Studies of the structures and redox chemistries of chromate zeolites

Abstract

Solid-state reactions of sodium chromate, sodium molybdate, and sodium tungstate and zeolites of the types A, X, Y, mordenite, and silicalite at temperatures between 465°C and 655°C resulted in substantial uptakes of the salts. The occluded oxyanions could not be removed by boiling water indicating either a strong interaction with the zeolite lattice or occlusion in β-cages. Factors found to influence the amount of oxyanion occluded were temperature, time, and zeolite to oxyanion mole ratio. Under certain conditions maximum loading was achieved within 24 hours in the absence of significant zeolite framework degradation. In the chromate system, solid-state reactions performed near 800°C resulted in collapse of the zeolite framework and crystallisation of chromate nosean. The kinetics of chromate uptake by 13X followed a √t dependence over the initial stages of reaction indicating a diffusion controlled rate determining step. An apparent activation energy of 124 ± 10 kJ mol⁻¹ was calculated. Occluded sodium chromate was removed from X and Y type zeolites upon heating in strong alkali or salt solutions. A salt-exchange mechanism for chromate removal is proposed. The physiochemical nature of chromate loaded zeolites was examined using a variety of experimental techniques including PXRD, EPR, XRF, DTA, SEM, XPS, FTIR and Raman spectroscopies, and BET surface area determination. All results were consistent with the occlusion of chromate within the α-cages of X, Y, and A type zeolites. No evidence for β-cage occlusion was observed. Maximum chromate uptakes per unit cell were 13.6 for 13X, 13.4 for Y52, and 32.3 for 5A. These uptakes correlated with half the number of S6R facing α-cages of these zeolites suggesting coordination across two S6R. A model consisting of the oxyanion coordinated across its two sodium cations located near the Sɪɪ sites is proposed. Coordination across S5R of mordenite and silicalite was proposed. The Si/ Al ratio had no effect on the maximum chromate uptake. Redox reactions of sodium chromate salt were investigated. Reduction by H₂₍g₎ formed NaCrO₂ and NaOH. Reoxidations were incomplete ceasing at an average oxidation state of Cr(+4.67) through the formation of Na₂CrO₄ and Na₂CrO₃ in a 1:2 ratio. The addition of NaOH to the partially oxidised product allowed full oxidation indicating a dependence of oxidation product on NaOH concentration. A mechanism consistent with the degree of reoxidation was deduced. Chromate occluded in zeolites was reduced by H₂₍g₎ and reoxidised by O₂₍g₎ The degree of loading affected the product of reduction. Reductions of chromates in lowly-loaded zeolites ceased at Cr(IV), chromates in moderately-loaded zeolites (MLZ) were reduced to Cr(III), while chromates in highly-loaded zeolites (HLZ) were reduced to an average oxidation state of Cr(+3.5) containing equimolar amounts of Cr(III) and Cr(IV). The formation of occluded Cr(III)-Cr(IV) “dimers” was postulated. Chromates in chromate nosean reacted with H₂₍g₎ to give a reduction product similar to that of HLZ, i.e. with an average oxidation state of Cr(+3.5). A unit cell edge contraction from 9.107(2) Å for chromate nosean to 8.916(6) Å upon reduction was shown to be consistent with the formation of [Na₄.CrO₂]³⁺, [Na₄.CrO₃]²⁺, [Na₄.OH]³⁺, and [Na₄]⁴⁺ clusters per two unit cells. Changes to the X-ray pattern upon reduction to an average oxidation state of Cr(+3.5) indicated transformation of chromate noselite to a reduced chromate sodalite. The movement of the reduced chromate species off the anion site and coordination to the lattice was envisaged. Reaction with O2(g) at 600°C formed chromate sodalite. Retention of the sodalite framework was attributed to the anion lying off the anion site. Heating the chromate sodalite in air at 1000°C for 2 hours reformed chromate noselite through movement of the chromate oxyanions back to anion sites. The effect of temperature, zeolite type, and degree of loading on the kinetics of redox reactions were investigated. Unlike the reactions in nosean, transport control was not a feature of the reactions and zeolite structure had no significant effect on the redox kinetics. Reductions generally followed second-order kinetics and were highly activated processes with apparent activation energies between 86 ± 2 kJ mol⁻¹ and 97 ± 10 kJ mol⁻¹. In contrast, reoxidations were usually described by chemisorption kinetics. Preliminary investigations indicated a lack of activity of chromate zeolites in ethylene polymerisation and oxidation of methanol reactions. Chromate zeolites were also unable to oxidise CO(g) and CHi(g).