The recently discovered mesoporous molecular sieve MCM-41 was synthesized, modified, and characterized and proposed as an alternative adsorbent for VOC control. The synthesis conditions for pure-silica and aluminosilicate MCM-41 were optimized as follows: 4.5Na2O:30SiO2:5.2C16H33(CH3)3N + :2500H2O and 7.5Na2O:30SiO2:xAl2O3:7.2C16H33(CH3)3N + :3500H2O (x < 1), respectively, and at 373 K for 4 days. Our studies showed that MCM-41 is not stable in the presence of water vapor. For example, a hydrothermal treatment of MCM-41 at 723 K for 2 hour resulted in 50 % of structure collapses. Again, when a template-free MCM-41 sample was exposed to air with a relative humidity of 60 % for three months, almost total pore structure collapses were observed. Adsorption equilibrium results showed that MCM-41 has a narrow pore size distribution and exhibits extraordinary pore volume compared to the classical microporous adsorbents, such as molecular sieves and activated carbons. Despite the impressive adsorption capacities of this material, the Type IV isotherm behavior requires the VOCs, in the gas phase, to be at high partial pressure. This is not the case with most industrial VOC streams. A real VOC stream requires an adsorbent with not only a high adsorption capacity but also a high adsorption affinity at a low VOC concentration. To overcome the above mentioned two problems, both the surface chemistry and the pore-opening sizes of MCM-41 were modified. To modify the surface chemistry, one has to better understand the surface chemistry. Our pioneering study of the surface chemistry of MCM-41 using FTIR, 29 Si CP/MAS NMR, pyridine-TPD, and TGA demonstrated that three types of silanol groups, i.e. single, (SiO)3Si-OH, hydrogen-bonded, (SiO)3Si-OH---OH-Si(SiO)3 and geminal, (SiO)2Si(OH)2 are distributed over the surface of MCM-41. The number of silanol groups per unit nm 2 , aOH, varies between 2.5 and 3.0 depending on the template-removal method. To improve the hydrothermal stability and enhance the hydrophobicity, the surface chemistry of MCM-41 was modified by silylation. Though both the free and hydrogen-bonded SiOH groups were found to be the active sites for adsorption of pyridine with desorption energies of 91.4 and 52.2 kJ mol -1 , respectively, only the free SiOH groups are highly accessible to the silylating agent, chlorotrimethylsilane. The surface coverage of the modifying agent was found to has a linear relationship with the surface free silanol groups which can be controlled by different heating temperatures. Modification by silyaltion can significantly improve hydrophobicity and stability. Rehydration/dehydration experiments demonstrate that the surface-silylated MCM-41 is highly tolerable to water vapor due to the complete replacement of surface-hydrophilic silanols. A novel modification method, namely selective tailoring (ST), was developed to tailor the pore-opening sizes of MCM-41 (rather than the entire pores). The novelty is that only the pore mouths at both ends of a cylindrical pore of MCM-41 was modified by deposition of some alkoxides. By doing so, the types of adsorption isotherms of VOCs can be changed from Type IV to Type I while the pore volume can be significantly preserved. This is of course significance in VOC removal since the adsorption affinity has been drastically enhanced. Adsorption equilibria and kinetics for VOCs in the pore-opening-modified MCM-41 materials were measured, modeled and compared to that of activated carbons and hydrophobic molecular sieves. The pore-modified MCM-41 has a much higher adsorption capacity than that of the traditional microporous adsorbents such as activated carbons and molecular sieves. The adsorption equilibrium data fit the Langmuir-Uniform distribution (Unilan) models very well. Upon the equilibrium parameters being obtained and considering the pore structure of our pore-modified MCM-41 adsorbents, the kinetic data were further modeled using the literature-existed models recently developed by Do and coworkers, i.e. the constant surface diffusivity macropore, surface and micropore diffusion (CMSMD) model and the macropore and surface diffusion (MSD) model. Results demonstrated that the CMSMD model can predict our kinetic uptake curves reasonably fine. Some key kinetic parameters including pore and surface diffusivities, apparent diffusivity, activation energy for adsorption, and pore tortuosity factor can be readily obtained. The porosity of the MCM-41 materials were primarily evaluated using the traditional methods based on nitrogen adsorption/desorption data. Results indicated that the BJH method always underestimates the true pore diameter of MCM-41. An comparison plot (t-plot or as-plot) method was suggested and improved. Plotting of nitrogen adsorption data at 77 K versus the statistical film thickness reveals three distinct stages, with a characteristic of two points of inflection. The steep intermediate stage is caused by capillary condensation occurred in the highly uniform mesopores. From the slope of the section after condensation, the external surface area can be obtained. Therefore, the true surface area of the mesopores is readily calculated. The linear portion of the last section is extrapolated to the adsorption axis of the comparison plot, and this intercept is used to obtain the volume of the mesopores. From the surface area and pore volume, average mesopore diameter is calculated, and the value thus obtained is in good agreement with the pore dimension obtained from powder X-ray diffraction measurements. The principle of pore size calculation, the thickness of adsorbed nitrogen film, and the problems associated with the BJH method were discussed in detail. It has been found that at a given relative pressure, the smaller the pore radius, the thicker the adsorbed film. Thermodynamics analysis established that the stability of the adsorbed film is determined by interface curvature and the potential of interaction between adsorbate and adsorbent. A semi-empirical equation is proposed to describe the state of stable adsorbed films in cylindrical mesopores. It is also shown to be useful in calculations of pore size distributions of mesoporous solids. The desorption of four representative volatile organic compounds (VOCs), i.e. n-hexane, cyclohexane, benzene, and methanol from MCM-41 were also investigated and compared with the hydrophobic zeolite, silicalite-1, using the technique of temperature programmed desorption (TPD). The desorption energies of these organics to MCM-41 were evaluated and compared with the adsorption isosteric heats. The affinity of organics to MCM-41 and silicalite-1, which represents surface hydrophobicity/hydrophilicity were studied and discussed. Results showed that only one desorption peak can be found for all organics from MCM-41, different from that from the microporous adsorbents (activated carbons and hydrophobic molecular sieves). The activation energies for desorption of non-polar molecules are slightly higher than their latent heats of evaporation, whereas the activation energy for desorption of methanol is well above its latent heat of evaporation. These results are consistent with those derived from the adsorption isotherm measurements. The very high activation energy for the desorption of methanol is due to the hydrogen bonds between methanol molecules and silanol groups over MCM-41 surfaces. The affinity of volatile organics to MCM-41 are in the order of methanol > n-hexane > benzene > cyclohexane.
Identifer | oai:union.ndltd.org:ADTP/253807 |
Creators | Zhao, Xiusong |
Source Sets | Australiasian Digital Theses Program |
Detected Language | English |
Page generated in 0.0022 seconds