Applied Catalysis A, General (v.335, #2)

Contents (iii-vi).

Modified Wacker TBHP oxidation of 1-dodecene by J.M. Escola; J.A. Botas; J. Aguado; D.P. Serrano; C. Vargas; M. Bravo (137-144).
A higher 1-olefin (1-dodecene) was oxidized to 2-dodecanone and other ketones in a modified Wacker system featured by omitting the use of copper chloride and employing tert-butyl hydroperoxide (TBHP) as oxidant. Acetonitrile was the chosen solvent due to its higher selectivity towards 2-dodecanone. At 40 °C, ∼90% conversion with 61% selectivity towards 2-dodecanone was attained after 3 h of reaction.▪A higher 1-olefin (1-dodecene) was oxidized to 2-dodecanone and other ketones in a modified Wacker system featured by omitting the use of copper chloride and employing tert-butylhydroperoxide (TBHP) as oxidant. Acetonitrile was the solvent of choice due to its higher selectivity towards 2-dodecanone, the reaction occurring mostly in the interface. A maximum in conversion and selectivity was measured for a ratio [CH3CN]/[1-dodecene] = 10. The temperature abatement enhances the 2-dodecanone selectivity (61% at 40 °C) because of the lower extent of the competing isomerization reactions. An optimum in 2-dodecanone selectivity was found for a [TBHP]/[1-dodecene] ratio = 7. The use of H2O2 as oxidant led towards slower oxidation rates than TBHP. Other palladium salts (PdSO4 and Pd(CH3COO)2) and several R–CN solvents (propionitrile, benzonitrile, isobutyronitrile) yielded lower performances than palladium chloride and acetonitrile, respectively. On the contrary, β-cyclodextrin, a phase transfer catalyst, improved the selectivity towards 2-dodecanone.
Keywords: Palladium chloride; Wacker oxidation; tert-Butylhydroperoxide; 2-Dodecanone; Acetonitrile;

The effect of the use of cerium (1 wt%)-doped alumina on the performance of Pt/Al2O3, Pt/CeO2/Al2O3 and Pt/CeZrO2/Al2O3 catalysts on the partial oxidation of methane (POM) was evaluated. The catalysts were characterized by X-ray diffraction (XRD), BET surface area and temperature-programmed reduction (TPR). Platinum dispersion was estimated through cyclohexane dehydrogenation. Pt/CeZrO2/Al-doped catalyst showed the best performance on the POM. XRD analysis revealed that the previous addition of ceria to alumina led to the formation of a homogeneous solid solution, which exhibited a high-oxygen storage capacity. The good activity and stability of this material was associated to its higher metal dispersion, higher oxygen storage capacity and larger degree of coverage of alumina by CeZrO2 oxide, which increase the effectiveness of the carbon cleaning mechanism.The effect of the use of cerium (1 wt%)-doped alumina on the performance of Pt/CeO2/Al2O3 and Pt/CeZrO2/Al2O3 catalysts on the partial oxidation of methane was evaluated. The doped alumina-supported catalysts exhibited the best performance on partial oxidation of methane due to a proper balance between a good metal dispersion, high-oxygen storage capacity and a good degree of coverage of alumina. (○) Pt/Al-1173, (●) Pt/Al-doped, (▵) Pt/CeO2/Al-1173, (▴) Pt/CeO2/Al-doped, (□) Pt/CeZrO2/Al-1173 and (■) Pt/CeZrO2/Al-doped. ▪
Keywords: Partial oxidation of methane; Cerium-doped alumina; Pt/CeO2/Al2O3 and Pt/CeZrO2/Al2O3 catalysts; Syngas production; Gas-to-liquid technology;

A thermodynamic investigation of the redox properties of ceria–titania mixed oxides by Gong Zhou; Jonathan Hanson; Raymond J. Gorte (153-158).
Ce0.9Ti0.1O2 and Ce0.8Ti0.2O2 were prepared by the citric-acid method and characterized using XRD, coulometric titration, and WGS reaction rates. The mixed oxides are shown to be significantly more reducible than bulk ceria, with enthalpies for re-oxidation being approximately −500 kJ/mol O2. However, WGS rates over 1 wt% Pd supported on ceria, Ce0.8Ti0.2O2, and Ce0.81Zr0.192O2 were nearly the same.Pictogram: ▪Oxygen-to-metal ratios for (○) CeO2, (□) Ce0.9Ti0.1O2, and (△)Ce0.8Ti0.2O2 as a function of P(O2) at 973 K, measured by coulometric titration, for samples were calcined at 973 K for 5 h. The closed squares (■) show oxygen-to-metal ratios for Ce0.9Ti0.1O2 at 973 K after the sample had been calcined at 1323 K. The closed triangles (▴) show the oxygen to metal ratio of Ce0.9Ti0.1O2 at 973 K measured using flow titration.Ceria–titania solutions with compositions of Ce0.9Ti0.1O2 and Ce0.8Ti0.2O2 were prepared by the citric-acid (Pechini) method and characterized using X-ray diffraction (XRD) for structure, coulometric titration for redox thermodynamics, and water–gas-shift (WGS) reaction rates. Following calcination at 973 K, XRD suggests that the mixed oxides exist as single phase, fluorite structures, although there was no significant change in the lattice parameter compared to pure ceria. The mixed oxides are shown to be significantly more reducible than bulk ceria, with enthalpies for re-oxidation being approximately −500 kJ/mol O2, compared to −760 kJ/mol O2 for bulk ceria. However, WGS rates over 1 wt% Pd supported on ceria, Ce0.8Ti0.2O2, and Ce0.8Zr0.2O2 were nearly the same. For calcination at 1323 K, the mixed oxides separated into ceria and titania phases, as indicated by both the XRD and thermodynamic results.
Keywords: Ceria–titania; Ceria; Coulometric titration; Thermodynamic properties; Oxidation enthalpy; Water–gas shift;

Novel bifunctional NiMo/Al-SBA-15 catalysts for deep hydrodesulfurization: Effect of support Si/Al ratio by Tatiana Klimova; Javier Reyes; Oliver Gutiérrez; Lilia Lizama (159-171).
NiMo catalysts supported on Al-SBA-15 with different Si/Al molar ratios were prepared and tested in the 4,6-dimethyldibenzothiophene HDS. These catalysts were able to isomerize methyl groups of 4,6-DMDBT prior to its desulfurization.▪A series of Al-containing mesoporous molecular sieves SBA-15 with different Si/Al molar ratios (50, 30, 20 and 10) was prepared by chemical grafting method using aluminum(III) chloride as alumina source. Mo and NiMo catalysts prepared using Al-SBA-15 supports were tested in the 4,6-dimethyldibenzothiophene (4,6-DMDBT) hydrodesulfurization (HDS) reaction. The synthesized solids were characterized by N2 physisorption, small- and wide-angle XRD, Py FT-IR, 27Al MAS-NMR, temperature-programmed reduction (TPR), UV–vis diffuse reflectance spectroscopy (DRS), chemical analysis and HRTEM. It was found that post-synthetic alumination of SBA-15 by reacting with AlCl3 has a little influence on the ordered mesoporous structure of the parent SBA-15 and leads to amorphous silicoaluminate materials with moderate Brönsted acidity. For all Si/Al molar ratios studied, tetrahedrally coordinated Al3+ cations predominated over octahedrally coordinated ones. However, as Al loading increases, the proportion of the latter ones was also increased. The dispersion of oxidic and sulfided Mo species increases with the incorporation of aluminum in the SBA-15 support due to the strong interaction of Mo and Ni species with aluminum atoms which serve as anchoring sites on the support surface. Catalytic activity of Mo and NiMo/SBA-15 catalysts increased with Al incorporation in the support reaching a maximum at Si/Al molar ratio of 20. High activity of NiMo/Al-SBA-15 catalysts in 4,6-DMDBT HDS can be attributed to a good dispersion of Ni and Mo active phases and to the bifunctional character of these catalysts. It was shown that both types of sites, coordinatively unsaturated sites of NiMoS active phase and Brönsted acid sites of the support participate in the catalytic transformations of 4,6-DMDBT.
Keywords: NiMo catalysts; Al-SBA-15; Deep hydrodesulfurization; 4,6-Dimethyldibenzothiophene; Bifunctional catalysts;

Mesostructured SBA-15 materials, with both pure silica and aluminosilicate compositions, have been investigated as supports for the heterogenization of the chiral Mn(salen) catalyst using different methods: impregnation, direct ion exchange, indirect ion exchange, and coordinative anchoring on organosilica-functionalized supports. In the last case, two different types of SBA-15 functionalization agents were tested: aminopropyltrimethoxysilane and aminophenyltrimethoxysilane. In all the methods studied, the incorporation of the Mn(salen) catalyst does not modify the SBA-15 structure although a significant decrease in the surface area, pore volume and pore size is observed. Enantioselective styrene epoxidation using m-CPBA as oxidant and NMO as co-oxidant has been used as test reaction for the different heterogeneous catalysts synthesized. The stability of the heterogenized systems has been studied by carrying out reusing tests. The main problem encountered is the extensive leaching of the Mn(salen) complex that occurs with most of the catalytic systems. However, a significant stability has been found for the catalyst prepared by anchoring the Mn(salen) complex on a SBA-15 support previously functionalized with aminophenyltrimethoxysilane. In this case, the extension of the Mn(salen) leaching has been reduced below 10% by increasing both the styrene/oxidant and catalyst/styrene ratios.Mesostructured SBA-15 materials, with both pure silica and aluminosilicate compositions, have been investigated as supports for the heterogenization of the chiral Mn(salen) catalyst using different methods: impregnation, direct ion exchange, indirect ion exchange, and coordinative anchoring on organosilica-functionalized supports. In the last case, two different types of SBA-15 functionalization agents were tested: aminopropyltrimethoxysilane and p-aminophenyltrimethoxysilane. In all the methods studied, the incorporation of the Mn(salen) catalyst does not modify the SBA-15 structure although a significant decrease in the surface area, pore volume and pore size is observed. Enantioselective styrene epoxidation using m-CPBA as oxidant and NMO as co-oxidant has been used as test reaction for the different heterogeneous catalysts synthesized. The stability of the heterogenized systems has been studied by using the same catalyst in repetitive reaction runs. The main problem is the extensive leaching of the Mn(salen) complex. After three reaction cycles, the best results were obtained with the catalyst prepared by anchoring the Mn(salen) complex on a SBA-15 support previously functionalized with aminophenyltrimethoxysilane. Moreover, the extension of the Mn(salen) leaching can be reduced in this catalyst system below 10% by increasing both styrene/oxidant and catalyst/styrene ratios. ▪
Keywords: Mn(salen); Chiral catalyst; Enantioselectivity; Epoxidation; SBA-15;

The activation of Phillips Cr/silica catalysts by M.P. McDaniel; K.S. Collins; E.A. Benham; T.H. Cymbaluk (180-186).
This report attempts to further define the underlying principles governing activation of the Phillips Cr/silica catalyst. Calcining the catalyst causes dehydroxylation or annealing of the silica surface that generates the active site population, and thereby the catalyst activity and polymer character. The extent of dehydroxylation during activation depends not only on the calcining temperature, but also on the calcining time, making the two variables seemingly inter-convertible in their effect on catalyst and polymer behavior. However, the rate of surface annealing also depends on temperature, so that the degree of time dependence can vary with the choice of activation temperature. These facts can be used to explain commercial behavior and to design more efficient activation recipes.The temperature used to activate Cr/silica catalyst determines its activity and polymer character, such as MW distribution. In this paper, activation time is shown to be equally important, affecting the catalyst much as temperature does. Commercial dehydroxylation of the silica is a time-dependent process in which both time and temperature can be considered as inter-convertible.▪
Keywords: Phillips chromium catalyst; Ethylene polymerization; Polyethylene; Catalyst activation;

Behavior of H2 chemisorption on Ru/TiO2 surface and its application in evaluation of Ru particle sizes compared with TEM and XRD analyses by Xiongfei Shen; Luis-Javier Garces; Yunshuang Ding; Kate Laubernds; Richard P. Zerger; Mark Aindow; Edward J. Neth; Steven L. Suib (187-195).
Ru is an active hydrogenation catalyst in many fields such as Fischer-Tropsch synthesis, fuel cell, bio-ethanol production etc. Most of these applications involve H2 gas absorption on Ru surface to carry further reactions. However, the behavior of H2 adsorption on Ru surface has seldom been investigated systematically. This manuscript presents a systematic study of H2 adsorption on Ru surface (Ru/TiO2) under different adsorption temperatures, equilibrium times, and H2 pressures. H2 chemisorption data were also used to calculate Ru particle sizes and are in agreement with TEM methods.▪Hydrogen adsorption on Ru surfaces of Ru/TiO2 catalysts has been investigated in a gas adsorption apparatus at temperatures from 35 to 150 °C, H2 partial pressures from 0.05 to 700 Torr, and adsorption equilibration time at each pressure from 3 to 60 min. Results showed that adsorption of H2 on the TiO2 support are negligible. TEM measurements suggested that there was no strong metal–support interaction (SMSI) on the synthesized Ru/TiO2 samples to inhibit H2 adsorption. A transition pressure point of ∼60 Torr was observed for H2 adsorption on Ru/TiO2. Most of the strong chemisorption occurs before this transition pressure and weak chemisorption happens thereafter. H2 adsorption increases with temperatures from 35 to 75 °C indicating an activation energy present for H2 chemisorption on Ru. Above 75 °C, the adsorption slightly decreases with further increase of temperature up to 150 °C. Monolayer coverage was attained at 75 °C for 30 min equilibration time with a H2 pressure higher than 300 Torr. Monolayer H2 chemisorption was used to determine Ru metal particle size in Ru/TiO2 systems compared with XRD and TEM analyses. Selected area electron diffraction (SAD) indicated that there was no preferred crystallographic orientation of the TiO2 supported Ru. Therefore, exposed Ru atoms equally contribute from the three low-index planes [(0 0 1), (1 0 0), and (1 1 0)] with the highest atomic density instead of only the (1 0 0) plane for conventional treatments. With this assumption, the average Ru particle size calculated from H2 chemisorption (4.6 nm) agrees with the TEM measurements (4.1 nm).
Keywords: Ru/TiO2; Particle size; H2 chemisorption; Monolayer coverage; Fischer-Tropsch catalyst;

Fresh, industrially spent, deactivated and regenerated VTiSbSiO x ammoxidation catalysts were used for nicotinonitrile synthesis from 3-picoline in various catalytic tests. The materials were characterized using different techniques such as XRD, UV/VIS-, and EPR-spectroscopy. The characterizations showed structural changes and active site reduction due to long-lasting operation that can be turned back by re-oxidation at elevated temperatures.▪Fresh and industrially spent VTiSbSiO x ammoxidation catalysts were used for nicotinonitrile manufacture from 3-picoline in various catalytic tests. The solid-state characterization of these materials was carried out using different techniques. Catalytic runs with spent samples showed a significant drop in catalytic activity in comparison to the fresh catalyst. However, neither coke-containing deposits nor loss of catalyst components could be detected in the spent samples. X-ray diffraction, nitrogen adsorption, UV/VIS- and EPR-spectroscopy were applied to uncover structural alterations during industrial long-term use and to explore possible reasons for the observed deactivation behavior. Characterization by UV/VIS revealed a partial reduction of vanadium (V) and (IV) towards Vanadium(III). EPR showed structural changes leading to a more pronounced formation of antiferromagnetically interacting vanadium (IV) oxide clusters and, thus, to a loss in vanadium dispersion. A slight increase in crystallinity of the spent samples could be observed by XRD. All these effects collectively lead to the observed deactivation; however, the original activity can be restored by re-oxidation under airflow at 600 °C. By means of regeneration, reduced vanadium species (mainly V(III)) can be partly re-oxidized and the dispersion of the vanadium (IV) oxide clusters can be enhanced again as evidenced by EPR. Interestingly, the catalytic properties of regenerated samples are comparable to the fresh solid.
Keywords: Ammoxidation; Nicotinonitrile; Catalyst deactivation; Oxidative regeneration; Catalyst characterization;

The effect of phosphorus on hydrotreating property of NiMo/γ-Al2O3 nitride catalyst by V. Sundaramurthy; A.K. Dalai; J. Adjaye (204-210).
γ-Al2O3 supported P-doped nickel molybdenum (NiMo) nitride hydrotreating catalysts with 12 wt.% of Mo, 3 wt.% Ni and 0–2.5 wt.% P were synthesized by means of temperature-programmed reaction, and characterized by BET surface area analysis, elemental analysis, CO uptake titration, DRIFT study of adsorbed CO, temperature-programmed oxidation (TPO), temperature-programmed reduction (TPR) and temperature-programmed desorption (TPD) of NH3. The activity of the nitride catalysts was studied in a trickle-bed reactor operated at 8.8 MPa and 385 °C in the hydrodenitrogenation (HDN) and hydrodesulfurization (HDS) of heavy gas oil (HGO) derived from Athabasca bitumen. The CO uptakes measurement showed that P addition to NiMo nitride improves the dispersion. DRIFT spectra of CO adsorption evidenced that the number of Mo2+ sites on the surface of NiMo nitride increases with increasing P concentration and reaches optimum with 2.0 wt.% of P. TPD of NH3 indicated an increase of strong acid sites and also strengthening of moderate acid sites in NiMo nitride on P doping. All the nitride catalysts have suffered partial sulfidation during initial period of hydrotreating of HGO. HDN activity of NiMo nitride was promoted greatly by the phosphorus content, reached a maximum with P content of 1.6 wt.%. An increased acidity of P-doped NiMo nitrides accelerated C–N bond breaking and thus enhanced the HDN activity. No significant change was observed in HDS of HGO over NiMo/Al2O3 nitride catalyst on P doping.The synthesized P-doped γ-Al2O3 supported nickel molybdenum (NiMo) nitride catalysts were characterized, and tested for HDN and HDS of heavy gas oil (HGO) derived from Athabasca bitumen at industrial conditions.▪
Keywords: γ-Al2O3 support; NiMo nitride; Phosphorus doping; DRIFT of CO adsorption; Hydrodesulfurization; Hydrodenitrogenation; Heavy gas oil;

Ortho-dichlorobenzene oxidation over Pd/Co loaded sulfated zirconia and mordenite catalysts by Beatriz H. Aristizábal; Cristian Maya; Consuelo Montes de Correa (211-219).
The performance of Pd/Co–HMOR for the oxidation of 1,2-dichlorobenzene (o-DCB) was significantly enhanced under wet conditions while the activity of Pd/Co–SZ was almost the same either under dry or wet conditions.▪The oxidation of 1,2-dichlorobenzene (o-DCB) has been examined over Co–HMOR, Pd/Co–HMOR, Co–SZ and Pd/Co–SZ under dry and wet conditions. Under dry conditions, Pd/Co–SZ was the most active catalyst. However, in the presence of water Pd/Co–HMOR exhibits a significant activity while Pd/Co–SZ was not affected. Besides, by-product formation such as polychlorinated compounds decreased on both catalyst types under wet conditions. Catalysts were characterized by XRD, BET, TGA, TPR, and NH3-temperature programmed desorption (TPD). The sulfation process greatly increased the surface areas of zirconia catalysts. No significant changes in the H2-TPR profiles of Co–SZ and Pd/Co–SZ under dry and wet conditions were observed while Co2+ ions at exchange positions of mordenite decreased on used mordenite catalysts. In spite of a higher total acidity (NH3-TPD) of mordenite catalysts compared to sulfated zirconia catalysts, zirconia-sulfated catalysts exhibited higher activity when dry feeds were used.
Keywords: Oxidation; Ortho-dichlorobenzene; Sulfated zirconia; Mordenite; Cobalt; Palladium cobalt-sulfated zirconia; Palladium cobalt mordenite;

The adsorption of ortho dimethyl benzene (o-DMB) at different adsorption temperatures T a (T a  > 300 K) on a SiO2 solid pretreated at 723 K is studied by Temperature Programmed Adsorption Equilibrium methods developed previously to provide the evolution of the adsorption equilibrium θ e with T a at constant adsorption pressures P a. These experimental curves are compared to theoretical curves associated to adsorption models developed with the statistical thermodynamics formalism to validate mathematical expressions for the adsorption coefficient K and the θ e  =  f(T a, P a) curves.▪The adsorption of ortho dimethyl benzene (o-DMB) at different adsorption temperatures T a (T a  > 300 K) on a SiO2 solid pretreated at 723 K is studied by the Temperature Programmed Adsorption Equilibrium methods developed previously. These methods provide the evolutions of the adsorption equilibrium coverage of the adsorbed species θ e (θ e  < 0.7) with the adsorption temperature T a in quasi isobar conditions. These experimental curves θ e  =  f(T a) are compared to theoretical curves associated to adsorption models developed with the statistical thermodynamics formalism. These models assume either localized or mobile adsorbed species without and with interactions. It is shown that the Langmuir model (localized species without interaction) provides theoretical isobars overlapped with the experimental data for different adsorption pressures P a considering a heat of adsorption of 61 kJ/mol consistent with the isosteric heat of adsorption. FTIR data show that the adsorption sites are mainly the free OH groups of SiO2 with a small contribution of superficial oxygen species. In line with the development of the experimental microkinetic approach of heterogeneous catalytic processes, and considering previous works dedicated to the adsorption of diatomic molecules such as CO, H2, NO on metal supported particles and metal oxides, it is concluded that adsorption models assuming localized adsorbed species without (Langmuir model) and with (Temkin model) interactions provide robust mathematical expressions, for (a) the adsorption coefficient and (b) the adsorption equilibrium coverage θ e  =  f(T a, P a), consistent with the experimental data for T a  > 300 K.
Keywords: Ortho di methyl benzene; Silica; Adsorption equilibrium; Heat of adsorption; Adsorption model; FTIR;

The relationship between acidity, Mo2C loading and carbon deposition during selective ring opening (SRO) of naphthalene over Mo2C/HY catalysts is reported. Increased Mo2C loading decreased catalyst acidity and carbon deposition on the acidic HY support, whereas no significant carbon deposition occurred on the Mo2C. A synergistic effect between the Mo2C and the acidic support (HY) was observed.▪The selective ring opening (SRO) of naphthalene over a series of Mo2C/HY catalysts with 7–27 wt% Mo2C, is reported. Pulsed adsorption and temperature programmed desorption of n-propyl amine showed that increased Mo2C content of the Mo2C/HY catalysts significantly decreased the amount and strength of the catalyst acid sites, especially the Brönsted acid sites necessary for SRO. Carbon deposition was the main cause of catalyst deactivation and thermogravimetric analysis of the used catalysts indicated that accumulation of carbonaceous species over the surface of the acidic HY support with time-on-stream led to deactivation of the Mo2C/HY catalysts, whereas no significant carbon deposition occurred on the Mo2C. Catalyst activity tests showed a synergistic effect between the Mo2C and the acidic support (HY) that was not readily obtained with a mechanical mixture of the two catalyst components.
Keywords: Ring opening; Hydrogenation; Catalyst; Molybdenum carbide; HY zeolite; Acidity; Carbon deposition; Deactivation;

The influence of surface functionalization of activated carbon on palladium dispersion and catalytic activity in hydrogen oxidation by V.Z. Radkevich; T.L. Senko; K. Wilson; L.M. Grishenko; A.N. Zaderko; V.Y. Diyuk (241-251).
Stone-fruit activated carbon functionalized with ethylamine groups facilitates higher Pd dispersions and catalytic activity in hydrogen oxidation than materials modified with NH3, H2O2, amide or HNO3.▪Stone-fruit activated carbon (SAC) and modified versions containing acidic oxygen and basic nitrogen groups have been used to prepare palladium catalysts by wet impregnation. Carbon supports and catalysts are investigated by thermo-gravimetric analysis, TPD, oxygen chemisorption, TEM and XPS. The influence of the nature of the functional groups on the dispersion and oxidation state of palladium and its activity in hydrogen oxidation is investigated. Pd dispersion is found to increase with the basic strength of functional groups on the support. XPS reveals that introduction of amine groups in SAC results in an increased proportion of Pd0, resistant to re-oxidation. Palladium catalysts supported on activated carbon modified by diethylamine groups are found to exhibit the highest metal dispersion and greatest activity in hydrogen oxidation.
Keywords: Activated carbon; Surface functional groups; Surface amine groups; Palladium supported catalysts; Hydrogen oxidation; XPS;

The activation of Phillips Cr/silica catalysts by M.P. McDaniel; K.S. Collins; E.A. Benham; T.H. Cymbaluk (252-261).
Although critical to activity and polymer properties, activation of the Phillips Cr/silica polymerization catalyst is never accomplished as efficiently on a commercial scale as in the laboratory. This is because Cr(VI) thermal stability depends on minimizing moisture released from the necessary dehydroxylation. This principle is used to successfully model commercial results from laboratory data. ▪This report attempts to define the underlying principles governing activation of the Phillips Cr/silica catalyst during fluid bed calcining. Cr(VI) is destabilized at high temperatures by water vapor, and therefore the key to a successful activation is in balancing two conflicting goals: (1) achieving extreme dehydroxylation of the silica surface, while (2) simultaneously minimizing Cr(VI) exposure to the H2O by-product. Moisture can lead to the growth of large crystallites of α-chromia which are difficult to reoxidize. Three main variables can be used to control moisture concentration in the gas stream during commercial activation—catalyst charge weight, air flow, and thermal ramp profile. In this study the sensitivity of Cr(VI)/silica to moisture, or to the absence of O2, has been defined as a function of temperature. Such laboratory data then allows one to model and predict the performance of Cr/silica catalyst activations when carried out on a commercial scale using a wide range of recipe variables.
Keywords: Phillips chromium catalyst; Ethylene polymerization; Polyethylene; Catalyst activation;