects of pore structure and surface chemistry on the water adsorption isotherms and the fact that the underlying mechanism of water adsorption in nanoporous carbons is still under investigation. Hence, in order to shed more light onto the potential of water adsorption for textural and surface characterization, we studied water adsorption in selected ordered mesoporous materials such as CMK carbon as well as in periodic mesoporous organosilicas (PMO). A detailed pore surface and pore structural analysis was performed based on nitrogen (77.4 K) and argon (87.3 K) adsorption coupled with QSDFT and NLDFT pore size analysis. To study the effect of pore structure on water adsorption, we chose CMK-1, CMK-3, and CMK-8, which have similar surface chemistry; i.e., they exhibit graphitic-like pore walls but differ in pore width and pore geometry (e.g., slit-like pores in the case of CMK-1 and CMK-8 and cylindrical-like pores for CMK-3). These advanced carbon materials are synthesized using mesoporous silica molecular sieves such as MCM-48 (CMK-1), SBA-15 (CMK-3), and KIT6 (CMK-8) as a matrix.38−40 The highly ordered mesoporous carbons obtained in this way can be considered as inverse replicas of the parent mesoporous silica materials exhibiting ordered mesoporosity. PMO materials have ordered pores with well-defined pore geometries and tailorable surface chemistry and are comprised Received: July 25, 2013 Revised: October 7, 2013 Published: November 4, 2013 14893

dx.doi.org/10.1021/la402832b | Langmuir 2013, 29, 14893−14902

Langmuir

Article

QSDFT methods (Figure 2a). A slit-pore model was chosen because the removal of the KIT-6 template during the CMK-8

of a silica network with organic bridges. To assess the effect of surface chemistry on water adsorption, we performed experiments on two PMO materials41−44 with essentially identical pore size (benzene PMO and divinylbenzene PMO)45−47 but differing surface chemistry. Systematic water sorption experiments on these carbon materials were performed in the temperature range from 298 to 318 K. Comparison between nitrogen/argon and water adsorption isotherms provides complementary structural information with regard to the chemical nature of the pore walls. Our results illuminate the potential of water as a complementary probe molecule in physical adsorption characterization.

2. EXPERIMENTAL SECTION CMK-1 and CMK-3 were synthesized according to the methods given by Ryoo38 and Huwe,48 CMK-8 according to Kleitz,40 and the PMOs by Fröba et al.45,46 High-resolution nitrogen (77.4 K) and argon (87.3 K) isotherm measurements were performed using an Autosorb-1 MP sorption instrument (Quantachrome Instruments, Boynton Beach, FL) over a relative pressure range from 10−7 to 1. Water adsorption experiments were performed in the temperature range from 298 to 318 K with both a volumetric (manometric) sorption analyzer (Hydrosorb1000, Quantachrome Instruments, Boynton Beach, FL) and a gravimetric sorption analyzer (Aquadyne, Quantachrome Instruments, Boynton Beach, FL). Errors in adsorbed amounts and pressures are within the size of the symbols. Water adsorption experiments were always performed on a fresh sample from the same homogeneous batch that was used for nitrogen and argon adsorption experiments. Before each sorption measurement the samples were outgassed for 12 h under turbomolecular pump vacuum at elevated temperatures (393 K for PMOs and 423 K for CMKs) and then weighed under an inert atmosphere.

3. RESULTS AND DISCUSSION 3.1. Effect of Pore Structure on Water Adsorption: CMK Carbons. 3.1.1. CMK-8. Figure 1 shows the nitrogen

Figure 2. (a) CMK-8 nitrogen (77.4 K) pore size distribution comparison of NLDFT and QSDFT methods. (b) CMK-8 nitrogen (77.4 K) isotherm plotted semilogarithmically together with the NLDFT and QSDFT theoretical isotherms.

synthesis process leads to pores that, similar to the case of CMK-1, are most closely modeled using slit pores.10 Figure 2b shows the theoretical fit of both the NLDFT and QSDFT methods to the experimental data. The QSDFT method takes into account the heterogeneity and surface roughness of the carbon and allows one to obtain a more reliable pore size distribution for these CMKs as compared to the NLDFT method.8,10 This is demonstrated in Figure 2b, which displays the almost perfect fit of the theoretical QSDFT isotherm to the experimental data. In contrast to QSDFT, the NLDFT isotherm reveals layering transitions in the relative pressure ranges from 10−5 to 10−3 and a less pronounced transition at a relative pressure of 0.1, which are expected for adsorption on perfectly smooth and chemically homogeneous surfaces. The mismatch between theoretical assumptions and experiment leads to the two artificial minima in the NLDFT pore size distribution (at ca. 1 and 2 nm), which are not observed in the QSDFT pore size distribution. The QSDFT pore size calculation indicates a narrow distribution of mesopores centered around 3 nm as well as some microporosity which are also both observed with NLDFT; hence, with the exception of the mentioned two artificial minima, the pore size distributions from NLDFT and QSDFT are in good agreement. Additionally, Figure 3 shows the QSDFT pore size distributions

Figure 1. Nitrogen (77.4 K) and argon (87.3 K) isotherms for CMK8.

(77.4 K) and argon (87.3 K) isotherms obtained for CMK-8. Both adsorption isotherms reveal pore condensation accompanied by hysteresis in the relative pressure (P/P0) range from 0.45 to 0.6, indicating pore filling of the primary mesopores. In both cases, secondary mesoporosity is also observed, which might include interparticle voids. From the nitrogen data we obtain a BET area of 1373 m2/g (calculated in the relative pressure range from 0.03 to 0.15). Pore size distributions were calculated from the nitrogen and argon adsorption isotherms by assuming a slit-like pore geometry and using both NLDFT and 14894


Langmuir

Article

low pressure adsorbs in layers along the pore walls, water adsorption happens around active sites within the material where association of the water molecules can occur, followed by further adsorption through the development of hydrogen bond networks that permeate from these active sites.21 Recent work highlights the existence of long-lived metastable states of adsorbed water clusters as a cause for hysteresis.33,34 As a consequence of the relatively weak water−carbon interactions, the relative pressure of micro- and mesopore filling is shifted significantly to higher relative pressures (mesopore filling occurs at a relative pressure around 0.875) as compared to nitrogen adsorption at 77.4 K, which corresponds to a complete wetting situation (capillary condensation into the CMK-8 mesopores accompanied by hysteresis occurs in the relative pressure range 0.45−0.6). In fact, water cannot fill the larger mesopores of the secondary mesoporosity, which is filled by nitrogen in the relative pressure range >0.8. 3.1.2. CMK-1. Nitrogen and water adsorption isotherms were measured on CMK-1 and activated carbon fiber (ACF-10), and the results are shown in Figures 5 and 6. From the nitrogen

calculated from the nitrogen and argon isotherms and good agreement is found between the two.

Figure 3. CMK-8 pore size distributions from nitrogen (77.4 K) and argon (87.3 K) QSDFT methods.

Figure 4 shows an overlay of water adsorption isotherms obtained at 298 K on CMK-8 with the manometric

Figure 5. Comparison of nitrogen (77.4 K) and water (298 K) isotherms on CMK-1. Figure 4. Nitrogen (77.4 K) isotherm and water (298 K) isotherms measured gravimetrically and volumetrically on CMK-8.

data, we obtain a BET area of 1411 m2/g (calculated in the relative pressure range from 0.07 to 0.25). Pore size analysis was performed by applying QSDFT methods assuming nitrogen adsorption in slit-pore carbons (i.e., for both ACF10 and CMK-1, a slit-pore geometry can be assumed). Both of the isotherms behave in a similar fashion as discussed above. Two steps are observed in the water adsorption isotherms for CMK-8 and CMK-1, one beginning around relative pressure of 0.4 that is attributed to the disordered micropores found in the CMK material and another steep step around P/P0 = 0.8 that is attributed to filling or partial filling of the mesopores. To prove that the first step corresponds to filling of the micropores, nitrogen and water adsorption isotherms measured for CMK-1 were compared to isotherms for ACF-10 (Figure 6). The nitrogen QSDFT pore size distribution plots (Figure 6b) reveal ACF-10 is comprised solely of micropores, and the identical low-pressure adsorption behavior in the water isotherms for the two materials (P/P0 < 0.4) indicates that the surface chemistry is comparable.28 The similarity of the surface chemistry of ACF10 and CMK-1 carbon is also evident by comparing a hydrophilicity index calculated according to the description given in ref 49 (obtained from adsorbed amounts at relative pressures P/P0 = 0.95). Briefly, the hydrophilicity index involves the comparison of the adsorption isotherms of an

(volumetric) technique (Hydrosorb) and the gravimetric technique (Aquadyne) together with nitrogen adsorption (at 77.4 K) on the same sample. The agreement between the two water adsorption isotherms is indeed remarkable and demonstrates the high accuracy of the water adsorption data. The interesting water adsorption isotherm (type V according to the IUPAC classification) reveals three separate parts: (i) a region where almost no adsorption occurs which extends up to a relative pressure of 0.4, (ii) a step which is associated with the filling of the disordered micropores (discussed in more detail in section 3.1.2), and (iii) a second step at relative pressures >0.8 associated with the filling of the mesopores (pore width ∼3 nm). Both micropore and mesopore filling are accompanied by hysteresis which, as in the case of hydrophobic active carbons, might be caused by different mechanisms of adsorption/pore filling and desorption (involving the formation of water clusters26,33,34). This different mechanism is due to the interplay between very weak, attractive water−carbon interactions coupled with surface heterogeneity and confined geometry effects (as discussed in the literature).26 Indeed, experiments and molecular models of water adsorption on activated carbon show that unlike for nitrogen which even at 14895


Langmuir

Article

Water adsorption and pore filling of the ACF-10 micropores occur in the relative pressure range from 0.4 to 0.7, which is identical with the first step in the CMK-1 adsorption isotherm, indicating that CMK-1 micropore filling is associated with the first step in the CMK-1 water adsorption isotherm. Hence, the comparison of water adsorption between ACF-10 and CMK-1 clearly indicates that the first step in the water adsorption isotherm is micropore filling. 3.1.3. CMK-3. One additional CMK, CMK-3, was also studied using both nitrogen and water adsorption. In this case, two different CMK-3 materials were used to illustrate the effect of small changes in pore size on water adsorption isotherms. The nitrogen isotherms for the two CMK-3’s are shown in Figure 7a, and both are type IV isotherms where condensation begins around a relative pressure of 0.4 and a small hysteresis loop is observed. From the nitrogen data, we obtain BET areas of 1205 m2/g (calculated in the relative pressure range from 0.08 to 0.3) for CMK-3 (318) and 1423 m2/g (calculated in the relative pressure range from 0.08 to 0.2) for CMK-3 (348). In order to analyze the pore structure and obtain a pore size distribution, a recently developed cylindrical pore QSDFT method was applied to the nitrogen data (Figure 7b). A cylindrical pore model was found to be applicable to describe the pore structure in CMK-3 carbon.50 The pore size distributions indicate these two CMK-3’s have identical micropore size distributions (centered around 1.5 nm) but slightly different mesopore sizes, where CMK-3 (318) has a mesopore distribution centered around 4 nm, while CMK-3 (348) has a mesopore distribution centered around 5 nm. In order to compare the surface chemistry in the pores of these two CMK-3 materials, we calculated again a hydrophilicity index (obtained from adsorbed amounts at relative pressures P/ P0 = 0.95) according to the description in ref 49, and it was found to be comparable for the two materials, 0.78 for CMK-3 (318) and 0.74 for CMK-3 (348). This indicates that the surface chemistries of these two carbons are nearly identical. Hence, comparing water adsorption in these two materials allows one to solely assess the influence of different pore size on water adsorption. Figure 7c shows the corresponding water isotherms for the two materials. Overlapping isotherms are observed up to a relative pressure of 0.7, corresponding to filling of the identical micropores in the two materials. However, above a relative pressure of 0.7, the isotherms deviate from each other, demonstrating that water adsorption is sensitive to small changes in mesopore size. The presented data reveal that water adsorption is sensitive to small differences in pore size. Furthermore, in the case of hydrophobic micro- and mesoporous carbons, the shape of the water adsorption isotherms (compared to the nitrogen isotherms) clearly reflects the essentially bimodal pore structure of these CMK carbons as observed in the QSDFT pore size distributions. The water adsorption isotherms clearly reveal a two-step isotherm indicative of the micro- and mesopores in the material. Genuine water adsorption hysteresis associated with the filling of both micropores and narrow mesopores as observed for hydrophobic activated carbons and ordered mesoporous CMK carbons (in a pore size range where nitrogen and argon isotherms are reversible) may also offer potential for pore network characterization for micro-mesopore carbon networks, but more work, such as scanning hysteresis experiments, is underway to explore in greater detail how pore structure and pore network characteristics affect water adsorption hysteresis.

Figure 6. (a) Nitrogen (77.4 K) isotherms on CMK-1 and ACF-10. (b) QSDFT pore size distributions calculated from the nitrogen (77.4 K) isotherms for CMK-1 and ACF-10. (c) Water (298 K) isotherms on CMK-1 and ACF-10.

adsorptive that is sensitive to surface chemistry and does not completely wet the adsorbent surface (i.e., water) with an adsorptive that does completely wet the surface (such as nitrogen or argon at their boiling temperatures) at a given relative pressure (here P/P0 = 0.95) The greater the deviation from 1, where 1 indicates complete pore filling with water, the greater the surface hydrophobicity. In this case, the hydrophilicity indices were found to be comparable for the two materials, 0.64 for ACF-10 and 0.62 for CMK-1, indicating that the surface chemistries of these two carbons are indeed nearly identical. 14896


Langmuir

Article

pore walls, the position of the adsorption branch (in relative pressure P/P0) of the water isotherms in CMK carbons is not appreciably affected by temperature. On the other hand, the position of the desorption branch shifts to higher relative pressures with increasing temperature similar as one observes for wetting adsorbates such as nitrogen and argon at their respective boiling temperatures. This may indicate that the mechanism of water desorption from the CMK mesopores is similar to capillary evaporation of nitrogen or argon (i.e., by a receding meniscus1). In fact, the temperature dependence of the evaporation pressures is in line with the prediction of the Kelvin equation as demonstrated in Figure 9 where the water evaporation pressures from the CMK-1 and CMK-3 isotherms are plotted in the form of ln(P/P0) as a function of γ/(Tρ), which reflects the temperature-dependent quantities of the Kelvin equation: ⎛P⎞ −2γ ln⎜ ⎟ = cos θ rRTρ ⎝ P0 ⎠

where r is the pore radius, R is the universal gas constant, γ is the surface tension, ρ is the molar liquid density, T is the temperature, and θ is the effective contact angle of the adsorbate on the pore surface. The fact that we do not observe an effect of temperature (in the range from 298 to 318 K) on the position of the adsorption branch clearly suggests that the mechanism of adsorption and pore condensation of water in CMK mesoporous carbons is significantly different than that of an adsorbate which completely wets the pore walls. The observed temperature dependence on capillary condensation for a wetting adsorbate is associated with the underlying adsorption mechanism. In the case of a wetting fluid, pore condensation occurs when, for a given pore width, a critical film thickness of the adsorbate is achieved; i.e., multilayer adsorption precedes capillary condensation. With increasing temperature, higher relative pressures are needed to reach this critical film thickness;1 hence, capillary condensation shifts to higher relative pressures. In contrast, experiments performed by various techniques (adsorption, SAXS, SANS) and theoretical work suggest that the adsorption behavior of water in hydrophobic microporous carbons is controlled by formation of certain size water clusters which adsorb on existing polar sites (e.g., oxygen) on the carbon walls.17,34 Pore filling of a pore of a given width is correlated with a critical water cluster size. The effect of temperature on water adsorption in microporous active carbons13,16,52,53 has been studied and is qualitatively in good agreement with the behavior we observe for CMK carbons. Based on this, it seems that pore condensation of water in the mesopores of CMK is not controlled by the formation of adsorption layers (i.e., multilayer adsorption does not occur prior to pore condensation), but by a mechanism similar to the mentioned mechanism for water adsorption in purely microporous carbons which is controlled by the formation of metastable water clusters. Very recent work which focused on the effect of temperature on water adsorption in micro- and mesoporous carbons derived from resorcinol−formaldehyde cryogels in the temperature range from 263 to 298 K again highlights the importance of the cluster formation mechanism,54 but also in connection with an associated kinetic effect which is important in the much lower temperature range the authors studied as compared to our work. Hence, in contrast to a mechanism where pore condensation occurs when a critical

Figure 7. (a) Nitrogen (77.4 K) isotherms on CMK-3 (318) and CMK-3 (348). (b) QSDFT pore size distributions calculated from the nitrogen (77.4 K) isotherms of CMK-3 (318) and CMK-3 (348). (c) Water (298 K) isotherms on CMK-3 (318) and CMK-3 (348).

3.1.4. Effect of Temperature on Water Adsorption in CMK Carbons. In order to further evaluate the details of the water adsorption mechanism in the CMK carbon materials, we studied the effect of temperature on the water adsorption isotherms of CMK-1 and CMK-3 (318) at 298, 308, and 318 K. As displayed in Figure 8, we observe that for both samples the width of the hysteresis loops associated with mesopore filling shrinks with increasing temperature as one would observe in the case of argon or nitrogen adsorption at their boiling temperatures.51 However, contrary to the known temperature behavior of capillary condensation and evaporation of argon and nitrogen, where the adsorbate phase completely wets the 14897


Langmuir

Article

Figure 8. Water adsorption isotherms at three different temperatures on (a) CMK-1 and (b) CMK-3 (318), plotted as a function of relative pressure and absolute pressure.

of absolute pressure, adsorption and desorption branches of the hysteresis loops both shift to higher pressures with increasing temperature. On the other hand, as discussed before, if plotted as a function of relative pressure (or relative humidity, Figures 8a and 8c), no temperature dependence on adsorption is observed. The water adsorption data shown in Figure 8 also clearly demonstrate that the adsorbed amount in the pseudoplateau regions of the isotherms does not appreciably depend on temperature in the range investigated (298−318 K). Interesting information can also be obtained from a plot of the isosteric heats of adsorption calculated from the water desorption isotherms shown in Figure 8 (the desorption branches of the hysteresis loops were used for the calculations). Figure 10 shows the isosteric heat for CMK-1 and CMK-3 (318) at higher adsorbed amounts (>50 cm3/g) as calculated from the desorption branches. Identical heat curves were observed despite the differences in pore shape and size for these two materials. The isosteric heat is constant over a wide range of loading, which is close to the condensation enthalpy of bulk water (44 kJ/mol at 298 K). This indicates that indeed a bulklike water phase forms in the core of the mesopores. This is interesting because the average density of water in the CMK-3 (318) mesopores corresponds to only 0.81 g/cm3 (instead of a density of ca. 1 g/cm3 for bulk water). The average water density was calculated from the adsorbed amount of water close to the saturation pressure and using the total pore volume obtained from nitrogen adsorption after filling of the main mesopores: 0.94 cm3/g for CMK-3 (318)). The low average density can be explained by the fact that due to the hydrophobic nature of the carbon surface, the water

Figure 9. Water evaporation/desorption pressures of CMK-1 and CMK-3 plotted in ln(P/P0) as a function of γ/(Tρ), where γ is the surface tension, T is experimental temperature, and ρ is the corresponding (bulk) liquid density of water.

thickness of the adsorbate multilayer is reached, our adsorption data on these ordered mesoporous carbons suggest that pore condensation of water in CMK occurs when a critical size of the water cluster is reached. The results suggest further that the size of these water clusters (in the studied temperature range) should only depend on the relative humidity (relative pressure), although at higher temperatures, larger absolute pressures are needed to form this critical size of the water cluster. This follows from Figures 8b and 8d; i.e., if one plots the adsorption isotherms as a function 14898


Langmuir

Article

isotherms were measured for these two PMOs and are shown in Figure 11a. In both cases the argon isotherms are fully

Figure 10. Isosteric heat of adsorption for CMK-1 and CMK-3 as a function ofadsorbed amounts (starting at an adsorbed amount of ca. 50 cm3/g STP).

density at the pore walls is depleted55−57 but approaches bulk liquid density in the core of the pore, leading to an average density smaller than the bulk density. Previous studies on nanoporous carbons have also found average water densities in pores of less than the bulk water density.58,59 This also explains why, at higher loadings, the isosteric heat of adsorption approaches the condensation enthalpy of bulk water, reflecting a bulk like water phase in the core of the pores. This is also in line with the observation that the adsorbed amount at the quasi-plateau of the water adsorption isotherm (at high relative pressures) remains virtually unchanged over this temperature range, reflecting that the density of bulk water does not change appreciably in the temperature range measured. In summary, our water adsorption data obtained in the range from 298 to 318 K on mesoporous carbons strongly suggest that the mechanism of water adsorption into hydrophobic carbon mesopores is similar to the cluster formation mechanism suggested for adsorption into micropores; i.e., pore filling of mesoporous carbons does not resemble conventional pore condensation as observed for wetting fluids such as argon and nitrogen. On the other hand, the effect of temperature on water evaporation from the mesopores resembles the behavior of capillary evaporation of nitrogen and argon. Hence, this indicates that the observed hysteresis is due to differences in pore filling and emptying mechanisms. Further work involving in-situ scattering experiments is necessary to study the mechanism of water adsorption and pore filling into ordered mesoporous carbons in more detail. Furthermore, our data demonstrate that water adsorption is sensitive to pore size and the existence of both micro- and mesopores is clearly visible in the shape of the water adsorption isotherm. The shape of the water adsorption isotherms confirms the bimodal pore size distribution of these CMK carbon materials as suggested by QSDFT pore size analysis of the nitrogen and argon adsorption isotherms. 3.2. Effect of Surface Chemistry on Water Adsorption: PMOs. To investigate the effect that surface chemistry has on water adsorption, we studied water adsorption in two ordered mesoporous molecular sieves, i.e., two PMOs with essentially identical pore sizesone containing benzene organic bridging ligands (benzene PMO) and one containing divinylbenzene bridging ligands (divinylbenzene PMO). Argon (87.3 K)

Figure 11. (a) Argon (87.3 K) isotherms for benzene and divinylbenzene PMO. (b) NLDFT pore size distributions calculated from the argon (87.3 K) isotherms for benzene and divinylbenzene PMO.

reproducible and pore condensation without hysteresis is observed. From the argon isotherms we derived BET areas for benzene PMO and divinylbenzene PMO of 576 and 616 m2/g, respectively. Pore size distributions calculated using NLDFT methods (Figure 11b) clearly reveal that these PMOs have nearly identical pore size distributions with mesopores which are centered around 3.6 nm. Because of their identical pore structure, but differing chemical surfaces, these two samples represent the ideal test of the effect that surface chemistry has on water adsorption. Figure 12 shows the water adsorption isotherms measured for the benzene and divinylbenzene PMOs. Both water adsorption isotherms reveal pore condensation and hysteresis of type H1 (in contrast to the argon adsorption isotherms), and compared to nitrogen and argon adsorption, pore condensation and hysteresis are shifted to much higher relative pressures, indicating the hydrophobic nature of both PMO materials. This is in line with water adsorption measurements obtained on various thin films of PMO.60 Despite essentially identical pore size of benzene PMO and divinylbenzene PMO, water adsorption in these two PMOs leads to completely different isotherms; i.e., the pore condensation and evaporation for divinylbenzene PMO occur at an appreciably higher pressure than for benzene PMO. Also, adsorption in the relative pressure 14899


Langmuir

Article

indicating micro- and mesopore structure which is consistent with the bimodal pore size distributions obtained by applying proper QSDFT methods on argon (87.3 K) and nitrogen (77.4 K) adsorption isotherms. We also clearly demonstrate, based on our results obtained on two PMO samples with identical pore size and structure, the potential of water to detect small variations in surface chemistry. In addition, our results suggest that the underlying porefilling mechanism of water into hydrophobic carbon mesopores deviates from conventional capillary condensation (for wetting adsorbates such as nitrogen and argon) but is consistent with the assumption of a clustering mechanism (which had been identified as the underlying mechanism for water adsorption into microporous carbons). Further work involving in-situ scattering experiments is necessary in order to study the mechanism of water adsorption and pore filling into ordered mesoporous carbons in more detail. For these carbons, genuine water adsorption hysteresis is observed for filling of micropores and narrow mesopores in these carbons in a pore size range where nitrogen and argon isotherms are reversible, offering potential for pore network characterization; more work is currently underway to explore this topic. It should be stressed that whereas nitrogen and argon adsorption allow one to obtain pore volume, pore size, and surface area information, these quantities cannot be determined by water adsorption in a straightforward way due to its sensitivity to surface chemistry (leading only to partial wetting on the surface). However, we have shown in this work the potential of water as a complementary probe molecule in physical adsorption characterization.

Figure 12. Water (298 K) adsorption isotherms for benzene and divinylbenzene PMO.

range prior to condensation is significantly smaller for divinylbenzene PMO. Both the shift of pore condensation to higher relative pressures for water in divinylbenzene PMO and the lower water adsorption amount prior to condensation compared to benzene PMO reflect differences in the surface chemistry of the two samples, indicating that divinylbenzene PMO is more hydrophobic than benzene PMO. In line with this is the observation that the adsorbed amount at the upper closure point of hysteresis is smaller for divinylbenzene PMO as compared to benzene PMO, although the total pore volume (based on nitrogen and argon adsorption) for divinylbenzene PMO is higher as compared to benzene PMO. This indicates that the average density of adsorbed water in the more hydrophobic divinylbenzene PMO is smaller than in the less hydrophobic benzene PMO sample, i.e., 0.88 g/cm3 vs 0.98 g/ cm3 (calculated from the adsorbed water amounts at the upper closure point of hysteresis and the pore volumes obtained from the argon data after filling of the main mesopores: 0.42 cm3/g for benzene PMO and 0.47 cm3/g for divinylbenzene PMO). Similar to porous carbon, the average water density is smaller than the bulk density. Another interesting observation is that the hysteresis loop for benzene PMO is narrower than for the more hydrophobic divinylbenzene PMO, indicating clearly (because of identical pore diameter) the effect of surface chemistry and wettability on the width of hysteresis for a given pore size and geometry (please note that quite wide hysteresis loops were also observed for the hydrophobic CMK mesoporous carbon).

■

ASSOCIATED CONTENT

S Supporting Information *

Table of surface areas, pore volumes, and mode mesopore diameters for all materials. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail [email protected] (M.T.). *E-mail [email protected] (M.F.). Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS M.T. thanks A.V. Neimark for helpful discussions.

4. SUMMARY AND CONCLUSIONS We present a systematic study combining nitrogen and argon (at 77.4 and 87.3 K, respectively) with water adsorption for an advanced characterization of CMK carbons and PMO materials. Accurate micro-mesopore analysis was obtained by analyzing nitrogen and argon adsorption data with advanced density functional theory methods (i.e., NLDFT, QSDFT). Our characterization allows for clear interpretation of the effect of pore size (CMK carbons) and surface chemistry (PMOs) on water adsorption, and the data indicate that water adsorption, including adsorbed amount, pore filling, and hysteresis behavior, is sensitive to small changes in both pore width and surface chemistry. In fact, water adsorption data obtained on hydrophobic CMK carbons reveal a two-step isotherm

REFERENCES

(1) Thommes, M. Physical Adsorption Characterization of Ordered and Amorphous Mesoporous Materials. In Nanoporous Materials: Science and Engineering; Lu, G. Q., Zhao, X. S., Eds.; Imperial College Press: London, 2004; pp 317−364. (2) Thommes, M. Textural Characterization of Zeolites and Ordered Mesoporous Materials by Physical Adsorption. In Introduction to Zeolite Science and Practice; Cejka, J., van Bekkam, H., Corma, A., Schüth, F., Eds.; Elsevier: Amsterdam, 2007; pp 495−523. (3) Kaneko, K.; Itoh, T.; Fujimori, T. Collective Interactions of Molecules with an Interfacial Solid. Chem. Lett. 2012, 41, 466−475. (4) Lastoskie, C.; Gubbins, K. E.; Quirke, N. Pore Size Distribution Analysis of Microporous Carbons: A Density Functional Theory Approach. J. Phys. Chem. 1993, 97, 4786−4796. (5) Seaton, N. A.; Walton, J. R. B.; Quirke, N. A New Analysis Method for the Determination of the Pore Size Distributions of

14900


Langmuir

Article

Porous Carbons from Nitrogen Adsorption Measurements. Carbon 1989, 27, 853−861. (6) Olivier, J. P.; Conklin, W. B.; Szombathely, M. V. Determination of Pore Size Distribution from Density Functional Theory: A Comparison of Nitrogen and Argon Results. Stud. Surf. Sci. Catal. 1994, 87, 81−89. (7) Landers, J.; Gor, G. Y.; Neimark, A. V. Density Functional Theory Methods for Characterization of Porous Materials. Colloids Surf., A 2013, DOI: 10.1016/j.colsurfa.2013.01.007. (8) Thommes, M.; Cychosz, K. A.; Neimark, A. V. Advanced Physical Adsorption Characterization of Nanoporous Carbons. In Novel Carbon Adsorbents; Tascon, J. M. D., Ed.; Elsevier: Amsterdam, 2012; pp 107− 145. (9) Ustinov, E. A.; Do, D. D.; Fenelonov, V. B. Pore Size Distribution Analysis of Activated Carbons: Application of Density Functional Theory Using Nongraphitized Carbon Black as a Reference System. Carbon 2006, 44, 653−663. (10) Neimark, A. V.; Lin, Y.; Ravikovitch, P. I.; Thommes, M. Quenched Solid Density Functional Theory and Pore Size Analysis of Micro-Mesoporous Carbons. Carbon 2009, 47, 1617−1628. (11) Jagiello, J.; Olivier, J. P. A Simple Two-Dimensional NLDFT Model of Gas Adsorption in Finite Carbon Pores. Application to Pore Structure Analysis. J. Phys. Chem. C 2009, 113, 19382−19385. (12) Lucena, S. M. P.; Paiva, C. A. S.; Silvino, P. F. G.; Azevedo, D. C. S.; Cavalcante, C. L. The Effect of Heterogeneity in the Randomly Etched Graphite Model for Carbon Pore Size Characterization. Carbon 2010, 48, 2554−2565. (13) Brennan, J. K.; Bandosz, T. J.; Thomson, K. T.; Gubbins, K. E. Water in Porous Carbons. Colloids Surf., A 2001, 187−188, 539−568. (14) Do, D. D.; Do, H. D. A Model for Water Adsorption in Activated Carbon. Carbon 2000, 38, 767−773. (15) Do, D. D.; Do, H. D. Modeling of Adsorption on Nongraphitized Carbon Surface: GCMC Simulation Studies and Comparison with Experimental Data. J. Phys. Chem. B 2006, 110, 17531−17538. (16) Kaneko, K.; Hanzawa, Y.; Iiyama, T.; Kanda, T.; Suzuki, T. Cluster-Mediated Water Adsorption on Carbon Nanopores. Adsorption 1999, 5, 7−13. (17) Kimura, T.; Kanoh, H.; Kanda, T.; Ohkubo, T.; Hattori, Y.; Higaonna, Y.; Denoyel, R.; Kaneko, K. Cluster-Associated Filling of Water in Hydrophobic Carbon Micropores. J. Phys. Chem. B 2004, 108, 14043−14048. (18) Lodewyckx, P.; Vansant, E. F. Water Isotherms of Activated Carbons with Small Amounts of Surface Oxygen. Carbon 1999, 37, 1647−1649. (19) Lodewyckx, P.; Raymundo-Piñero, E.; Wullens, H.; Vaclavikova, M.; Beguin, F. Water Isotherms of Structurally Identical Carbons with Different Amounts of Surface Oxygen Groups. In Proceedings of the International Carbon Conference, Nagano, Japan, 2008. (20) Liu, J.-C.; Monson, P. A. Does Water Condense in Carbon Pores? Langmuir 2005, 21, 10219−10225. (21) Liu, J.-C.; Monson, P. A. Monte Carlo Simulation Study of Water Adsorption in Activated Carbon. Ind. Eng. Chem. Res. 2006, 45, 5649−5656. (22) Slasli, A. M.; Jorge, M.; Stoeckli, F.; Seaton, N. A. Modelling of Water Adsorption by Activated Carbons: Effects of Microporous Structure and Oxygen Content. Carbon 2004, 42, 1947−1952. (23) Slasli, A. M.; Jorge, M.; Stoeckli, F.; Seaton, N. A. Water Adsorption by Activated Carbons in Relation to Their Microporous Structure. Carbon 2003, 41, 479−486. (24) Stoeckli, F.; Lavanchy, A. The Adsorption of Water by Active Carbons, in Relation to their Chemical and Structural Properties. Carbon 2000, 38, 475−477. (25) Striolo, A.; Gubbins, K. E.; Gruszkiewicz, M. S.; Cole, D. R.; Simonson, J. M.; Chialvo, A. A.; Cummings, P. T.; Burchell, T. D.; More, K. L. Effect of Temperature on the Adsorption of Water in Porous Carbons. Langmuir 2005, 21, 9457−9467.

(26) Ohba, T.; Kanoh, H.; Kaneko, K. Cluster-Growth-Induced Water Adsorption in Hydrophobic Carbon Nanopores. J. Phys. Chem. B 2004, 108, 14964−14969. (27) 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. (28) Thommes, M.; Morlay, C.; Ahmad, R.; Joly, J. P. Assessing Surface Chemistry and Pore Structure of Active Carbons by a Combination of Physisorption (H2O, Ar, N2, CO2), XPS and TPDMS. Adsorption 2011, 17, 653−661. (29) Rouquerol, F.; Rouquerol, J.; Sing, K. Adsorption by Powders and Porous Solids, 1st ed.; Academic Press: London, 1999; pp 276−279. (30) Kiselev, A. V. In The Structure and Properties of Porous Materials; Everett, D. H., Stone, F. S., Eds.; Butterworths: London, 1958; p 51. (31) Kiselev, A. V. In Proceedings of the Second International Congress on Surface Activity; Butterworths: London, 1957; p 219. (32) Dubinin, M. M. In Chemistry and Physics of Carbon; Walker, P. L., Ed.; Marcel Dekker: New York, 1966; p 51. (33) Ohba, T.; Kaneko, K. Kinetically Forbidden Transformations of Water Molecular Assemblies in Hydrophobic Micropores. Langmuir 2011, 27, 7609−7613. (34) Horikawa, T.; Sekida, T.; Hayashi, J.; Katoh, M.; Do, D. D. A New Adsorption-Desorption Model for Water Adsorption in Porous Carbons. Carbon 2011, 49, 416−424. (35) Ohba, T.; Kanoh, H.; Kaneko, K. Facilitation of Water Penetration through Zero-Dimensional Gates on Rolled-Up Graphene by Cluster-Chain-Cluster Transformations. J. Phys. Chem. C 2012, 116, 12339−12345. (36) Nguyen, T. X.; Bhatia, S. K. How Water Adsorbs in Hydrophobic Nanospaces. J. Phys. Chem. C 2011, 115, 16606−16612. (37) Lodewyckx, P. The Effect of Water Uptake in Ultramicropores on the Adsorption of Water Vapour in Activated Carbon. Carbon 2010, 48, 2549−2553. (38) Ryoo, R.; Joo, S. H.; Jun, S. Synthesis of Highly Ordered Carbon Molecular Sieves via Template-Mediated Structural Transformation. J. Phys. Chem. B 1999, 103, 7743−7746. (39) Jun, S.; Joo, S. H.; Ryoo, R.; Kruk, M.; Jaroniec, M.; Liu, Z.; Ohsuna, T.; Terasaki, O. Synthesis of New, Nanoporous Carbon with Hexagonally Ordered Mesostructure. J. Am. Chem. Soc. 2000, 122, 10712−10713. (40) Kleitz, F.; Choi, S. H.; Ryoo, R. Cubic Ia3d Large Mesoporous Silica: Synthesis and Replication to Platinum Nanowires, Carbon Nanorods and Carbon Nanotubes. Chem. Commun. 2003, 2136−2137. (41) Inagaki, S.; Guan, S.; Fukushima, Y.; Ohsuna, T.; Terasaki, O. Novel Mesoporous Materials with a Uniform Distribution of Organic Groups and Inorganic Oxide in Their Frameworks. J. Am. Chem. Soc. 1999, 121, 9611−9614. (42) Melde, B. J.; Holland, B. T.; Blanford, C. F.; Stein, A. Mesoporous Sieves with Unified Hybrid Inorganic/Organic Frameworks. Chem. Mater. 1999, 11, 3302−3308. (43) Asefa, T.; MacLachlan, M. J.; Coombs, N.; Ozin, G. A. Periodic Mesoporous Organosilicas with Organic Groups Inside the Channel Walls. Nature 1999, 402, 867−871. (44) Hoffmann, F.; Cornelius, M.; Morell, J.; Fröba, M. Silica-Based Mesoporous Organic-Inorganic Hybrid Materials. Angew. Chem., Int. Ed. 2006, 45, 3216−3251. (45) Inagaki, S.; Guan, S.; Ohsuna, T.; Terasaki, O. An Ordered Mesoporous Organosilica Hybrid Material with a Crystal-like Wall Structure. Nature 2002, 416, 304−307. (46) Cornelius, M.; Hoffmann, F.; Fröba, M. Periodic Mesoporous Organosilicas with a Bifunctional Conjugated Organic Unit and Crystal-like Pore Walls. Chem. Mater. 2005, 17, 6674−6678. (47) Hoffmann, F.; Fröba, M. Vitalising Porous Inorganic Silica Networks with Organic Functions − PMOs and Related Hybrid Materials. Chem. Soc. Rev. 2011, 40, 608−620. (48) Huwe, H.; Fröba, M. Iron (III) Oxide Nanoparticles within the Pore System of Mesoporous Carbon CMK-1: Intra-Pore Synthesis and Characterization. Microporous Mesoporous Mater. 2003, 60, 151−158. 14901


Langmuir

Article

(49) Thommes, M.; Mitchell, S.; Pérez-Ramírez, J. Surface and Pore Structure Assessment of Hierarchical MFI Zeolites by Advanced Water and Argon Sorption Studies. J. Phys. Chem. C 2012, 116, 18816− 18823. (50) Gor, G. Y.; Thommes, M.; Cychosz, K. A.; Neimark, A. V. Quenched Solid Density Functional Theory Method for Characterization of Mesoporous Carbons by Nitrogen Adsorption. Carbon 2012, 50, 1583−1590. (51) Lowell, S.; Shields, J. E.; Thomas, M. A.; Thommes, M. Characterization of Porous Solids and Powders: Surface Area, Pore Size and Density; Springer: Dordrecht, The Netherlands, 2004. (52) Ohba, T.; Kanoh, H.; Kaneko, K. Water Cluster Growth in Hydrophobic Solid Nanospaces. Chem.Eur. J. 2005, 11, 4890−4894. (53) 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. (54) 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. (55) Brovchenko, I.; Geiger, A. Water in Nanopores in Equilibrium with a Bulk Reservoir − Gibbs Ensemble Monte Carlo Simulations. J. Mol. Liq. 2002, 96−97, 195−206. (56) Brovchenko, I.; Geiger, A.; Oleinikova, A. Water in Nanopores: II. The Liquid-Vapour Phase Transition Near Hydrophobic Surfaces. J. Phys.: Condens. Matter 2004, 16, S5345−S5370. (57) Brovchenko, I.; Oleinikova, A. Interfacial and Confined Water; Elsevier: Oxford, UK, 2008. (58) 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. (59) Alcañiz-Monge, J.; Linares-Solano, A.; Rand, B. Mechanism of Adsorption of Water in Carbon Micropores As Revealed by a Study of Activated Carbon Fibers. J. Phys. Chem. B 2002, 106, 3209−3216. (60) Wang, W. D.; Grozea, D.; Kohli, S.; Perovic, D. D.; Ozin, G. A. Water Repellent Water Repellent Periodic Mesoporous Organosilicas. ACS Nano 2011, 5, 1267−1275. Wang, W. Periodic Mesoporous Organosilica and Silica, Ph.D. Thesis, Department of Chemistry, University of Toronto, 2011.

14902


Combining Nitrogen, Argon, and Water Adsorption for Advanced

Recommend Documents