Synthesis, Characterization, and Adsorption ... - ACS Publications

Aug 18, 2004 - Industrial & Engineering Chemistry Research 0 (proofing), .... Synthesis of cytocompatible Fe 3 O 4 @ZSM-5 nanocomposite as magnetic re...
0 downloads 0 Views 109KB Size
Langmuir 2004, 20, 8301-8306

8301

Synthesis, Characterization, and Adsorption Properties of Nanocrystalline ZSM-5 W. Song, R. E. Justice, C. A. Jones, V. H. Grassian,* and S. C. Larsen* Department of Chemistry, University of Iowa, Iowa City, Iowa 52242 Received February 24, 2004. In Final Form: June 28, 2004 Nanocrystalline ZSM-5 with a Si/Al ratio of 20 was synthesized using clear solutions and a hydrothermal synthesis procedure. The resulting ZSM-5 materials were characterized by powder X-ray diffraction, scanning electron microscopy (SEM), nitrogen adsorption isotherms, solid-state nuclear magnetic resonance, and toluene adsorption. A commercial ZSM-5 sample was similarly characterized for comparison with the synthesized materials. The particle sizes of the synthesized ZSM-5 samples were calculated using the measured external surface areas and were determined to be 15 and 60 nm. SEM images indicated that the ZSM-5 samples consist of agglomerated and possibly intergrown particles. Toluene adsorption measurements showed that the ZSM-5 sample with a particle size of 15 nm adsorbed approximately 50% more toluene than the other ZSM-5 samples, most likely due to the adsorption of toluene on the external surface. For the toluene adsorbed on the internal zeolite surface, approximately one toluene molecule was adsorbed per channel intersection for each of the ZSM-5 samples.

Introduction Microporous zeolite materials are widely used in applications such as catalysis and chemical separations and as adsorbents. Zeolites are unique materials with pores of various shapes, sizes, and topologies that control the reactivity and shape selective properties of these materials. Zeolites with discrete uniform particle sizes of less than 100 nm are referred to as nanocrystalline zeolites. Nanocrystalline zeolites have two dimensions in the nanometer size regime: the pores which are nanometer to subnanometer in size and the crystal sizes which are also in the nanometer size range (100 nm or below). Nanocrystalline zeolites are promising catalytic and adsorbent materials that have higher surface areas and reduced diffusion path lengths relative to conventional micrometer-sized zeolites. Nanocrystalline zeolites can also be assembled into thin films and other porous nanostructures for use as separation membranes, chemical sensors, and photochemical hosts.1-6 ZSM-5 is a zeolite that is often used as a catalyst in petroleum refining. ZSM-5 has an intersecting pore network composed of straight channels and zigzag channels with pore diameters of ∼0.55 nm. Nanocrystalline ZSM-5 exhibits increased selectivity and toluene conversion into cresol and decreased coke formation relative to conventional ZSM-5 materials.7 Several strategies for the synthesis of nanocrystalline ZSM-5 zeolites have been * To whom correspondence should be addressed. E-mail: [email protected], [email protected]. Phone: 319-3351346. Fax: 319-335-1270. (1) Metzger, T. H.; Mintova, S.; Bein, T. Microporous Mesoporous Mater. 2001, 43, 191-200. (2) Mintova, S.; Bein, T. Microporous Mesoporous Mater. 2001, 50, 159-166. (3) Huang, L.; Wang, Z.; Sun, J.; Miao, L.; Li, Q.; Yan, Y.; Zhao, D. J. Am. Chem. Soc. 2000, 122, 3530-3531. (4) Jung, K. T.; Hyun, J. H.; Shul, Y. G. Zeolites 1997, 19, 161-168. (5) Wang, Y. J.; Tang, Y.; Wang, X. D.; Yang, W. L.; Gao, Z. Chem. Lett. 2000, 1344-1345. (6) Tang, Y.; Wang, Y. J.; Wang, X. D.; Yang, W. L.; Gao, Z. In Zeolites and Mesoporous Materials at the Dawn of the 21st Century: Proceedings of the 13th International Zeolite Conference; Studies in Surface Science and Catalysis, Vol. 135; Elsevier: Amsterdam, 2001; pp 3325-3332. (7) Vogel, B.; Schneider, C.; Klemm, E. Catal. Lett. 2002, 79, 107112.

described in the literature.8-17 Verduijn13 and Schoeman and co-workers15,18 reported the synthesis of nanocrystalline ZSM-5 using clear solutions in open beakers at low temperature and ambient pressure in contrast to the conventional hydrothermal routes. These syntheses resulted in the formation of colloidal solutions of zeolites with uniform, nanometer particle sizes of 100 nm or less. Van Grieken and co-workers recently reported a similar synthesis for ZSM-5 based on clear solutions under hydrothermal conditions.17 Reding and co-workers recently surveyed the synthetic methods for nanocrystalline ZSM-5 and reported a new procedure based on two crystallization steps, both under atmospheric pressure, that yielded crystals with sizes ranging from 100 to 300 nm.12 Alternative strategies based on confined space and templating methods have been adopted by other groups.8,16 Jacobsen and co-workers devised a method in which the zeolite synthesis mixture was impregnated into a porous carbon black material.16,19 The growth of the zeolite crystals was then restricted by the pore size of the carbon black matrix. This method was hindered by the fact that the carbon black material is disordered and therefore zeolites with rather broad size and shape distributions were produced. Pinnavaia recently reported a method for (8) Kim, S. S.; Shah, J.; Pinnavaia, T. J. Chem. Mater. 2003, 15, 1664-1668. (9) Yamamura, M.; Chaki, K.; Wakatsuki, T.; Okado, H.; Fujimoto, Z. Zeolites 1994, 14, 643-649. (10) Lovallo, M. C.; Tsapatsis, M. Advanced Catalysis and Nanostructure Materials; Academic Press: San Diego, 1996. (11) Mintova, S.; Bein, T. Adv. Mater. 2001, 13, 1880-1883. (12) Reding, G.; Maurer, T.; Kraushaar-Czarnetzki, B. Microporous Mesoporous Mater. 2003, 57, 83-92. (13) Verduijn, J. P. Exxon Chemical Patents Inc., Patent No. WO97103019, 1993. (14) Mintova, S.; Petkov, N.; Karaghiosoff, K.; Bein, T. Mater. Sci. Eng., C 2002, C19, 111-114. (15) Schoeman, B. J.; Sterte, J.; Ottstedt, J. E. Chem. Commun. 1993, 994-995. (16) Schmidt, I.; Madsen, C.; Jacobsen, C. J. H. Inorg. Chem. 2000, 39, 2279-2283. (17) Van Grieken, R.; Sotelo, J. L.; Menendez, J. M.; Melero, J. A. Microporous Mesoporous Mater. 2000, 39, 135-147. (18) Schoeman, B. J.; Sterte, J.; Ottstedt, J. E. Zeolites 1994, 14, 110-116. (19) Jacobsen, C. J. H.; Madsen, C.; Janssens, T. V. W.; Jakobsen, H. J.; Skibsted, J. Microporous Mesoporous Mater. 2000, 39, 393-401.

10.1021/la049516c CCC: $27.50 © 2004 American Chemical Society Published on Web 08/18/2004

8302

Langmuir, Vol. 20, No. 19, 2004

Song et al. Table 1. Synthesis Conditions for ZSM-5

samplea

gel composition

vessel

temp (°C)

time (h)

Si/Al

ZSM-5 (15 nm) ZSM-5 (60 nm) Zeolyst ZSM-5

9TPAOH/0.16NaOH/Al/25Si/495H2O/100EtOH 9TPAOH/0.16NaOH/Al/25Si/300H2O N/A

autoclave autoclave N/A

165 165 N/A

120 120 N/A

20 20 20

a

The number in parentheses denotes the average particle size (nm) of the sample calculated from the external surface area.

synthesizing highly uniform ZSM-5 crystals using colloidimprinted carbon templates (CICs).8 The method is based on the imprinting of carbon pitch with colloidal silica to create a carbon material with uniform pore sizes. The ZSM-5 was then synthesized in the uniform carbon pores followed by burning off the porous carbon. ZSM-5 with systematically varied particle sizes of 13, 22, 42, and 90 nm were synthesized using the CIC method.8 In the work reported here, nanocrystalline ZSM-5 was synthesized using modifications of the clear solution based procedures. The procedures were modified to produce ZSM-5 with uniform crystals with sizes as small as 15 nm and with Si/Al ) 20, which is a lower ratio than previously reported. A commercial ZSM-5 material was used for comparison of the properties of the synthesized nanocrystalline ZSM-5 materials. The resulting materials were extensively characterized by powder X-ray diffraction (XRD), scanning electron microscopy (SEM), nitrogen adsorption isotherms, and solid-state nuclear magnetic resonance (NMR). The adsorption properties for a representative volatile organic compound (VOC), toluene, on ZSM-5 samples with different particle sizes were also investigated. Experimental Section Materials. Tetraethyl orthosilicate (TEOS), tetrapropylammonium hydroxide (TPAOH, 40 wt % aqueous solution), and aluminum isopropoxide were purchased from Alfar Aesar. Sodium hydroxide (NaOH) was purchased from Aldrich. A commercial ZSM-5 sample (Si/Al ) 20) was provided by Zeolyst International. Nanocrystalline ZSM-5 Synthesis. All synthesized ZSM-5 samples were prepared from clear gel solutions following the general procedure described by Van Grieken and co-workers.17 Detailed conditions such as gel composition, synthesis vessel, temperature, and time duration are listed in Table 1. The relative ratio of reagents in the synthesis gel was similar to that in our previous report on silicalite-1 synthesis,20 except that an aluminum source, aluminum isopropyloxide, was added. ZSM-5 samples synthesized in this study have a starting Si/Al ) 25 which is lower than that used by Van Grieken (Si/Al ) 60).17 Overall, the general procedure follows Van Grieken, but the specific synthesis gel composition and synthesis conditions were modified to prepare nanocrystalline ZSM-5 with a lower Si/Al ratio. To start a synthesis, measured amounts of chemicals (H2O, NaOH, TPAOH solution, TEOS, and aluminum isopropyloxide) were mixed and stirred at room temperature (RT) overnight to ensure complete TEOS and aluminum isopropyloxide hydrolysis to ethanol and isopropyl alcohol, respectively. For the larger particle size (60 nm), the ethanol and isopropyl alcohol were removed by heating the solution to 80 °C until the weight loss expected for complete alcohol removal was reached. For the smaller particle size (15 nm), the alcohols were not removed from the solution. Next, the clear solution was transferred into a 25 mL autoclave equipped with a Teflon liner for the hydrothermal treatment specified in Table 1. ZSM-5 crystals were recovered after three cycles of centrifuging and washing with deionized water, followed by drying at 120 °C. The ZSM-5 crystals were calcined at 600 °C under air flow to remove the TPAOH template. The ZSM-5 samples are labeled according to average particle size as calculated from their external surface areas, measured (20) Song, W.; Justice, R. E.; Jones, C. A.; Grassian, V. H.; Larsen, S. C. Langmuir 2004, 20, 4696-4702.

using the Brunauer-Emmett-Teller (BET) method. For example, the sample labeled ZSM-5 (15 nm) consists of ZSM-5 with an average particle size of 15 nm. Scanning Electron Microscopy. SEM images of the ZSM-5 crystals were acquired using a Hitachi S-4000 scanning electron microscope. To prepare the sample for SEM, a drop of dilute colloidal solution of the sample was dropped onto the SEM sample stud surface and the sample stud was then dried at 60 °C for 3 h. Shortly before an SEM image was acquired, the sample was coated with gold. Elemental Analysis. A Perkin-Elmer Plasma 400 inductively coupled plasma atomic emission spectrometer (ICP/AES) was used to determine the Si/Al ratio of the ZSM-5 samples. ZSM-5 samples were acid digested by dilute HF solution followed by neutralization in NaBO3. Four standard solutions with known silicon and aluminum concentrations were prepared as calibration standards. Atomic emission peak intensities at wavelengths of 309.2 nm for aluminum and 203.9 nm for silicon were recorded for all four standard solutions and the sample solutions. Exact concentrations of aluminum and silicon in the sample solution were obtained by projection from the working curve generated from standard solution data. X-ray Diffraction. A Siemens D5000 X-ray diffractometer with a Cu KR target and a nickel filter was used to collect XRD powder patterns for the samples. XRD patterns were collected between 2θ angles of 5 and 55°. To estimate crystal sizes from Scherrer’s equation, a slower, high-resolution scan between 7.7 and 7.8° for each sample was also collected. The full width at half-maximum (fwhm) of the peak from the slow scan of each sample was obtained by simulation and was used to estimate the crystal size of each ZSM-5 sample from Scherrer’s equation (with K ) 1.0) given below:

T)

Kλ β cos θ

where T is the crystal size (nm), K is the crystal shape factor, λ is the wavelength of X-rays (for the Cu target, λ ) 1.542 Å), β is the full width at half-maximum, and θ is Bragg’s angle. Nitrogen Adsorption Isotherms. Nitrogen adsorption isotherms were obtained on a Quantachrome Nova 1200 multipoint BET apparatus using approximately 0.2 g of sample for each measurement. Immediately prior to the N2 adsorption, each sample was vacuum degassed at 120 °C for 1 h. The specific surface area was measured by the BET method, which was performed automatically by the instrument. BET adsorption isotherms were collected for ZSM-5 samples before and after calcination to remove the template. Solid-State Magic Angle Spinning NMR. 29Si (59.595 MHz) and 27Al (78.172 MHz) solid-state magic angle spinning (MAS) NMR spectra were obtained using a 300 MHz wide bore magnet with a TecMag Discovery Console with a Chemagnetics doubleresonance 7.5 mm pencil MAS probe with a spinning speed of ∼6 kHz. Typically, 0.2 g of sample was used to load the rotor. Each single pulse spectrum was acquired by signal averaging 1000 scans with a 3 s pulse delay for 27Al and 1000 scans with a 60 s pulse delay for 29Si. Probe Molecule Adsorption. Toluene was selected to examine VOC adsorption as a function of the particle size of ZSM-5. The experimental setup has been described previously.20 Each experiment consists of three regions: Region I, roomtemperature adsorption; Region II, room-temperature purge with helium; and Region III, temperature-programmed desorption (TPD). During each experiment, 0.1 g of calcined ZSM-5 sample was loaded into a 1/4 in. quartz tube. The sample bed was capped

Nanocrystalline ZSM-5

Langmuir, Vol. 20, No. 19, 2004 8303

Table 2. Particle Sizes of ZSM-5 Determined from SEM Images and XRD Line Widths

sample

BET area (as synthesized)a

BET area (calcined)a

ZSM-5 (15 nm) ZSM-5 (60 nm) Zeolyst ZSM-5

208 53 N/A

556 419 396

differenceb

particle size (calculated)c

XRD Scherrer particle sized (nm)

348 363 N/A

15 60 N/A

14 28 32

a All surface areas are provided in m2/g. Standard deviations of surface areas were measured to be between 0.4 and 1.4 m2/g; the larger errors are associated with the higher surface areas measured. b Difference ) BET area (calcined) - BET area (as synthesized). c Calculated particle sizes are obtained using the following equation: particle size (nm) ) 3216/BET surface area of as-synthesized ZSM-5 samples. d K ) 1.0 in Scherrer’s equation.

Figure 1. XRD patterns of three ZSM-5 samples: (a) Zeolyst ZSM-5; (b) ZSM-5 (60 nm); (c) ZSM-5 (15 nm). 2θ angles between 5 and 55° were collected. with quartz wool to minimize breakthrough. The sample bed was heated at 300 °C under helium prior to the adsorption at room temperature. During room-temperature adsorption, 25 sccm helium was directed through a toluene bubbler and then through the sample bed. After the sample was saturated with toluene, the toluene bubbler was bypassed to start purging the sample by helium flow at room temperature; then the sample was heated at 5 K/min to do the TPD experiment. Throughout the experiment, the concentration of toluene in the gas stream was monitored by the thermal conductivity detector (TCD) of a Varian 3400CX gas chromatograph (GC), which was calibrated to give a quantitative measurement of toluene passing through the detector.

Results and Discussion

Figure 2. SEM images of three ZSM-5 samples: (a) Zeolyst ZSM-5; (b) ZSM-5 (60 nm); (c) ZSM-5 (15 nm). In each image, the scale bar is equal to 100 nm.

Synthesis and Characterization of Nanocrystalline ZSM-5. The two nanocrystalline ZSM-5 samples have different average particle sizes and were synthesized by varying the relative amounts of water in the synthesis solution as listed in Table 1. In addition, the alcohol hydrolysis products were removed prior to the thermal synthesis treatment for the larger ZSM-5 crystal size, but not for the smaller crystal size sample. Aluminum was added to the synthesis gel in order to prepare ZSM-5 with a Si/Al ratio of 20. The powder XRD patterns of two synthesized ZSM-5 samples and a Zeolyst ZSM-5 sample for comparison are shown in Figure 1. The diffraction patterns agree with the diffraction pattern expected for ZSM-5.21 Close inspection of the diffraction patterns in Figure 1 reveals that the line widths decrease from top to bottom, suggesting that the ZSM-5 particle size is increasing such that the Zeolyst ZSM-5 sample has the largest particle size. The particle sizes estimated from the XRD line width and Scherrer’s equation are listed in Table 2 with the shape factor K equal to 1.0. Using Scherrer’s equation, the particle sizes are estimated to be 14, 28, and 32 nm for the two synthesized ZSM-5 samples and the Zeolyst sample, respectively.

The SEM images of the ZSM-5 samples are shown in Figure 2. The SEM images show the size, particle morphology, and aggregation of the synthesized and commercial ZSM-5 crystals. The ZSM-5 particles grow into larger aggregates as shown in the SEM images, and it is therefore difficult to determine the primary particle size from the SEM images alone. The larger aggregates may be composed of many individual zeolite particles or of intergrown zeolite particles. The larger aggregates range from 100-200 nm for ZSM-5 (15 nm) to 400-600 nm for ZSM-5 (60 nm) to 700-1000 nm for Zeolyst ZSM-5. Nitrogen adsorption isotherms for the synthesized ZSM-5 samples were measured both before and after calcination to remove the template. Total surface areas of both calcined and as-synthesized samples were obtained from the nitrogen adsorption isotherms using the BET method and are listed in Table 2. The total surface area of the as-synthesized samples in which the internal surface is blocked by template molecules represents the external surface area of the zeolite sample.20,22 The total surface area obtained from the calcined samples should contain

(21) Meier, W. M.; Olson, D. H. Atlas of Zeolite Structure, 2nd revised ed.; Butterworth: Cambridge, 1987.

(22) Camblor, M. A.; Corma, A.; Valencia, S. Microporous Mesoporous Mater. 1998, 25, 57-74.

8304

Langmuir, Vol. 20, No. 19, 2004

Figure 3. Correlation between external surface area in m2/g and particle size in nm for ZSM-5. The solid line represents the calculated external surface area as a function of particle size assuming a cubic particle as described in the text. The open triangles represent the data points for the two ZSM-5 samples synthesized in this study assuming the relationship between external surface area and particle size described in the text. The dotted lines indicate how the theoretical curve is used to determine the particle size from the external surface area for the two synthesized ZSM-5 samples.

contributions from both the internal and the external surfaces. Therefore, the difference between the surface area of the uncalcined and the calcined ZSM-5 should provide the internal surface area. For the Zeolyst ZSM-5 sample, only the total surface area was measured since the material was only available in the calcined form. Examination of the data in Table 2 shows that the total surface area of the calcined ZSM-5 samples increases as particle size decreases, as expected. The total surface area can be separated into contributions from the internal and external surface area. The internal surface area obtained by taking the difference in total surface area of the calcined and the as-synthesized ZSM-5 samples is constant with an average value of approximately 355 m2/g. The external surface area increases from a value of approximately 53 to 208 m2/g for the two synthesized ZSM-5 samples. Only the total surface area of 396 m2/g was obtained for the Zeolyst sample. Using a simple approximation for external surface area, the particle size for the synthesized ZSM-5 samples can be estimated from the external surface areas measured from the BET method.20 To establish a correlation between external surface area and the particle size, a simple model assuming a cubic particle shape for ZSM-5 was used. Then, a particle with a size of x nm will have an external surface area of 6x2 (nm2). ZSM-5 has the MFI framework structure with a unit cell volume of 5.21 nm3 and a chemical formula for the unit cell of 92Si/4Al/192O, which gives a unit cell formula weight of 5852. For a ZSM-5 particle with x nm size, the chemical formula weight can then be calculated to be (x3/5.21) × 5852 ) 1123x3. For 1 g of ZSM-5, the number of crystals is 6.02 × 1023/1123x3, and the total external surface area can then be calculated as 6x2 × 6.02 × 1023/1123x3. After adjusting the units, a particle of size x of a ZSM-5 sample is

x ) 3216/Sext where Sext is the external surface area in m2/g and x is the ZSM-5 particle size in nm. Using this relationship, the particle size can be estimated from external surface area as illustrated in Figure 3 and listed in Table 2. Using the values for the external surface area of 208 and 53 m2/g,

Song et al.

Figure 4. 29Si solid-state NMR spectra of calcined ZSM-5 samples: (a) Zeolyst ZSM-5; (b) ZSM-5 (60 nm); (c) ZSM-5 (15 nm).

Figure 5. 27Al solid-state magic angle spinning NMR spectra of as-synthesized and calcined ZSM-5 samples: (a) calcined Zeolyst ZSM-5; (b) as-synthesized ZSM-5 (60 nm); (c) assynthesized ZSM-5 (15 nm); (d) calcined ZSM-5 (60 nm); (e) calcined ZSM-5 (15 nm).

the particle sizes for the synthesized ZSM-5 samples were determined to be 15 and 60 nm, respectively. An analogous relationship was derived for nanocrystalline silicalite, and the experimentally determined external surface areas and particle sizes (from SEM) agreed remarkably well with the results of the calculation assuming cubic silicalite crystals.20 The method of using the external surface area to determine the particle size for the nanocrystalline zeolites appears to provide a very good approximation of the particle size. The particle size obtained from the external surface area is similar to the particle size obtained from Scherrer’s equation for the smallest synthesized ZSM-5 sample but deviates significantly for the larger sample. This was also found to be the case in our study on nanocrystalline silicalite-120 and was attributed to the intergrowth of multiple particles.22 As observed in the SEM images in Figure 2, the intergrown particles are present in these samples. Solid-State 29Si and 27Al MAS NMR. The 29Si MAS NMR spectra of calcined ZSM-5 samples are shown in Figure 4. Peaks at -110 and -100 ppm are observed and can be assigned to Q4 and Q3 silicon atoms, respectively. The Qn nomenclature is used, where n represents the number of siloxane bonds and (4 - n) provides the number of silanol (SiOH) groups. There are at least two kinds of Q3 sites: a siloxy site, which is present in as-synthesized

Nanocrystalline ZSM-5

Langmuir, Vol. 20, No. 19, 2004 8305

Figure 6. Toluene adsorption on nanocrystalline ZSM-5 and commercial ZSM-5. Each experiment consists of three regions: Region I, room-temperature adsorption; Region II, room-temperature purge with helium; and Region III, temperature-programmed desorption. Solid line, ZSM-5 (15 nm); dotted line, ZSM-5 (60 nm); dashed line, Zeolyst ZSM-5.

samples to compensate the charge from the templating agent, and a terminal silanol site.23-25 Only a small number of Q3 sites remain after calcinations as can be seen in Figure 4. The 27Al MAS NMR spectra of calcined and assynthesized ZSM-5 samples are shown in Figure 5. Peaks at ∼50 and ∼0 ppm in 27Al NMR spectra are assigned to aluminum in tetrahedral sites and extraframework aluminum in octahedral coordination, respectively. The main peak present in the 27Al NMR spectra of as-synthesized and calcined samples is at ∼50 ppm and is assigned to aluminum in tetrahedral sites. This peak broadens as the particle size decreases, as has been observed previously for HZSM-5.26 A second peak is present in the 27Al NMR spectrum of Zeolyst ZSM-5 at approximately 0 ppm and is assigned to extraframework aluminum. Very little extraframework aluminum is observed for the ZSM-5 samples synthesized as part of this study. VOC Adsorption. The adsorption of a typical VOC, toluene, on ZSM-5 was investigated using the flow apparatus described previously.20 Representative results from toluene adsorption/desorption on ZSM-5 samples are shown in Figure 6. The TCD signal of the GC is at a maximum when the sample bed is bypassed or when the sample is saturated and at a minimum when no toluene passes through the sample bed. In Region I of the experiment, toluene adsorption on the ZSM-5 sample was monitored. In Region II of the experiment, the ZSM-5 sample bed was purged with helium while monitoring the desorption of toluene, and in Region III of the experiment, a temperature-programmed desorption of the toluene was conducted. The amount of toluene adsorbed was obtained by measuring the area between the maximum baseline and the experimental curve, while the amount of toluene desorbed was obtained from the area between the minimum baseline and the experimental curve profiles after the temperature ramp. Using this method, the amounts of toluene adsorbed and desorbed in this experiment were calculated and are listed in Table 3. The total toluene adsorption in Region I of the experiment ranged from 1.65 to 1.04 to 1.03 mmol/g for the ZSM-5 (23) Koller, H.; Lobo, R. F.; Burkett, S. L.; Davis, M. E. J. Phys. Chem. 1995, 99, 12588-12596. (24) Dessau, R. M.; Davis, M. E.; Kerr, G. T.; Wooley, G. L.; Alemany, L. B. J. Catal. 1987, 104, 484-489. (25) Kragten, D. D.; Fedeyko, J. M.; Sawant, K. R.; Rimer, J. D.; Valachos, D. G.; Lobo, R. F. J. Phys. Chem. B 2003, 107, 10006-10016. (26) Zhang, W.; Bao, X.; Wang, X. Catal. Lett. 1999, 60, 89-94.

Table 3. Amount of Toluene Adsorption/Desorption on ZSM-5

samples

adsorption (mmol/g)

RT purge (mmol/g)

TPD (mmol/g)

total desorption (mmol/g)

ZSM-5 (15 nm) ZSM-5 (60 nm) Zeolyst ZSM-5

1.65 1.04 1.03

1.07 0.46 0.46

0.66 0.65 0.67

1.73 1.11 1.13

(15 nm), ZSM-5 (60 nm), and Zeolyst ZSM-5 samples, respectively. The amount of toluene removed during the helium purge was 1.07 mmol/g for ZSM-5 (15 nm), 0.46 mmol/g for ZSM-5 (60 nm), and 0.46 mmol/ g for Zeolyst ZSM-5. The amount of toluene desorbed during the roomtemperature helium purge trends with the external surface area for the synthesized ZSM-5 samples, suggesting that the toluene desorbed during the helium purge is due to toluene adsorbed on the external surface of the ZSM-5. The area under the desorption peak in Region III of the experiment is approximately the same for all three samples and is 0.65-0.67 mmol/g. This corresponds extremely well with the calculated density of channel intersections of the ZSM-5 framework which is 0.68 mmol/ g, suggesting that approximately one toluene molecule is adsorbed per channel intersection of ZSM-5. In a previous study of toluene adsorption on siliceous ZSM-5, toluene was found to adsorb at channel intersections at low toluene loadings (1-4 molecules/unit cell) and at both channel intersections and midsections of channels at higher toluene loadings (>4 molecules/unit cell).27 However, while the area of the TPD peak is approximately the same for all three samples, the shape of the TPD peak is different for the different samples. The TPD peak for ZSM-5 (60 nm) and Zeolyst ZSM-5 has two maxima, suggesting that the toluene is adsorbed at two different sites in the zeolite. However, the TPD peak for ZSM-5 (15 nm) only has the lower temperature desorption component. The shapes of the peaks suggest that the toluene is adsorbing at different sites in the zeolite such as proton acid sites versus sodium cation sites or channel intersection sites versus channel sites. Given that the concentration of adsorbed toluene corresponds so well with the number of channel intersections, it seems most likely that the toluene is adsorbed at different sites in the channel intersection such as proton acid sites or sodium cation sites. (27) Huang, Y.; Havenga, E. A. Chem. Mater. 2001, 13, 738-746.

8306

Langmuir, Vol. 20, No. 19, 2004

These results indicate that the overall adsorption capacity of ZSM-5 (15 nm) is approximately 50% greater than that of the other samples tested, largely due to the higher external surface area and the weakly adsorbed toluene that desorbs during the room-temperature helium purge of the sample. The strongly adsorbed toluene that desorbs during the temperature-programmed desorption is quantitatively the same for all of the samples, suggesting that this toluene is adsorbed on the internal zeolite surfaces, most likely at the channel intersections.27 Conclusions Nanocrystalline ZSM-5 samples with different particle sizes and Si/Al ) 20 were synthesized. The resulting ZSM-5 samples had particle sizes of 15 and 60 nm and were compared to a commercial ZSM-5 sample. For each of the samples, the framework structures, particle sizes, internal and external surface areas, and VOC adsorption properties were characterized by XRD, SEM, solid-state NMR, and N2 and toluene adsorption. The external surface areas of the samples were determined from the BET measurement and were used to determine the particle sizes for the two synthesized ZSM-5 samples. The external surface areas

Song et al.

of as-synthesized samples were 208 and 53 m2/g for the 15 and 60 nm samples, respectively. No extraframework aluminum was detected by 27Al MAS NMR of the synthesized nanocrystalline ZSM-5 zeolites in contrast to the commercial ZSM-5 sample, which exhibited a peak due to extraframework aluminum. ZSM-5 with a 15 nm average particle size showed significantly higher adsorption capacity for toluene relative to the other ZSM-5 samples examined in this study due to adsorption of toluene on the external zeolite surface. The toluene adsorbed on the internal surface was the same for all three ZSM-5 samples examined and quantitatively corresponded to the number of channel intersections in ZSM-5, suggesting that the toluene adsorbed on the internal zeolite surface was preferentially located at the channel intersections. Acknowledgment. The research described in this article has been funded by the Environmental Protection Agency through EPA Grant No. R82960001 to S.C.L. and V.H.G. Gonghu Li is acknowledged for assistance in setting up the flow reactor used for the adsorption experiments. LA049516C