Search

The purpose of this investigation was to examine the effect of sulfur impurity on 1%Pt/KL-zeolite catalyst by co-feeding 500 ppbv hydrogen sulfide (H2S) during hexane aromatization under industrially relevant conditions using a plug flow reactor. Product selectivity and hexane conversion were measured with time-on-stream and compared to a clean run carried out under otherwise identical conditions. Sulfur addition to the feed accelerated the rate of deactivation as observed by rapid declines in both hexane conversion and benzene selectivity; hexene selectivity, the product of the less structurally sensitive dehydrogenation reaction, increased significantly. After 20 h, which was enough time to observe sufficient deactivation, the reaction was stopped. For the purpose of catalyst characterization, after cooling to 150 °C, the catalyst was preserved in Polywax 725 to prevent catalyst oxidation. XANES analysis at the potassium K-edge suggests that the local environment for potassium was not significantly altered by sulfur addition, while sulfur K-edge results indicate that sulfur bound to platinum to form platinum sulfide (PtS, not PtS2). Platinum sulfide is likely responsible for accelerating Pt growth, as observed in DRIFTS of adsorbed CO and HR-TEM/STEM micrographs.

XANES K-edge spectra of potassium promoter in precipitated Fe catalysts were acquired following activation by carburization in CO and as a function of time on-stream during the course of a Fischer–Tropsch synthesis run for a 100Fe:2K catalyst by withdrawing catalysts, sealed in wax product, for analysis. CO-activated and end-of-run spectra of the catalyst were also obtained for a 100Fe:5K catalyst. Peaks representing electronic transitions and multiple scattering were observed and resembled reference spectra for potassium carbonate or potassium formate. The shift in the multiple scattering peak to higher energy was consistent with sintering of potassium promoter during the course of the reaction test. The catalyst, however, retained its carbidic state, as demonstrated by XANES and EXAFS spectra at the iron K-edge, suggesting that sintering of potassium did not adversely affect the carburization rate, which is important for preventing iron carbides from oxidizing. The method serves a starting point for developing better understanding of the chemical state and changes in structure occurring with alkali promoter.

The effect of co-fed hydrogen chloride (HCl) in syngas on the performance of iron and cobalt-based Fischer–Tropsch (FT) catalysts was investigated in our earlier studies (Ma et al. in ACS Catal 5:3124–3136, 2015; Davis et al., DOE final report, 2011; Gnanamani et al. in Catal Lett 144:1127–1133, 2014). For an iron catalyst, lower HCl concentrations (

The effect of 1.0 ppm H2S in the syngas feed on initial activity and selectivity of a 0.5%Pt–25%Co/Al2O3 catalyst was studied by comparing the catalyst performance under H2S and sulfur free conditions. The reaction tests were performed using a 1-L slurry phase reactor for 141–212 h under constant reaction conditions: 220 °C, 2.0 MPa, H2/CO = 2.0 and 6.0 Nl/g-cat/h. In the H2S poisoning test, an H2S in N2 gas mixture was added to the syngas feed (1.0 ppm) after running the Fischer–Tropsch synthesis (FTS) reaction for ca. 6.0 h; as such, the impact of H2S on the initial deactivation of the cobalt catalyst (unsteady state) was successfully assessed. The used catalysts were characterized by XANES to investigate if Co–S surface species were formed during the deactivation. The initial deactivation under 1.0 ppm H2S condition was significantly higher (by 2.0–2.4 times) than that under clean conditions. CH4 selectivity increased substantially and C5+ selectivity decreased greatly with time regardless of whether H2S was added or not; however, the addition of H2S accelerated the changes in the hydrocarbon selectivities, which were likely caused by the sharp deactivation of the catalyst in the presence of H2S. After co-feeding the sulfur for 141 h, a comparison was made at similar conversions by adjusting space velocity; the sulfur pretreated cobalt catalysts favored heavier hydrocarbons (C5+) formation and suppressed lower hydrocarbon formation. The addition of H2S to the feed increased CO2 selectivity and the secondary reaction of 1-olefins. The XANES results revealed that surface species involving Co–S bonding formed on the cobalt catalyst after exposure to H2S during FTS. This was likely the primary reason for the abnormal selectivity trends observed during and after the deactivation of the catalyst by sulfur. This study points out a possible approach to increase the selectivity to heavy hydrocarbons by properly sulfiding the cobalt catalyst prior to the FTS reaction.

Silver addition, in low amounts (0.0054 or 0.013 %), to a 1 %Pt/KL catalysts prepared by a sequential CVD procedure using acetylacetonates resulted in improved benzene selectivity at similar conversion levels compared to the undoped 1 %Pt/KL CVD catalyst. Both hydrogenolysis and aromatization require an ensemble of atoms to constitute the active site, but hydrogenolysis is more structure sensitive. Silver, which is a relatively inert metal catalyst, blocks a fraction of Pt surface sites, lowering the overall n-hexane rate. When added in low levels, silver disrupts to a greater degree the Pt ensembles responsible for hydrogenolysis relative to those required for aromatization. This was confirmed by DRIFTS of adsorbed CO measurements, which showed that higher wavenumber bands attributed to larger Pt ensembles were suppressed to a greater extent relative to lower wavenumber bands (i.e., arising from smaller ensembles).

Fischer–Tropsch (FT) synthesis continues to receive widespread attention. Even after 90 years of investigation, the mechanistic route has yet to be fully defined. FT, as a polymerization process, uses CO and H2 as the reactants to produce a broad spectrum of hydrocarbons. Since the conception of the FT synthesis, several different isotopic routes have been employed for mechanistic studies. Various isotopes, such as 13C, 14C, 18O, and 2H, have been utilized through different types of experiments to shed light on the active site(s), the rate-limiting step, and the catalytic pathways. Direct evidence in the FT mechanism has been uncovered by utilizing experiments such as H2/D2 switching trials, as these experiments attempt to shed light on the rate-limiting step of CO hydrogenation. However, before hydrogen participates in the mechanism of CO hydrogenation, it may first dissociatively adsorb on the catalyst surface. The aim of this work is to ascertain to what extent H and D partition on the surface. This is accomplished by passing an equimolar H/D gas mixture over the activated FT catalyst, followed by a TPD method to determine if the active carbide surface displays a partitioning preference toward one of the isotopes. If a preference is observed, then the interpretation of kinetic isotopic effect (KIE) data ascertained in the CO hydrogenation switching experiments could potentially be affected. However, only a very slight isotopic preference toward deuterium was observed, and it is deemed not significant enough to affect an interpretation of the KIE based on H/D switching.

The effect of the reducing agent on the performance of a ruthenium nanoparticle catalyst was investigated during aqueous-phase Fischer–Tropsch synthesis using a 1 L stirred tank reactor in the batch mode of operation. For the purpose of comparison, the activity and selectivity of NaY zeolite supported Ru catalyst were also studied. NaBH4 and hydrogen were used as reducing agents in this study, and hydrogen reduced catalysts exhibited higher activities than the NaBH4 reduced catalysts, because of higher extent of reduction and a relatively lower tendency toward agglomeration of Ru particles. The Ru nanoparticle catalyst displayed higher activities than the NaY zeolite supported Ru catalyst for both reducing agents. NaBH4 reduced catalysts are less active and the carbon dioxide selectivity is higher than the hydrogen reduced catalysts. Nevertheless, the activity of the supported Ru catalyst (Ru/NaY) was 75 % of that of the Ru nanoparticle catalyst, and has the benefit of easy wax/catalyst slurry separation by filtration. The hydrogen reduced supported Ru catalyst exhibited superior selectivity towards hydrocarbons (higher C5+ selectivity and lower selectivity to methane) than all other catalysts tested.

In a recent review article, it was suggested that a redox mechanism best describes observations regarding the WGS reaction for metal promoted ceria systems, as well as how molecules such as CO, CO2, H2O, and H2 interact with the ceria surface. In this contribution, the pretreated surface of Pt/ceria was subjected to steaming or H2 treatment at 250 °C followed by CO adsorption after six different pre-treatment sequences were performed. Subjecting a H2-treated surface directly to CO, or followed by steaming, led to the highest concentration of surface formates by reaction of CO with Ce3+ defect-associated Type II bridging OH groups. The Type II OH groups were also generated by CO treatment followed by steaming. Using CO as a probe led to a significant coverage of formate species, though less than the previous cases (CO added to a H2 treated surface or a surface treated with H2, then H2O). Surprisingly, even subjecting an oxidized surface (pre-reduced and subjected to O2 at 400 °C) to steam still produced a significant density of Type II OH groups, as evidenced by the formates produced upon exposure to CO. The formates were easily decomposed in steam at 130 °C and converted to carbonate species. Thus, focusing on the turnover of formate via C–H bond breaking offers important possibilities for future catalyst development.

The effect of potassium promoter loading (0, 0.5, 1.0 and 2.0 atomic ratio) on the performance of precipitated iron catalysts was investigated during Fischer–Tropsch synthesis using a continuously stirred tank reactor. Characterization by temperature-programmed reduction with CO, Mossbauer effect spectroscopy, and transmission/scanning transmission electron microscopy were used to study the effect of potassium promoter interactions on the carburization, phase transformation and carbon layer formation behavior of the catalysts. Under similar reaction conditions, all four catalysts exhibited similar initial CO conversions (~85 %), whereas stability was found to increase with potassium loading up to 0.5 % (atomic ratio related to the iron), and further increases in potassium led to decreased activity. Unpromoted and excessively K loaded (2.0K/100Fe) catalysts exhibited similar deactivation trends with time and followed essentially similar conversion levels with time-on-stream. The selectivity of various potassium promoted catalysts was found to increase the average molecular weight of hydrocarbon products with increasing potassium loading. The deactivation rate was related to carbon deposition which could embed the iron carbide particles. If not enough K is present, Fe carbides tend to oxidize with TOS; with excessive K-loading, carbon deposition/site blocking become problematic.

The effects of activation gas and copper promoter loading on the Fischer–Tropsch synthesis performance of potassium promoted precipitated iron-based catalysts were investigated using a continuously stirred tank reactor. In this study, CO and syngas (H2/CO = 0.7) activated catalysts were tested after varying the copper promoter loading (0, 2 and 5 %, atomic ratios relative to iron). After attaining a steady-state conversion for the CO-activated catalysts, similar or slightly higher CO conversions were exhibited with increasing copper loading, and the induction period was reduced with increasing copper loading. Partial pressure of hydrogen in the activation gas influenced the resulting activity of the catalysts. For syngas activated catalysts, CO conversion was found to increase with increasing copper loading up to 2 %, and slightly decrease with further increases in copper (5 %) loading. For similar CO conversion levels, the selectivities were similar for the CO activated catalysts, whereas for the syngas activated catalysts, the selectivity varied with copper loading. With increasing copper loading, lower hydrocarbon (methane and C2–C4) selectivities decreased and the corresponding higher hydrocarbon (C5+) selectivity increased.

The effect of halides and potassium addition on cobalt Fischer–Tropsch synthesis (FTS) was studied using 1L CSTR at 493 K, 1.99 MPa and H2/CO of 2.0. The potential catalyst poison was co-fed with syngas after reaching steady-state FTS. Halides were added in an aqueous solution of HX (HF, HCl, HBr) and potassium as aqueous potassium nitrate solution along with syngas at various concentration levels ranging from 500 ppbv to 100 ppmv. Fluoride addition did not affect FT performance of cobalt up to 1.0 ppmv level. However, the presence of other halides (Cl− and Br−) decreased the FT activity proportionally. Addition of potassium decreased cobalt FT activity considerably and increases syngas conversion to methane and CO2 at higher potassium concentrations. The observed poisoning effects for metallic cobalt could be attributed to a blocking effect of cobalt metal sites by halides and potassium in agreement with results of H2 chemisorption on used cobalt FT catalysts.

The effect of reaction temperature on the performance of a traditional Fischer–Tropsch cobalt catalyst (0.5 % Pt–25 % Co/Al2O3) was investigated during aqueous-phase Fischer–Tropsch synthesis (AFTS) using a 1 L stirred tank reactor in the batch mode of operation. The CO conversion rate of the catalyst was found to increase monotonically with increasing reaction temperature. At lower temperatures oxygenate selectivity was high. With increasing the reaction temperature, oxygenate selectivity decreased and the selectivity to hydrocarbons increased. Carbon dioxide and methane selectivity also increased with reaction temperature and the corresponding higher hydrocarbon (C5+) selectivity decreased. For comparison, the CO conversion rate of the catalyst was also tested using C30 oil as a solvent, and similar activation and reaction conditions were utilized in the batch mode of operation. Slightly higher CO rate was observed with C30 oil as a solvent than with the water.

Accumulation of products is a known phenomenon associated with a continuously stirred tank reactor (CSTR). Secondary reactions of α-olefins due to prolonged bed/pore residence time can significantly change the primary Fischer–Tropsch product distribution. Using D2 as a tracer, this study first investigated the significance of Fischer–Tropsch product accumulation in a CSTR. Secondly, the D2 tracer study was used to investigate primary product distribution and olefin to paraffin ratios. Based on the D2 study, it was found that Fischer–Tropsch synthesis with a 2.5 % Ru/NaY catalyst follows a single α mechanism with a chain growth probability of about 0.74. Both olefins and paraffins are primary products and the ruthenium catalyst produced a similar olefin/paraffin ratio for each carbon number. The apparent decline of the O/P ratio for higher carbon number products was shown to be due to secondary reactions of the olefin at prolonged residence times. D2 tracing was shown to be a versatile tool to investigate product accumulation and to define primary product distribution which is very important for mechanistic interpretation and kinetic modeling.

The active sites for Fischer–Tropsch synthesis are surface cobalt metal atoms. In order to obtain a high dispersion for reaction, ultra-small cobalt clusters (e.g., 1 nm) were supported on KL-zeolite using a chemical vapor deposition procedure. A Pt promoter was added to facilitate reduction of the cobalt oxide species. Upon activation in hydrogen, a high dispersion was obtained, but after exposure to realistic FT conditions in a pressurized continuously stirred tank reactor reactor, XANES and EXAFS results indicate that the fine cobalt clusters were susceptible to reoxidation, resulting in adverse activity and selectivity (30–37 % CH4 and 3.5–4.0 % CO2 on a C% basis).

The effect of CO conversion on methane selectivity of a 10 % Co/TiO2 catalyst was investigated using a 1L continuously stirred tank reactor at 493–503 K, 2.4 MPa, H2/CO = 2 and GHSV = 0.4–6 Nl g cat −1 h−1. The cobalt catalysts were activated by H2 and CO, respectively, in two separate runs. For the H2 activated catalyst, the methane selectivity decreased (15–6 %) and C5+ selectivity increased (80–91 %) linearly with increasing CO conversion from 10 to 60 %. The C2–C4 hydrocarbon selectivity decreased slightly with CO conversion. The enhancement in C5+ selectivity was mainly due to the decrease in methane selectivity. The CO activated cobalt catalyst was initially in a carburized form after activation, and then converted to a form with greater metallic character when exposed to syngas at Fischer–Tropsch conditions. The CO activated catalyst at pseudo steady-state had higher methane selectivity in general as compared to the H2 activated one, possibly due to the presence of the cobalt carbide phase. Two kinetic regions appear to exist for the CO activated catalyst. When CO conversion is lower than 20 %, the methane selectivity increased exponentially with CO conversion, which is attributed to enhanced hydrogenation. When CO conversion is higher than 20 %, the CO activated catalyst behaves more like a H2 activated catalyst, with a decrease in methane and an increase in C5+ selectivity with CO conversion.

Heterogeneous nickel catalysts play an important role in numerous processes. In the past, a number of H/D isotopic labeling studies have probed catalytic mechanisms during specific associative and dissociative steps. The current work investigates the adsorption of hydrogen on Ni0 sites, the active surface where H-transfer reactions take place, to determine if preferential partitioning of H or D occurs on the metal surface. This is important for understanding whether differences in surface coverages between H and D could potentially impact KIE analyses. Overall, when H2 and D2 were added competitively to either H-activated NiO or Ni/alumina catalysts, no significant isotopic preference was identified. However, when the catalyst was pretreated with H2 (or D2) prior to flowing the equimolar mixture, a slight isotopic preference for H2 (or D2) was observed. This residual effect suggests that, in both cases, hydrogen exchange with hydroxyl groups on the support occurred. That is, in the case of hydrogen activated NiO, the resulting catalyst is likely a mixture of Ni and residual NiO, the latter presumably containing surface hydroxyl groups. The isotopic preference when H2 (or D2) pretreatment is conducted prior to running the H2/D2 (50–50 %) mixture is enhanced when alumina is used as a support due to greater OH/OD exchange. A similar, but even less pronounced effect, was observed previously with cobalt/alumina catalysts.

The effect of start-up solvent on the performance of cobalt-based catalysts and potassium-promoted precipitated iron catalysts was investigated during Fischer–Tropsch (FT) synthesis by using a continuously stirred tank reactor. In this study, four different molecular weight start-up solvents were tested: Polywax-3000 ( $$\overline{MW}$$ (average molecular weight) = 3,000), Polywax-2000 ( $$\overline{MW}$$ = 2,000), Polywax-500 ( $$\overline{MW}$$ = 500) and C-30 oil ( $$\overline{MW}$$ = 420). The conversion and selectivities (methane, C5+ and CO2) were essentially the same for all the start-up solvents tested for the potassium promoted precipitated iron catalysts suggesting that there was no significant effect of startup solvent for the iron catalyst tested. However, with the cobalt catalyst, conversion varied with solvent, with the conversion increasing with decreasing molecular weight of the solvent. This is considered to be due to the particle size of the cobalt alumina catalysts being large relative to that of iron that was used. Under synthesis conditions, the iron catalyst attrits to form a measurable fraction of 1–3 micron sized particles in the lower range of the particle size distribution, whereas the alumina support retains essentially the same, larger size during synthesis. Thus, the decrease in conversion with time is likely to be a result of pore filling with solvent into the interior of the catalyst, which increases with increasing molecular weight of the start-up solvent. In obtaining conversion and aging data for FT catalysts, wax accumulation needs to be considered.

The hydroisomerization and hydrocracking of long chain n-paraffins and a Fischer–Tropsch wax produced with a cobalt catalyst were accomplished over a Pt–amorphous silica–alumina catalyst. The relative conversion of the n-hexadecane and n-octacosane mixed feed greatly favored the higher carbon number compound even though the conversions of the pure hydrocarbons were the same within a factor of two or less when converted separately. Thus, vapor equilibrium plays a role for the conversion of the heavier alkanes and in this case the conversion essentially occurs with only the compound present in the liquid phase. The single branched cracked products show a peak at the mid-carbon number, C8 and C14 for the two reactants, but the peak for the multi-branched product occurs at a higher carbon number. Thus, it appears that the multi-branched products are primarily produced in a series reaction with the singly branched compounds being formed as the primary products. The data for wax conversion are consistent with the competitive conversion operating for the higher carbon number compounds; however, the transport of intermediate carbon number products from the reactor occurs more rapidly than their formation rates by cracking reactions. The data clearly show that the hydrocracking of wax is dominated by vapor–liquid equilibrium and that hydrocracking is initially controlled by the compounds present in the liquid phase. Figure shows the catalyst pore filling with low boiling (left) and high boiling (right) hydrocarbons. Each reactant saturates the catalytic sites and the breaking of C–C bond occurs. Once the products from cracking of the liquid phase go into the vapor phase, it should rapidly pass the catalyst bed. This short contact time on gas phase hydrocarbons relative to the liquid phase, limits the conversion of low boiling point hydrocarbons.

The effect of CO conversion (35–91 %) on hydrocarbon product distribution (CH4, C5 + and olefin selectivities) and CO2 selectivity was studied in a fixed bed reactor on 15 % Co/Al2O3 catalyst at conditions of industrial relevance. Selectivity to CH4 decreased, whereas those of C5 + and CO2 increased with increasing CO conversion (and consequently, partial pressure of water produced during the Fischer–Tropsch synthesis). 1-Olefin content of C2–C5 olefins decreased and 2-olefin content of C4 hydrocarbon increased with increasing CO conversion.

Pt-promoted tungsten-hafnia catalysts containing different amounts of tungsten (15 and 20 wt% WO3) were prepared and tested for hydroisomerization of n-hexadecane in a trickle-bed reactor. All catalysts were calcined under dry air flow at 723 K for 3 h. Pt-promoted sulfated hafnia was also prepared for the purpose of comparison. The catalysts were characterized using BET surface area, X-ray diffraction, and DRIFTS for the adsorption of pyridine. Tungsten-based hafnia catalysts exhibit moderate activity ( 95 %), which had an operating temperature of 493 K. However, tungsten-based hafnia catalysts were more selective towards iso-C16 even at higher n-hexadecane conversions. More cracking products were obtained with sulfated hafnia in which the selectivity to iso-C16 decreases exponentially with n-hexadecane conversion. The less acidic tungsten–hafnia catalysts may be suitable for producing more isomerized products compared to the sulfated analog.

In the preparation and evaluation of Fischer–Tropsch (FT) catalysts, active catalysts formed by both incipient wetness impregnation (IWI) and atomic layer deposition (ALD) of major components were demonstrated. ALD-deposited Co on a silica support was more effective than a similar catalyst deposited upon a support of ALD-deposited Al2O3 on silica. The addition of Co reduction promoters including Pt, Ir and Ru using either ALD or IWI has been shown to strongly affect the catalyst pre-conditioning step. CO conversion results were consistent with previously reported Temperature Programmed Reduction X-ray Absorption Near-edge Structure/Extended X-ray Absorption Fine Structure Spectroscopy (TPR-XANES/EXAFS) experiments observing the nature of chemical transformations occurring during the activation of cobalt-based FT catalysts in hydrogen. Specifically, there exists a 2-step reduction process involving Co3O4 to CoO and CoO to Co0 transformations. The extent of catalyst preconditioning was strongly affected by the reduction temperature (with 400 °C preferred) and the loading of the promoter. This was demonstrated using a continuous-flow catalytic-bed unit with a 2:1 molar blend of H2:CO, at temperatures ranging from about 260 to 300 °C, pressures averaging 1.3 MPa (190 psia), and gas space velocities about 24 NL/h-g.

A series of platinum loaded catalysts supported on cation (Me)-doped cerium dioxide (Me = Ba, La, Y, Hf and Zn) was prepared by co-precipitation of the Me-nitrates and impregnation of a Pt precursor. Low temperature water–gas shift activity depends on the nature of dopant employed, varying in the order of Ba > Y > Hf > La > undoped ceria > Zn. TPR-XANES measurements with flowing hydrogen reveal that adding dopants to ceria facilitate ceria reduction and increases the extents of both surface shell and bulk reduction of ceria. Experimental results confirm past theoretical models that dopants enhance both O-mobility and reducibility of ceria. DRIFTS measurements of the transient decomposition of formates in steam suggest that formate half-life follows the trend Zn > undoped ceria > La > Hf > Y > Ba, indicating that the formate decomposition rate is enhanced by the addition of most of the dopants tested. Taken together, the results suggest that dopant addition improves the WGS rate by increasing the O-mobility of O-bound associated intermediates. Therefore, less Pt and Ce, which are expensive, is required to achieve comparable levels of activity.

The hydrogenation of CO2 using Pt promoted Co/γ-Al2O3 and doubly (Cu, K) promoted iron catalysts exhibits an inverse isotope effect (r H/r D

The price of iridium currently trends at about half the cost of platinum, the latter being a typical reduction promoter for Co/Al2O3 Fischer–Tropsch (FT) synthesis catalysts in gas-to-liquids (GTL) technology. In the current contribution, both fixed-bed catalytic FT and TPR-EXAFS/XANES experiments were carried out over 0.1% iridium-doped 25% Co/Al2O3 catalysts in order to (1) assess the effectiveness of Ir as a promoter of cobalt oxide reduction and (2) evaluate the effectiveness of the incipient wetness impregnation (IWI) technique for adding the Ir precursor by comparing a catalyst prepared by IWI to one prepared by atomic layer deposition (ALD). Ir was demonstrated to be an effective promoter for facilitating the second step of cobalt oxide reduction, CoO to Co0, and the IWI method was found to be superior to ALD.

The effect of water on the performance of potassium-promoted precipitated iron catalyst was investigated during Fischer–Tropsch synthesis (FTS) using a continuously stirred tank reactor (CSTR) at two different reaction temperatures. Water was added in such a manner as to replace an equivalent amount of inert gas so that all other reaction conditions (e.g., reactant partial pressure, space velocity) remained the same before, during, and after water addition. The externally added water had a positive effect on CO conversion at 270 °C whereas, for the reaction carried out at 230 °C, the added water decreased CO conversion and deactivated the catalyst. From these findings, the addition of water at 230 °C oxidized the catalyst, transforming the iron carbide to the Fe3O4 phase. When the reaction was carried out at 270 °C, severe oxidization did not take place and a carbide phase was retained. The loss in activity and the rate of deactivation were more pronounced at 230 °C compared to the 270 °C condition for the same catalyst. Mossbauer spectroscopic measurements revealed that for the reaction carried out at 230 °C, the catalyst had 85% of the iron present as Fe3O4 and the remaining as Hagg carbide (χ-Fe5C2), whereas at the higher temperature reaction condition the catalyst had about 66% of the iron present as έ-Fe2.2C, with the remaining as Fe3O4. These findings were also supported by XANES analysis, where a high white line intensity was observed for the sealed used catalyst sample for the low temperature reaction condition, indicating a higher extent of oxidation. A low white line intensity was recorded for the used sample for the high temperature reaction condition, indicative of a higher extent of reduction. The effect of water on the performance of potassium-promoted precipitated iron catalyst was investigated during Fischer–Tropsch synthesis (FTS) by using a continuously stirred tank reactor (CSTR) at two different reaction temperatures. In these studies, the added water replaced an equivalent amount of inert gas so that all other reaction conditions remained the same before, during, and after water addition. The externally added water had a positive effect on CO conversion at 270 °C whereas, for the reaction carried out at 230 °C, the added water decreased CO conversion and deactivated the catalyst. From these findings, the addition of water at 230 °C oxidized the catalyst, completely transforming the iron carbide to the Fe3O4 phase. When the reaction was carried out at 270 °C, oxidization did not take place and the carbide phase was retained. The loss in activity and the rate of deactivation were more pronounced at 230 °C compared to the 270 °C condition for the same catalyst. Mossbauer spectroscopic measurements and X-ray absorption near edge spectroscopy (XANES) studies further supported these findings.

KIT-6 molecular sieve was used as a support to prepare cobalt catalyst for Fischer–Tropsch synthesis (FTS) using an incipient wetness impregnation method to produce cobalt loadings of 15 and 25 wt%. The catalysts were characterized by BET surface area, X-ray diffraction, scanning transmission election microscopy (STEM), extended X-ray absorption fine spectroscopy and X-ray absorption near edge spectroscopy. The catalytic properties for FTS were evaluated using a 1L CSTR reactor. XRD, pore size distribution, and STEM analysis indicate that the KIT-6 mesostructure remains stable during and after cobalt impregnation and tends to form smaller cobalt particles, probably located inside the mesopores. The mesoporous KIT-6 exhibited a slightly higher cobalt dispersion compared to amorphous SiO2 supported catalyst. With the higher Co loading (25 wt%) on KIT-6, partial structural collapse was observed after the FTS reaction. Compared to an amorphous SiO2 supported cobalt catalyst, KIT-6 supported cobalt catalyst displayed higher methane selectivity at a similar Co loading, likely due to diffusion effects. KIT-6 molecular sieve was used as a support to prepare cobalt catalyst for Fischer–Tropsch synthesis (FTS) using an incipient wetness impregnation method to produce cobalt loadings of 15 and 25 wt%. The catalysts were characterized by BET surface area, X-ray diffraction (XRD), scanning transmission election microscopy (STEM), extended X-ray absorption fine spectroscopy (EXAFS) and X-ray absorption near edge spectroscopy (XANES). The catalytic properties for FTS were evaluated using a 1L CSTR reactor. The mesoporous KIT-6 exhibited a slightly higher cobalt dispersion compared to amorphous SiO2 supported catalyst. With the higher Co loading (25 wt%) on KIT-6, partial structural collapse was observed after the FTS reaction. XRD, pore size distribution, and STEM analysis indicate that the KIT-6 mesostructure remains stable during and after cobalt impregnation and tends to form smaller cobalt particles, probably located inside the mesopores. Compared to an amorphous SiO2 supported cobalt catalyst, KIT-6 supported cobalt catalyst displayed higher methane selectivity at a similar Co loading, likely due to diffusion effects.

Nano-ridge and valley structures formed as a result of self-aligned individual ceria nano-crystals, and HRSTEM/EELS revealed the 3-dimensional morphology profiles of the ridge and valley structures of the ceria support. After preparation of a Pt/ceria catalyst, HRSTEM indicated the preferential alignment of Pt nanoparticles inside the nano-valleys of the ceria support. EELS analyses confirmed changes in the oxidation state of ceria in the vicinity of Pt nanoparticles that are present at different depth in the valleys.

A series of mixed Hf X Zr1−X O2 oxide catalysts was prepared according to a recipe that yields the monoclinic structure. The samples were examined by EXAFS spectroscopy at the Zr K and Hf LIII edges. A fitting model was used that simultaneously fits data from both edges, and makes use of an interdependent mixing parameter X mix to take into account substitution of the complementary atom in the nearest metal-metal shell. For XPS analysis, Scofield factors were applied to estimate the relative atomic surface concentrations of Zr and Hf. EXAFS results suggested that a solid bulk solution was formed over a wide range of X for Hf X Zr1−X O2 binary oxides, and that the relative ratio was retained in the surface shell (i.e., including some subsurface layers by XPS) and the surface (e.g., by ISS). The increase in selectivity for the 1-alkene from dehydration of alcohols at high Zr content does not correlate smoothly with the tuned relative atomic concentration of Hf to Zr. The step change at high Zr content appears to be due to other indirect factors (e.g., surface defects, oxygen vacancies).

Doping Pt/ceria catalysts with the Group 1 alkali metals was found to lead to an important weakening of the C–H bond of formate and methoxy species. This was demonstrated by a shift to lower wavenumbers of the formate and methoxy ν(CH) vibrational modes by DRIFTS spectroscopy. Li and Na-doped Pt/ceria catalysts were tested relative to the undoped catalyst for low temperature water–gas shift and methanol steam reforming using a fixed bed reactor and exhibited higher catalytic activity. Steaming of formate and methoxy species pre-adsorbed on the catalyst surface during in-situ DRIFTS spectroscopy suggested that the species were more reactive for dehydrogenation steps in the catalytic cycle for the Li and Na-doped catalysts relative to undoped Pt/ceria. However, with increasing atomic number over the series of alkali-doped catalysts, the stability of a fraction of the carbonate species was found to increase. This was observed during TPD-MS measurements of the adsorbed CO2 probe molecule by a systematic increase of a high temperature peak for a fraction of the CO2 desorbed. This result indicates that alkali-doping is an optimization problem—that is, while improving the dehydrogenation rates of methoxy and formate species, the carbonate intermediate stability increases, making it difficult to liberate the CO2. Infrared spectroscopy results of CO adsorbed on Pt and ceria suggest that the alkali dopant is located on, and electronically modifies, both the Pt and ceria components. The results not only lend further support to the role that methoxy and formate species play as intermediates in the catalytic mechanisms, but also provide a path forward for improving rates by means other than resorting to higher noble metal loadings.

Rubidium promoted iron Fischer–Tropsch synthesis (FTS) catalysts were prepared with two Rb/Fe atomic ratios (1.44/100 and 5/100) using rubidium nitrate and rubidium carbonate as rubidium precursors. Results of catalytic activity and deactivation studies in a CSTR revealed that rubidium promoted catalysts result in a steady conversion with a lower deactivation rate than that of the corresponding unpromoted catalyst although the initial activity of the promoted catalyst was almost half that of the unpromoted catalyst. Rubidium promotion results in lower methane production, and higher CO2, alkene and 1-alkene fraction in FTS products. Mossbauer spectroscopic measurements of CO activated and working catalyst samples indicated that the composition of the iron carbide phase formed after carbidization was χ-Fe5 C2 for both promoted and unpromoted catalysts. However, in the case of the rubidium promoted catalyst, ɛ′-Fe2.2C became the predominant carbidic phase as FTS continued and the overall catalyst composition remained carbidic in nature. In contrast, the carbide content of the unpromoted catalyst was found to decline very quickly as a function of synthesis time. Results of XANES and EXAFS measurements suggested that rubidium was present in the oxidized state and that the compound most prevalent in the active catalyst samples closely resembled that of rubidium carbonate.

Doping Pt/ceria catalysts with alkali metals was found to lead to an important weakening of the formate C–H bond, as demonstrated by a shift to lower wavenumbers of the ν(CH) vibrational mode. However, with high alkalinity (∼2.5%Na or equimolar amounts of K, Rb, or Cs), a tradeoff was observed such that while the formate became more reactive, the stability of the carbonate species, which arises from the formate decomposition, was found to increase. This was observed by TPD-MS measurements of the adsorbed CO2 probe molecule. Increasing the amount of alkali to levels that were too high also led to lower catalyst BET surface area, the blocking of the Pt surface sites as observed in infrared measurements, and also a shift to higher temperature of the surface shell reduction step of ceria during TPR. When the alkalinity was too high, the CO conversion rate during water–gas shift decreased in comparison with the undoped Pt/ceria catalyst. However, at lower levels of alkali, the abovementioned inhibiting factors on the water–gas shift rate were alleviated such that the weakening of the formate C–H bond could be utilized to improve the overall turnover efficiency during the water–gas shift cycle. This was demonstrated at 0.5%Na (or equimolar equivalent levels of K) doping levels. Not only was the formate turnover rate found to increase significantly during both transient and steady state DRIFTS tests, but this effect was accompanied by a notable increase in the CO conversion rate during low temperature water–gas shift.

NO oxidation was studied over Pt/CeO2 and Pt/SiO2 catalysts. Apparent activation energies (Ea) of 31.4 and 40.6 kJ/mole were determined for Pt/CeO2 and Pt/SiO2, respectively, while reaction orders for NO and O2 were fractional and positive for both catalysts. Pre-treatment of the catalysts with SO2 caused a decrease in the Ea values, while the reaction orders were only slightly changed. In situ DRIFTS measurements indicated that high concentrations of nitrate species were formed on the surface of Pt/CeO2 during NO oxidation, while almost no surface species could be detected on Pt/SiO2. The addition of SO2 resulted in the formation of a highly stable sulfate at the expense of nitrate species and caused an irreversible loss of catalytic activity for Pt/CeO2.

The loss in activity of Pt-promoted and unpromoted 25 wt% Co–Al2O3 catalysts has been compared under identical conditions except for adjustment of the space velocity to give the same initial CO-conversion. Both catalysts underwent a 200 h period of rapid, initial decline in CO conversion and then a slower, linear decline during the next 1000 h. Pt-promotion did not alter the cobalt dispersion (or average particle size) from that of the unpromoted catalyst but did increase the amount of cobalt that was reduced. When compared not by time-on-stream, but by the moles of Co converted per unit weight of catalyst, both the Pt-promoted and unpromoted catalysts decline in activity at the same rate.

The stability of surface formates generated by reaction of bridging OH groups with CO is an important first criterion supporting the idea that the rate limiting step of WGS involves formate decomposition. The second important factor is that, in the presence of water, shown directly by the measurements obtained during this steady state isotope switching study, the forward decomposition of surface formates to CO2 and H2 is strongly auto-catalyzed by H2O, in agreement with the findings of Shido and Iwasawa. Based on a normal kinetic isotope effect previously observed with H2O:D2O switching and the response of surface formate coverages to the WGS rate under steady state conditions when a high H2O:CO ratio is employed, the conclusion is drawn that a surface formate mechanism is likely operating for the low temperature water gas shift reaction.

Recently, the role of ceria vacancies on water–gas shift activity has been explained in terms of a redox mechanism, whereby CO adsorbed on a metal reduces the ceria surface to generate CO2, and water reoxidizes the ceria surface to CeO2, liberating hydrogen in the process. In this study, we examine the possibility of a ceria-mediated redox mechanism by examining more closely the evolution of carbonate and formate bands under different controlled treatment environments, and utilizing different reduction procedures. Earlier it was claimed that the decomposition of carbonates by water was consistent with a redox process, whereby the CO2 product could spillover to the support. We found that the observation of carbonate formation and decomposition by water was a result of the treatment procedure used in the earlier work, and that, once bridging OH groups are produced in the presence of water, the reaction more likely proceeds via a formate intermediate, which is produced by reaction of CO with the active bridging OH groups. However, the vacancies appear to play an important role in generating these active sites. Possible pathways to active site generation are discussed.