Gasification of biomass residue, a thermochemical conversion process, is fast becoming an attractive method of syngas production by extracting energy from the non-food based biomass feedstocks. Unlike the more popular biochemical processes, like fermentation of cellulosic biomass, it can operate with a wide range of biomass-feedstock in terms of feed’s quality and consistency, thus expanding the available pool of feedstock and thereby reducing the overall costs. Moreover, by doing almost complete depolymerization of the source biomass to useful gaseous products like syngas, gasification currently is the only established technology that can theoretically make the most of the energy stored in the raw biomass.However, on gasification, along with the syngas, gaseous impurities like sulfur containing species (H2S, COS), ammonia, alkali oxides, halides etc are also generated due to the volatile contaminants present in the biomass residue. Many downstream processes, like Fischer Tropsch synthesis, Solid Oxide Fuel Cells and methanol production, use catalysts that have little tolerance with gaseous contaminants. Transition metal / metal oxide catalysts, typically employed for such value-addition processes, are especially vulnerable to sulfur containing gaseous species like hydrogen sulfide. Therefore, before this raw syngas can be used for different downstream applications, it needs to be cleaned. Among the different processes used for H2S removal, wet amine scrubbing has been the most popular. However, such wet processes require comparatively low operating temperatures (35 - 55⁰C). The cleaned syngas has to be subsequently reheated for the downstream processes (300 - 800⁰C). Such consecutive cooling and heating can cause considerable thermal losses.To avoid such energy losses, it is important to desulfurize the gas stream at suitably high temperatures. A solid-phase sulfur-sorbent material that has a high reactivity, good structural stability, and easy regenerability at such high temperatures can provide such an alternative.In past work, different bulk sorbents with different chemical and structural properties have been tried for high temperature desulfurization.However, in spite of numerous efforts to modify chemical and compositional properties, only a limited success has been achieved.It is due to sorbent's failure in meeting one or more of the aforementioned criteria. For instance, in the past, it has been demonstrated that due to mass transfer limitations, if a bulk sorbent gets completely sulfided, it is almost impossible to achieve complete regeneration afterwards, regardless of process's favorable thermodynamics and chemical kinetics. This incomplete regeneration, partly due to mass transfer limitation and subsequent loss of sorbent's surface area, leads to underutilization of the sorbent material, which is typically a transition or rare earth metal oxide. To overcome these limitations, it seems necessary that one needs to go beyond doing modifications in sorbent chemical composition alone. One option that has been little explored involves integration of tailored design and morphology of the sorbent with the process's favorable kinetics. The attempt in this work will be to describe how the structure-based modification of sorbents, specifically nanostructuring, can help in achieving improved sorbent performance for a process requiring reusable sorbents. Since the reaction time has been shown to be comparatively short, the approach here will be to come up with a sorbent design that facilitates short contact time so that deep sulfidation of sorbent can be avoided without affecting the sulfur removal capacity. As this would require high gas velocity along with sorbent's high specific surface area, the use of conventional bulk sorbent will lead to incomplete regeneration and significant pressure drop. Sorbents with tailored designs and sizes, however, can help in overcoming such limitations. The criteria for such designs should be to maximize available specific surface area along with short diffusion lengths. Nanosizing of sorbent seems as a good alternative.However, sorbent in the form of nanopowders tend to aggregate and cause mass transfer limitations similar to bulk sorbents. On the other hand, nanostructures having high aspect ratio, like nanofibers, can remain isolated; thus, potentially providing a more suitable framework for carrying out frequent cyclic sulfidation-regeneration operation. These high aspect ratio nanoscale structures can not only retain the properties from their bulk form such as favorable thermodynamics, chemical affinity etc., but they also tend to develop useful properties due to highly anisotropic geometry and confined grain size. Because of the confinement of the grain size and short contacting time, nanofibers will tend to limit the large volume changes and accompanying grain boundary collisions which are typical during repeated sulfidation/regeneration. Thus, it is expected that the use of nanostructured sorbent will not only lead to high specific surface area and improved mass-transfer, it can also lead to an improved mechanical behavior during high temperature cyclic operation. This work will present the results from the experimental investigation of such high-aspect ratio nanofibers for their potential to serve as a regenerable sorbents. Sol-gel based electrospinning was used to synthesize composite zinc and titanium oxide adsorbents in the form of non-woven fiber-mats. Two such samples, with different zinc-to-titanium atomic ratio were selected for further investigation. The Zn-to-Ti ratio for the first sample (Sample-1) was 3.69 and for the second sample (Sample-2a) it was 1.17. Salt solutions of the respective metals, with PVP (Mw ~ 1300000) as the binding polymer, were used to prepare the corresponding sol-gels, which were then used as the precursor solutions for electrospinning of the fibers. Calcination of the as-spun fibers at 600⁰C for 4hrs resulted in zinc-titanate fiber-mats free from polymer. These fiber mats were characterized by substantially high specific surface areas: 151.7 and 90.1 m2/g respectively. The average fiber diameter for Sample-1 (post-sintering) was found to be 435 nm and for the second sample it was 714 nm. Fibers were then characterized for their internal crystal structure and surface morphology using the techniques of XRD, SEM and TEM. XRD results and selected-electron-diffraction done using TEM revealed that the electrospun fibers, obtained after the heat-treatment, are multi-phase and polycrystalline in nature. The crystallites formed within a fiber are within the range of 10 – 15 nm in size. SEM allowed observation of surface morphology of the fiber mats. Fiber diameters, ranging from 165 nm to 830 nm, were obtained. It was found that the use of inorganic binders, like lithium polysilicate, can help in reducing the spread in the fiber diameters. These specimens were subsequently tested for their high temperature reduction behavior. Testing of the reduction behavior is important as it determines the durability and resistance of the sorbent specimens in a cyclic sulfidation/regeneration operation. Temperature controlled reduction was carried out using a Thermo-gravimetric Analyzer (TGA). Fiber-based sorbents with Zn-to-Ti ratio closer to one were found to be more resistant to deactivation caused by reduction. During the reduction reaction, different compositions of the mixed oxide sorbent exhibited different regimes with different rate-controlling steps. More zinc content in the sorbent implied that the overall reaction rate was controlled by the gas-film diffusion step. Product layer diffusion step becomes rate controlling if there is comparable zinc-to-titanium content (as in Sample -2a). In such sample, shrinking-core mechanism can be seen in operation in which inert product layer grows as the active core shrinks. TGA was again used for the sulfidation experiments, carried out at isothermal conditions of 600C with 1% H2S (rest N2) gas stream (with flow rate of 200 ml/min). It was found that the composite oxides (Zn-Ti ratio ≈ 1.1) based nanofibrous sorbent specimens (Sample-2a) were slightly more reactive than the specimens rich in free zinc oxide (e.g. Sample-1 with Zn-Ti ratio ≈ 3.7, both specimens being pre-reduced). This may be an indication of higher density of active sites and surface defects in polycrystalline complex oxides as compared to simple metal oxides. Equivalent grain model was used for identification of the rate-controlling step. For majority of the reaction duration, chemical reaction step was found to be the rate-controlling step rather than the product layer diffusion – an indication of minimal mass transfer resistance offered by the nanofibrous sorbent specimens. During sulfidation, dendritic growth from the parent composite fibers was observed. This was attributed to specific local occurrences of the polar planes of ZnS/ZnO as fiber outer surface along with locally high gas (H2S) / vapor (Zn) concentration. These structures may have contributed to increased specific surface area. Also, no sulfate formation was detected post-sulfidation, as confirmed by the results from the EDX and XPS analysis . Absence of formation of any sulfate compounds during the sulfidation of nanofibrous zinc titanate adsorbent is expected to improve the chances of achieving complete regeneration. In addition, formation of wurtzite phase during sulfidation, a distinct crystal form of zinc sulfide, was seen as another potential advantage of using nanostructured sorbent morphology. Although bulk wurtzite is only metastable at temperatures less than 1020C, wurtzite, in its nanocrystalline form, was apparently stable at lower temperatures (≈ 600C). Wurtzite tend to oxidize directly to zinc oxide whereas oxidation of sphalerite (another ZnS crystal form) has been reportedly linked with the formation of ZnSO4 and Zn3O(SO4)2.Regeneration experiments were carried out at the same temperature as the sulfidation (600⁰C) with a gas stream containing 3% O2 (rest nitrogen) flowing at 200 ml /min. Comparing the reaction data for the two different specimens (Sample-1 vs Sample-2a), sorbent specimens rich in zinc titanates (Sample-2a) showed faster kinetics than the zinc oxide sorbents (both pre-reduced). It was found that it is possible to achieve complete regeneration without raising the sulfidation temperature (600C). Again, results from the XPS analysis confirmed no formation of sulfates on oxidation even when the regeneration temperature was kept same as the sulfidation temperature. Structural properties of the sorbent specimen like the fibrous morphology, grain size of the fresh sorbent mostly remained intact in the regenerated specimens. Multi-cycle sulfidation and regeneration tests found that the nanofibrous design of zinc titanate sorbents can successfully withstand repeated sulfidation-regeneration operations whithout suffering microstructure degradation while maintainng high sulfur capacities. Progressive improvement in the textural properties of the fiber was seen as the reason for better performance. Activation energy associated with the sulfidation reaction was also found to be atleast 3 times smaller than that needed for the other sorbent geometries.All these observations promise an enhanced performance of composite oxides with nanofibrous morphology in a process requiring regenerable sorbents. These findings confirm the main propositions of this work that the sorbent morphology and overall structure influence the sorbent performance as much as the compositional modifications. The above-discussed modifications in the sorbent morphology helped in overcoming some problem areas generally associated with the compositionally-enhanced metal oxides; for instance their lower reactivity, the requirement of high regeneration temperatures, mass transfer controlled overall rate and the consequent incomplete regeneration, all werefound to be absent in the above nanostructured sorbent specimens. Future studies will explore the extent to which these improvements can be sustained in a many-cycle operation.
【 预 览 】
附件列表
Files
Size
Format
View
Electrospun nanofibrous metal oxides as regenerable adsorbents for desulfurization of biomass-derived syngas