Trifluoroethanol and 19F Magic Angle Spinning Nuclear Magnetic

Jun 23, 2011 - Studied here are three zirconium hydroxide samples, two unperturbed commercial materials, and one commercial material that is crushed b...
0 downloads 11 Views 3MB Size
Download PDF
ARTICLE pubs.acs.org/Langmuir

Trifluoroethanol and 19F Magic Angle Spinning Nuclear Magnetic Resonance as a Basic Surface Hydroxyl Reactivity Probe for Zirconium(IV) Hydroxide Structures Jared B. DeCoste,*,† T. Grant Glover,‡ Gregory Mogilevsky,† Gregory W. Peterson,† and George W. Wagner† †

Edgewood Chemical Biological Center, 5183 Blackhawk Road, Building 3549, Aberdeen Proving Ground, Maryland 21010-5424, United States ‡ Science Applications International Corporation (SAIC), Post Office Box 68, Gunpowder, Maryland 21010, United States

bS Supporting Information ABSTRACT: A novel technique for determining the relative accessibility and reactivity of basic surface hydroxyl sites by reacting various zirconium(IV) hydroxide materials with 2,2,2trifluoroethanol (TFE) and characterizing the resulting material using 19F magic angle spinning (MAS) nuclear magnetic resonance (NMR) is presented here. Studied here are three zirconium hydroxide samples, two unperturbed commercial materials, and one commercial material that is crushed by a pellet press. Factors, such as the ratio of bridging/terminal hydroxyls, surface area, and pore size distribution, are examined and found to affect the ability of the zirconium(IV) hydroxide to react with TFE. X-ray diffraction, nitrogen isotherms, and 1H MAS NMR were used to characterize the unperturbed materials, while thermogravitric analysis with gas chromatography and mass spectrometry along with the 19F MAS NMR were used to characterize the materials that were reacted with TFE. Zirconium hydroxide materials with a high surface area and a low bridging/terminal hydroxyl ratio were found to react TFE in the greatest amounts.

1. INTRODUCTION In recent years, there has been significant interest in porous materials for the removal of toxic industrial chemicals/materials (TICs/TIMs) and potential chemical warfare agents/simulants (CWAs).1 8 Porous materials are sought out because they possess high surface areas, selectivity, and potential for high concentrations of surface functional groups. We report here a novel method to determine reactive basic sites that are accessible to small organic molecules in porous materials. Basic hydroxyl sites are of specific interest because of their reactivity with acid gases, such as cyanogen chloride and sulfur dioxide.7,9,10 There are many instrumental methods for analyzing the structures and surfaces of metal hydroxides. Typically, these methods give insight into either the structural or chemical properties of the metal hydroxide but not both at the same time. For instance, the number of basic surface hydroxyl sites on a given metal hydroxide has been studied by directly probing a given material with 1H magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy and X-ray photoelectron spectroscopy (XPS).7,11 These techniques give insight into the total number of basic surface hydroxyl groups but do not tell about their accessibility to sorbates and toxic chemicals. At the other end of the spectrum, metal hydroxide materials have been studied using X-ray diffraction (XRD) and nitrogen isotherms.12 These methods lend information about the crystal structure and surface area of the material but tend not to tell much about its chemical makeup. The accessibility of basic surface hydroxyl sites is very important in the reactive acid base chemistry and sorption of small r 2011 American Chemical Society

molecules by zirconium(IV) hydroxide.7,10 Few studies have been performed to date probing the reactive surface sites of zirconium hydroxide materials.13,14 Highly crystalline metal oxide materials do not have pores accessible even to small molecules e5 Å, allowing for minimal sorption. One of the most appealing characteristics of zirconium hydroxide materials is its micro-/mesoporosity that allows small molecules to diffuse through the pores and access adsorption/reaction sites for either adsorption or chemical reactivity. The structure and morphology of zirconium hydroxide materials vary widely and depends upon factors such as the synthesis process (e.g., aqueous precipitation versus sol gel methods), reaction conditions (e.g., pH and solvent), and temperature of calcination.12,15,16 The synthesized structure and post-synthetic treatment of zirconium hydroxide varies immensely from one commercial supplier to another. It has been shown that the zirconium hydroxide materials are typically composed of a combination of localized tetragonal, monoclinic, and amorphous phases. There is a stark decrease in the amount of amorphous and tetragonal phases and an increase in the amount of monoclinic phase as the material is calcined at higher temperatures.12 The building blocks of zirconium hydroxides are composed of zirconium atoms, which are linked together by either bridging hydroxyl groups in the R-type structure or oxo bridges in the Received: May 31, 2011 Revised: June 23, 2011 Published: June 23, 2011 9458

dx.doi.org/10.1021/la202029e | Langmuir 2011, 27, 9458–9464

Langmuir

ARTICLE

Figure 1. Simple model structures are shown for (a) R-, (b) γ-, and (c) β-type hydroxyl and oxo bridges. The valences of Zr4+ shown are not complete for simplistic purposes.

Scheme 1. Basic Surface Hydroxyl Groups of Zirconium Hydroxide React Stoichiometrically with TFE

γ-type structure (Figure 1). The β type is a hybrid of the R and γ types, having both bridging hydroxyls and oxo atoms. The R type is the building block for the tetragonal zirconium hydroxide, which has gained much interest for its use in catalysis.17,18 The γ type is the building block for the monoclinic form, which does not seem to have the same reactive capabilities as the tetragonal form. Most commercially available zirconium hydroxides have some combination of the two forms. In this study, 2,2,2-trifluoroethanol (TFE) in conjunction with 19 F MAS NMR is used not only to elucidate the structure of commercially available zirconium hydroxide materials but also to reveal the accessibility of reactive sites in the material. Fluorinecontaining probe molecules have been used by others to quantify reactive surface sites with 19F MAS NMR.19,20 TFE is expected to act as a weak acid (pKa = 11.4)21 in the zirconium hydroxide environment and probe basic surface OH sites by following Scheme 1.

2. EXPERIMENTAL SECTION 2.1. TFE Loading. Zirconium hydroxide samples from Sigma Aldrich (ZHSA) and MEL Chemicals (ZHMEL) were obtained and heated to 100 °C for 24 h to remove bulk water on the surface of the sample. A total of 0.75 g of the zirconium hydroxide was then stirred in a round-bottom flask with 5.0 mL of chloroform (Aldrich, 99.8%) and 2.5 mL of TFE (Aldrich, 99%, boiling point of 77 80 °C) in excess to ensure that all accessible sites are probed for 16 h at room temperature. The sample was vacuum-filtered, and the white powder was washed 3 times with 5 mL of chloroform. The sample was then heated to 100 °C to remove any residual organic material and water that may be physisorbed to the surface. A third sample, ZHMEL-P, was prepared by crushing the ZHMEL material into a pellet under approximately 5 tons of pressure with a SPEX Industries, Inc. X-PRESS pellet press and a Chemplex Industries 35 mm die. ZHMEL-P was ground into a powder before it was heated to remove bulk water. 2.2. 19F MAS NMR. A known amount of the reacted zirconium hydroxide sample was loaded into a 5 mm thick walled silicon nitride rotor with long Aurum caps. The selection of rotor materials free of fluorine is important to accurately quantify reacted TFE. A known amount of hexafluorobenzene (Aldrich, 99.5%), a NMR standard, was added to the sample in the rotor before it was packed to quantify the fluorine signal. The packed sample was spun at ∼11 kHz to minimize broadening effects as well as the number and intensity of spinning side bands. The 19F MAS NMR spectra were obtained at 564.4 MHz using a 26 μs pulse (90°) and a relaxation delay of 0.200 s, followed by an acquisition time of 1.747 s, repeated for 1000 transients on a Varian 600 NMR spectrometer with a magnetic field of 14.1 T, and equipped with a DOTY Scientific XC-5 5 mm VT-MAS NMR probe. Spectra were

referenced to the internal hexafluorobenzene signal (δ = 163 ppm). The experiment was repeated a minimum of 4 times for each sample. 2.3. TGA GC MS. A known amount of reacted ZHMEL with TFE was loaded into a platinum pan for analysis with a TA Instruments Q 500 thermogravimetric analyzer (TGA). The temperature of the sample was raised from room temperature to 400 °C at a rate of 3 °C/ min under a constant N2 flow of 20 mL/min. The vapors expelled by the sample were passed to an Agilent 6890 G1530 gas chromatograph (GC). The GC was equipped with a 250 μL sample loop, which was filled with analyte vapors and expelled every 2.5 min onto an Agilent HP-5MS column (30 m  0.25 mm  1.0 μm), with a flow rate of 1.5 mL/min at a temperature of 35 °C. After passing through the column, gases were analyzed via a Hewlett-Packard 5973 mass spectrometer (MS) at a rate of 6.48 scans/s from m/z 40 to 250. 2.4. Physical Structure Analysis. The surface area and pore volume of each zirconium hydroxide sample was measured using nitrogen adsorption isotherms measured on a Quantachrome Instruments Autosorb 1C analyzer. The samples were degassed at 100 °C for at least 4 h prior to analysis. The structure of each material was determined using a PANalytical X’Pert X-ray powder diffractometer (XRD) with an X’celerator detector. Samples were scanned at 45 kV and 40 mA, using Cu KR radiation (λ = 1.54 Å), a step size (10.16 s/step) of 2θ = 0.0167° over the 2θ range of 5 75°. 2.5. 1H MAS NMR. The density of terminal and bridging hydroxyl groups for each sample are determined using 1H MAS NMR with a method described in detail elsewhere.11 A known amount of the zirconium hydroxide sample was first packed into a 5 mm thick walled silicon nitride rotor, evacuated to 250 mTorr, and heated at 175 °C overnight to remove bulk and physisorbed water. The sample was weighed, and two Kel-F end caps were used to seal the rotor. The 1H MAS NMR spectra were obtained at 599.9 MHz using a 3.0 μs pulse (45°), 10 kHz spinning, and a relaxation delay of 5.0 s, followed by an acquisition time of 0.500 s, and repeated for 128 transients with the same spectrometer and probe as the 19F MAS NMR experiments. Spectra were referenced to an external tetramethylsilane signal (δ = 0 ppm). Peak deconvolution and integration was carried out with Acorn NUTS software.

3. RESULTS AND ANALYSIS 3.1. TFE Reaction with Zr(OH)4 Quantified with 19F MAS NMR. Representative 19F MAS NMR spectra for ZHMEL,

ZHMEL-P, and ZHSA after treatment with TFE are shown in Figure 2. The broad signal at δ ∼ 77.5 ppm with spinning side bands is consistent with immobile, covalently bound TFE species to a zirconium atom. The signal and spinning side bands were integrated and quantified using the integral of the hexafluorobenzene internal standard. The calculations to determine reactivity are explained more thoroughly in the Supporting Information. Reactivity of TFE with ZHMEL, ZHMEL-P, and ZHSA was determined to be 1.22, 0.55, and 0.36 mmol/g of zirconium hydroxide, respectively. A sample of zirconium(IV) oxide (Sigma 9459

dx.doi.org/10.1021/la202029e |Langmuir 2011, 27, 9458–9464

Langmuir

ARTICLE

Scheme 2. Reaction between Zirconium Hydroxide Terminal Hydroxyl Groups and TFE Results in a Bound Trifluoroethoxy Group with the Loss of Water

Figure 2. Representative 19F MAS NMR spectra of zirconium hydroxide materials + TFE referenced to the internal standard hexafluorobenzene peak at δ ∼ 163 ppm.

Figure 3. 19F MAS NMR spectra of ZHMEL + TFE from δ = 140 to 0 ppm with (a) no heating and heating to 85 °C for (b) 30, (c) 120, (d) 180, (e) 220, (f) 300, and (g) 360 min.

Figure 4. 19F MAS NMR spectra of ZHMEL + TFE upon heating to various temperatures.

Aldrich, 99.99%) was treated as a control in the same manner as the zirconium hydroxide samples, resulting in no appreciable signal from bound TFE (not shown). Crystalline zirconium oxide has no basic hydroxyl groups for TFE to react with and, therefore, would not be expected to show any signal in the 19F MAS NMR. When a ZHMEL sample was treated directly with excess TFE (no solvent), we saw the presence of a sharp signal in the 19F MAS NMR at δ ∼ 76.5 ppm, as seen in Figure 3. This sharp peak is consistent with free TFE not bound to surface hydroxyl groups. Upon heating the sample, the free TFE was driven off and the sharp peak disappeared, giving way to bound TFE, which has a much broader signal because of fewer degrees of freedom stemming from its restricted environment.22 The fluorine signal from the sample with no heat is a combination of bulk TFE, TFE physisorbed on the surface, and bound TFE as determined by the presence of spinning side bands. Further confirmation that the majority of TFE was covalently bound as a trifluoroethoxy species (as seen in Scheme 2) to the surface and not physisorbed was achieved by heating the ZHMEL sample to a given temperature for 2 h and taking the 19F MAS NMR spectra again, as presented in Figure 4. There was minimal

loss of the fluorine signal up to 200 °C, but at 250 °C, the signal decreased and ultimately disappeared. This is consistent with the breaking of Zr O bonds in coordinate covalent compounds.23,24 Also, at temperatures above 200 °C, zirconium hydroxide samples have local structural changes, which could be driving off the trifluoroethoxy species.12 TGA runs were performed on the ZHMEL + TFE samples, and the derivative of the weight loss was taken. An effort was made to understand the species that were liberated from the sorbent, and the TGA gas stream was analyzed via GC MS. The TGA of ZHMEL with and without the addition of TFE (Figure 5a) shows weight loss prior to 100 °C corresponding to the loss of free water. Weight loss also occurs in both samples at 130 °C, which may correspond to the loss of coordinated water. The ZHMEL + TFE weight losses at 170, 220, and 350 °C are not consistent with the base ZHMEL material. Representative chromatograms of ZHMEL + TFE from the TGA GC MS experiment can be seen for m/z 61, the base peak of TFE, in Figure 5b. Figure 5c shows the mass spectrum of the peak at a retention time of 1.67 min for the spectra with the most intense signal (temperature = 218 °C) and confirms that the species being expelled by the ZHMEL + TFE sample is CF3CH2OH. GC MS shows 9460

dx.doi.org/10.1021/la202029e |Langmuir 2011, 27, 9458–9464

Langmuir

ARTICLE

Figure 6. Nitrogen adsorption isotherms for zirconium hydroxide samples. Closed markers indicate adsorption, and open markers indicate desorption.

Figure 5. (a) Thermogravimetric analysis, (b) gas chromatogram of m/z 61 ion, and (c) mass spectrum at retention time = 1.67 min of ZHMEL + TFE.

small amounts of loosely bound TFE expelled at 80 and 120 °C. This represents a small fraction of the total TFE being physisorbed to the surface. There is a moderate presence of TFE liberation at 170 °C and a more significant presence of TFE liberation at 220 °C. This represents a loss of two species with different binding energies. These results are consistent with the presence of at least two basic hydroxyl groups with differing pKa values reacting with TFE. Hydroxyl groups with different pKa values are common in metal hydroxides.25 Further speculation is beyond the scope of this study. 3.2. Physical Properties of Zr(OH)4 Structures. Powder XRD patterns (for spectra, see the Supporting Information) for

each of the zirconium hydroxide materials show that the zironcium hydroxide materials are amorphous to XRD. On the other hand, the XRD pattern for the zirconium oxide control sample showed the presence of highly crystalline material with the presence of both monoclinic (2θ = 28° and 32°) and tetragonal (2θ = 30°) phases.12 The accessibility of basic hydroxyl sites is directly related to the surface area of the zirconium hydroxide material. A higher surface area leads to more exposed surface sites that can be reacted with TFE. Figure 6 shows the N2 adsorption isotherms of the three zirconium hydroxide materials. The Brunauer Emmett Teller (BET) surface area of ZHMEL decreased from 429 to 333 m2/g upon crushing into the pellet form.26 The ZHSA material had a surface area of only 249 m2/g. All three nitrogen isothems are classified as International Union of Pure and Applied Chemistry (IUPAC) type IV, which is consistent with the presence of mesopores. Furthermore, the presence of an IUPAC H4-type hysteresis loop is consistent with the slit-like pores that arise in layered materials.27 The nitrogen isotherm for ZHSA does show more type-I characteristics than ZHMEL and ZHMEL-P and has a much narrower hysteresis loop, which shows that the porosity of ZHSA is more microporous in nature, instead of a wide distribution of pore sizes as present in the other zirconium hydroxide samples. A nitrogen isotherm of ZHMEL treated with TFE was taken and showed no significant changes in the shape of the isotherm. This indicates that the integrity of the zirconium hydroxide structure was maintained from the reaction with TFE. The total pore volume for ZHMEL, ZHMEL-P, and ZHSA was determined to be 0.50, 0.37, and 0.18 cm3/g, respectively. The pore size distribution was calculated using a nonlocal density functional theory (NLDFT) using a silica cylindrical pore kernel. This kernel was used, as in other publications, because a zironia polymorph kernel has not been commercially developed.9 The results of the NLDFT calculations can be seen in Figure 7. It can be seen that there is a significant decrease in the number of pores between 25 and 50 Å upon crushing of the ZHMEL sample. Also, there is a qualitative decrease in the larger macropores above 500 Å. Interestingly, there is relative consistency in the pore size distribution between ZHMEL and ZHMEL-P in the mesopores between 50 and 500 Å. When the ZHSA sample is compared to the ZHMEL samples, it becomes evident that there is a 9461

dx.doi.org/10.1021/la202029e |Langmuir 2011, 27, 9458–9464

Langmuir

ARTICLE

Figure 7. Pore size distribution for zirconium hydroxide materials was calculated from 10 to 750 Å using a NLDFT method using a silica cylindrical pore kernel. Note the change in the vertical scale at 100 Å.

Figure 8. Deconvoluted 1H MAS NMR spectra of (left) ZHMEL, (middle) ZHMEL-P, and (right) ZHSA from δ =

substantial lack of pores greater than 50 Å in the ZHSA sample, which prohibits diffusion through the pores. 3.3. Determination of the Bridging/Terminal Hydroxyl Ratio. The 1H MAS NMR spectra were taken of each unperturbed sample to determine the relative number of bridging and terminal hydroxyl groups. Figure 8 shows the deconvoluted 1H MAS NMR spectra of each zirconium hydroxide sample. The assignment of the 1H NMR peaks has been reported elsewhere.11,28 The peaks between δ ∼ 1 and 2 ppm are part of the background from the rotor and probe, which was confirmed by performing a blank run. The peaks at δ ∼ 3, 4, and 6 ppm were assigned to the terminal hydroxyl groups, bridging hydroxyl groups, and physisorbed water, respectively. 1 H NMR signal intensity provides quantitative information on the amount of protons. From the 1H MAS NMR deconvoluted peaks, the ratio of bridging/terminal hydroxyl groups for each sample was determined. ZHMEL was not chemically perturbed upon crushing to the pellet form ZHMEL-P, because both the ZHMEL and ZHMEL-P samples have bridging/ terminal hydroxyl ratios of approximately 3. However, the peak representing the physisorbed water decreased significantly, which is consistent with the decrease in the pore volume and surface area, leading to a decrease in H2O capacity. The ZHSA sample on the other hand had significantly fewer terminal hydroxyl groups, with a bridging/terminal hydroxyl ratio of approximately 8. This is possibly from higher calcination temperatures, different pH, and/or different starting materials used by the supplier, causing the formation of larger zirconium hydroxide sheets. The surface area of larger sheets tends to

10 to 15 ppm.

have a much less accessible surface area, because the sheets can pack in a much more regular configuration.

4. DISCUSSION All three amorphous zirconium hydroxide samples yielded significant reaction with TFE when compared to the crystalline zirconium oxide control sample. However, the materials with higher surface areas, higher pore volumes, and lower bridging/terminal hydroxyl ratios had a much higher reactive capacity, as seen in Table 1. TFE is expected to react with only the basic terminal hydroxyl groups. The decrease in reactivity of ZHMEL-P when compared to ZHMEL was expected upon crushing of the sample. The surface area decreased by 22%, but the reactivity of TFE decreased by 55%, while the ratio of bridging/terminal OH groups remained constant. The DFT analysis shows that there are significant decreases in the pores smaller than 5 nm and pores larger than 50 nm upon crushing the sample. However, the mesopores from 5 to 50 nm show minimal change. This is consistent with, upon application of pressure, the less structured macropores collapsing, causing the partial and complete blocking of micropores, making them inaccessible, as illustrated in Figure 9. The micropores are likely still present in the materials; however, many of them are not accessible to nitrogen, and substantially more of them are not accessible to TFE. The ZHSA sample has not only the inaccessibility of the reactive basic surface hydroxyl sites because of a low surface area (42% less than ZHMEL) adversely effecting the reactivity but also a lower amount of total terminal hydroxyl groups for TFE to react. The DFT analysis shows that ZHSA still has pores smaller than 9462

dx.doi.org/10.1021/la202029e |Langmuir 2011, 27, 9458–9464

Langmuir

ARTICLE

Table 1. Quantity of TFE Reacted, Surface Area, Total Pore Volume, and Bridging/Terminal Hydroxyl Ratio for Each Zirconium Hydroxide Sample mmol of TFE reacted/g of

BET surface area

total pore volume

bridging/terminal OH

sample

Zr(OH)4 (19F MAS NMR)

(nitrogen isotherm) (m2/g)

(nitrogen isotherm) (cm3/g)

(1H MAS NMR)

ZHMEL ZHMEL-P

1.22 ( 0.07 0.55 ( 0.02

429 333

0.50 0.37

3 3

ZHSA

0.36 ( 0.02

249

0.18

8

Associateship Awards at the Edgewood Chemical Biological Center. This work was completed under the Joint Science and Technology Office for Chemical Biological Defense (JSTO CBD) Project BA07PRO104.

’ REFERENCES Figure 9. Illustration, not to scale, of collapsing macropores of ZHMEL upon pressure to block the smaller micropores.

5 nm, but there is a lack of pores greater than 10 nm. The micropores and small mesopores possess a high concentration of basic surface hydroxyl sites for reactivity, but the lack of larger pores substantially hinders the diffusion of even small organic molecules through the material. The combination of a high bridging/terminal hydroxyl ratio, low surface area, and tight pores hindering the transport of TFE through the structure all attribute to the 70% decrease in total TFE reactivity for ZHSA when compared to ZHMEL.

5. CONCLUSION Reacting zirconium hydroxide with TFE creates a surfacebound trifluoroethoxy species, which is stable above 200 °C, and can be quantified using 19F MAS NMR, allowing for a semiquantitative approach for probing basic surface hydroxyl site reactivity. The reactivity depends upon not only the number of basic surface hydroxyl sites available for reaction but also their accessibility. The accessibility of sites is directly affected by the surface area and pore size distribution. The commercially available zirconium hydroxide samples have markedly different physical and chemical properties, lending way to much different degrees of reactivity. The three samples studied here showed how surface area, pore volume, and number of reactive sites can affect the reactivity of zirconium hydroxide with acidic organic analytes, such as TFE. ’ ASSOCIATED CONTENT

bS

Supporting Information. Reactivity of trifluoroethanol calculations and powder XRD patterns of ZHMEL, ZHMEL-P, ZHSA, and ZrO2. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT This research was performed while Jared B. DeCoste and Gregory Mogilevsky held National Research Council Research

(1) Kanyi, C. W.; Doetschman, D. C.; Schulte, J. T. Microporous Mesoporous Mater. 2009, 124, 232. (2) Kanyi, C. W.; Doetschman, D. C.; Yang, S. W.; Schulte, J. T. Microporous Mesoporous Mater. 2009, 119, 23. (3) Sambur, J. B.; Doetschman, D. C.; Yang, S. W.; Schulte, J. T.; Jones, B. R.; DeCoste, J. B. Microporous Mesoporous Mater. 2008, 112, 116. (4) DeCoste, J. B.; Doetschman, D. C.; Lahr, M. J.; Kanyi, C. W.; Schulte, J. T. Microporous Mesoporous Mater. 2011, 139, 110. (5) Mahle, J. J.; Peterson, G. W.; Schindler, B. J.; Smith, P. B.; Rossin, J. A.; Wagner, G. W. J. Phys. Chem. C 2010, 114, 20083. (6) Wagner, G. W.; Chen, Q.; Wu, Y. J. Phys. Chem. C 2008, 112, 11901. (7) Peterson, G. W.; Wagner, G. W.; Keller, J. H.; Rossin, J. A. Ind. Eng. Chem. Res. 2010, 49, 11182. (8) Glover, T. G.; Peterson, G. W.; Schindler, B. J.; Britt, D.; Yaghi, O. Chem. Eng. Sci. 2011, 66, 163. (9) Peterson, G. W.; Rossin, J. A.; Karwacki, C. J.; Glover, T. G. J. Phys. Chem. C 2011, 115, 9644. (10) Peterson, G. W.; Karwacki, C. J.; Feaver, W. B.; Rossin, J. A. Ind. Eng. Chem. Res. 2009, 48, 1694. (11) Mogilevsky, G.; Karwacki, C.; Peterson, G.; Wagner, G. Chem. Phys. Lett. 201110.1016/j.cplett.2011.06.072. (12) Navio, J. A.; Colon, G.; SanchezSoto, P. J.; Macias, M. Chem. Mater. 1997, 9, 1256. (13) Jung, K. T.; Bell, A. T. J. Mol. Catal. A: Chem. 2000, 163, 27. (14) Wang, G. W.; Hattori, H.; Tanabe, K. Bull. Chem. Soc. Jpn. 1983, 56, 2407. (15) Southon, P. D.; Bartlett, J. R.; Woolfrey, J. L.; Ben-Nissan, B. Chem. Mater. 2002, 14, 4313. (16) Brinker, C. J.; Smith, D. M.; Deshpande, R.; Davis, P. M.; Hietala, S.; Frye, G. C.; Ashley, C. S.; Assink, R. A. Catal. Today 1992, 14, 155. (17) Moles, P. J. Catal. Today 1994, 20, R5. (18) Osendi, M. I.; Moya, J. S.; Serna, C. J.; Soria, J. J. Am. Ceram. Soc. 1985, 68, 135. (19) Fry, R. A.; Tsomaia, N.; Pantano, C. G.; Mueller, K. T. J. Am. Chem. Soc. 2003, 125, 2378. (20) Strynar, M.; Dec, J.; Benesi, A.; Jones, A. D.; Fry, R. A.; Bollag, J. M. Environ. Sci. Technol. 2004, 38, 6645. (21) Ballinger, P.; Long, F. A. J. Am. Chem. Soc. 1959, 81, 1050. (22) Maricq, M. M.; Waugh, J. S. J. Chem. Phys. 1979, 70, 3300. (23) Donia, A. M. Thermochim. Acta 1998, 320, 187. (24) Lu, G.; Purvis, K. L.; Schwartz, J.; Bernasek, S. Langmuir 1997, 13, 5791. (25) Stumm, W. Aquatic Surface Chemistry; John Wiley and Sons, Inc.: New York, 1987. (26) Brunauer, S.; Emmett, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309. 9463

dx.doi.org/10.1021/la202029e |Langmuir 2011, 27, 9458–9464

Langmuir

ARTICLE

(27) Sing, K. S. W.; Everett, D. H.; Haul, R. A. W.; Moscou, L.; Pierotti, R. A.; Rouquerol, J.; Siemieniewska, T. Pure Appl. Chem. 1985, 57, 603. (28) Chadwick, A. V.; Mountjoy, G.; Nield, V. M.; Poplett, I. J. F.; Smith, M. E.; Strange, J. H.; Tucker, M. G. Chem. Mater. 2001, 13, 1219.

9464

dx.doi.org/10.1021/la202029e |Langmuir 2011, 27, 9458–9464