American Mineralogist, Volume 84, pages 1870–1882, 1999
Chabazite: Energetics of hydration, enthalpy of formation, and effect of cations on stability SANG-HEON SHIM,1 ALEXANDRA NAVROTSKY,2,* THOMAS R. GAFFNEY,3 AND JAMES E. MACDOUGALL3 2
1 Department of Geosciences, Princeton University, Princeton, New Jersey 08544, U.S.A. Thermochemistry Facility, Department of Chemical Engineering and Materials Science, University of California at Davis, Davis, California 95616, U.S.A. 3 Air Products and Chemicals, Inc., Allentown Pennsylvania 18195, U.S.A.
ABSTRACT The stability of synthetic cation-exchanged zeolites having the chabazite framework (CHA) and their cation-water interaction were studied using high-temperature reaction calorimetry. Four cations (K, Na, Li, and Ca) were exchanged into CHA. The enthalpies of formation were determined for all samples, and the partial molar enthalpy of hydration was measured by varying the water content of one Ca-CHA. The enthalpy of formation depends strongly on the exchanged cation, becoming more exothermic in the order Ca, Li, Na, K. The integral hydration enthalpy does not depend strongly on the nature of the cation, but becomes slightly less exothermic with increasing aluminum content. For the one Ca-CHA studied in detail, the average enthalpy of hydration is –34.6 ± 1.2 kJ/mol relative to liquid water. A – quadratic fit to the transposed temperature drop data gives ∆ hH2O = –52.97 (± 4.74) + 2.94 (± 0.68) n – (kJ/mol), where ∆ hH2O is the partial molar enthalpy of hydration and n is the number of water molecules per 24 oxygen formula.
INTRODUCTION Zeolites [(Li,Na,K)a(Mg,Ca,Sr,Ba)d(Al (a+2d)Si n–(a+2d)O2n)· mH2O] are hydrated framework aluminosilicates that incorporate molecular H2O in structural pores. The interaction of water with cations and with the framework in a zeolite cage is important in a variety of applications, including pollution abatement, catalysis, cation exchange, and gas separation. Thermodynamic data are essential to evaluate zeolite stability and characterize the hydration process. Zeolite minerals occur in several near-surface geologic settings. Thermodynamic data help determine their paragenesis. Extra-framework cations and the Al/Si ratio affect stability and hydration of zeolites. Barrer and Langley (1958) and Barrer and Baynham (1956) found that the thermal stability (temperature at which the zeolite decomposes on heating at a constant rate) of chabazite increased in the order Li, Na, K, Rb, Cs, and water retentivity increased in the order Rb, K, Na, Li. Cartlidge and Meier (1984) observed that Na-exchanged natural chabazite showed solid-state transformation to a sodalite structure at 600 °C in a dry nitrogen atmosphere, whereas K-exchanged natural chabazite did not transform. According to their Rietveld refinement, Na and K occupied different positions and coordination states with H2O molecules and framework O atoms.
*E-mail:
[email protected] 0003-004X/99/1112–1870$05.00
Although these observations are mostly qualitative and both kinetic and thermodynamic factors are probably involved, it is obvious that the cations play an important role in zeolite stability. Many studies have shown that there are two major types of water in the zeolite cage: one has strong interaction and may be associated with the hydration of the extra-framework cations, and the other has little interaction with the zeolite and may simply fill the voids. For example, leonhardite (CaAl2Si4 O12·3.5H2O) shows significant differences in the enthalpy of hydration associated with H2O molecules in different positions (Kiseleva et al. 1996a). Although the loss of one mole of H2O in the initial stage of dehydration has a near zero enthalpy of interaction (6.4 ± 9.2 kJ/mol, relative to liquid water), removal of the water molecules from the Ca coordination sphere requires 40.2 ± 2 kJ/mol. Carey and Bish (1996, 1997) studied the clinoptilolite-H2O equilibrium and measured the partial molar enthalpies of hydration. The hydration enthalpies became less exothermic in the sequence Ca, Na, K at low water content, but converged to similar values at saturation. Water having weak interaction with the framework was also observed in cordierite, Mg2Al4Si5O18·nH2O. Cordierite absorbs water on a well-defined crystallographic site without hydration of any cations. Its enthalpy of hydration (relative to liquid H 2O) is zero over a range of water contents (Carey and Navrotsky 1992). This implies that the molecular environment of H2O in cordierite is energetically similar to that of H2O in liquid water. 1870
SHIM ET AL.: THERMODYNAMIC PROPERTIES OF CHABAZITE
Hydration may be a complex process of H2O gain and structural rearrangement coupled with extra-framework cation movement (chabazite: Smith 1962; Smith and Rinaldi 1963; Mortier et al. 1977; zeolite Y: Dooryhee et al. 1991; heulandite: Armbruster and Gunter 1991). Although some structural studies proposed the sites for extra-framework cations and H2O molecules (Alberti et al. 1982), the numbers and locations of such sites are often difficult to deduce even from single-crystal X-ray structure refinements because of ambiguity in site assignments, positional disorder of H2O, and difficulty in accounting for the effects of different cations (Carey and Bish 1996). Vieillard (1995) proposed a method of predicting enthalpy of formation based on known crystal structures and applied it to zeolites. However, ambiguity in the determination of cation sites and H2O location in zeolite cages hinders the accurate prediction of energetics using such models. Therefore, to understand better the cation-water interaction in zeolite cages, it is necessary to determine accurate energetic data for the hydration process as well as structural variation during hydration. Such an understanding is essential to predicting the energetics, mechanisms, and rates of hydration and dehydration of zeolites in natural environments ranging from soils to volcanic glass. A better understanding of the influence of the cation on the stability of a zeolite is desirable to allow design of catalysts and adsorbents with the high thermal stability required to survive activation in commercial processes. By choosing appropriate mixtures of cations, adsorbents with good performance and improved thermal stability can be developed. This study focuses on chabazite (ideal form Ca 2Al 4 Si8O24·12H2O) and reports heats of formation and heats of hydration for several synthetic materials having the CHA-framework type. The enthalpies of formation for the cation-exchanged chabazites were measured using high-temperature, drop-solution calorimetry in molten lead borate to investigate the effect of cation type and Al/Si ratio on energetics. To observe the hydration energetics in detail, transposed-temperature, drop calorimetry was performed for one Ca-CHA sample with varying water content. For all other samples, an integral enthalpy of hydration, obtained from transposed-temperature, drop calorimetry, is reported.
EXPERIMENTAL METHODS Synthesis and characterization of initial chabazite samples All the chabazite (CHA) samples were synthesized in the potassium form, using different methodologies depending on the Si/Al ratio desired. At the highest Si/Al ratio, the K-CHA was prepared via a zeolite to zeolite conversion similar to that of Bourgogne (1985). A 2 M KOH solution was prepared. To 300 cc of that caustic solution, 142.22 g of a colloidal silica solution (Nalco 2326, which is an ammonia-stabilized, colloidal silica solution, nominally 14.5 wt% SiO2) was added. This mixture was also clear and colorless. Fifty grams of a commercial hydrogen zeolite (HY) powder (LZY-62 analyzed by the manufacturer as 72.2 wt% SiO2, 24.0 wt% Al2O3, 2.55 wt% Na2O) was added and the mixture was briefly homogenized by shaking the container. The sealed container was placed in a forced-air oven set at 100 °C. After 5 days, the solid was separated from the clear solution using vacuum filtration, and the
1871
solid product was washed with deionized water while on the filter. The solid was allowed to air dry on the filter and was recovered. X-ray powder diffraction (XRD) showed a pattern typical for a K-CHA with an Si/Al ratio above 2.5. 29Si NMR determined the framework Si/Al ratio to be 2.65, in good agreement with the elemental analysis (Si/Al = 2.7, Na/Al = 0.02, K/Al = 1.07). The other K-CHA samples were prepared using standard gel synthesis techniques (Barrer et al. 1959). To prepare CHA samples with Si/Al ≅ 1.5 some Sr(NO3)2 was used in the synthesis. These samples retain Sr after ion exchange. Chemical analyses of all zeolite samples were done by inductively coupled plasma mass spectrometry (ICP-MS) and atomic absorption spectrometry (AA). Ion exchange and characterization Cation-exchanged chabazite samples were prepared by Air Products and Chemicals, Inc. For example, the Ca ion-exchanged form of CHA 084-32 was prepared using batch ion exchange, by contacting the K-CHA with a 2 M solution of CaCl2 at a solution to solid ratio of 25 cc/g. The mixture was heated overnight at 100 °C, washed with deionized water, and air dried on the filter prior to the next exchange. A total of five exchanges were conducted. This was necessary to overcome the unfavorable exchange of K out of the zeolite. Elemental analysis revealed an excess of Ca (Ca/Al = 1.06 on an equivalent fraction basis) in the solid. This excess suggests the possibility of imbibed salt into the chabazite, as is often the case with unfavorable ion-exchange isotherms, however, no chloride was detected. In addition, ~4% of the original K ions remained after this exchange procedure. Other ion-exchanged forms were prepared using similar standard batch techniques with excesses of the exchanging cations. Thermal analysis to determine H2O content was done on a TGA 2951, TA Instruments, using an N2 purge and a 10 °C/ min ramp rate. Analytical data are summarized in Table 1. Throughout this paper, we use the term hydrated or “fully hydrated” to mean samples having water contents representative of their initial preparation followed by equilibration for several weeks in our laboratory (23 ± 1 °C and 50 ± 5% relative humidity). Preparation of partially hydrated samples To investigate the energetics of hydration in more detail, we selected a sample (084-32) that does not change its structure during transposed-temperature drop calorimetry (TTD) to 700 °C. This was confirmed by measuring the powder XRD patterns using a Scintag PAD-V diffractometer before and after heating for one hour at 700 °C. The pattern of the heated samples showed only minor changes, namely, peak shifts to higher angle, likely due to water loss (see Fig. 1). To determine the water content of the sample, thermogravimetric analysis was performed on a Netzsch Thermal Analyzer STA 409 in static air using a pellet weighing about 40 mg. The water content of the initial sample was 22.27 wt%. Because dehydrated Ca-CHA hydrates spontaneously under ambient conditions, its hydration rate was determined to control the water content. The sample was completely hydrated overnight and then heated at 700 °C for 1 hour. The mass
1872
SHIM ET AL.: THERMODYNAMIC PROPERTIES OF CHABAZITE
TABLE 1. Chemical compositions of CHA-zeolites (moles per 24-oxygen formula) Major cation Ca
Li
Na
K
Sample 113-91 084-32 084-99 084-11 084-48 084-84 084-51 113-92 084-34 084-27 084-60 084-96 084-87 113-90 084-53 084-16 084-95 084-85 113-89 084-14 084-93 084-77
Si 8.66 8.77 7.09 7.93 7.20 7.29 8.75 8.68 7.11 8.04 7.96 7.04 6.25 8.69 8.00 8.02 7.14 6.15 8.69 8.82 7.11 6.13
Al 3.34 3.23 4.91 4.07 4.80 4.71 3.25 3.32 4.89 3.96 4.04 4.96 5.75 3.31 4.00 3.98 4.86 5.85 3.31 3.18 4.89 5.87
Ca 1.58 1.63 2.24 2.29 2.80 2.84 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.02 0.00 0.00 0.00 0.08
Li 0.00 0.00 0.00 0.00 0.00 0.00 3.05 3.17 3.75 3.27 3.76 4.70 5.60 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Na 0.00 0.03 0.00 0.04 0.00 0.00 0.13 0.15 0.15 0.62 0.20 0.24 0.02 3.33 3.95 4.28 4.40 5.51 0.00 0.00 0.00 0.00
K 0.36 0.13 0.52 0.04 0.06 0.08 0.01 0.01 0.02 0.02 0.01 0.01 0.00 0.01 0.01 0.02 0.01 0.01 3.23 3.39 4.79 5.39
Mg 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Zn 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00
Sr 0.00 0.00 0.00 0.00 0.08 0.00 0.00 0.00 0.17 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
H 2O 12.83 12.54 14.68 13.40 12.95 15.16 11.79 10.88 12.52 12.59 12.70 14.52 15.17 11.69 12.57 11.89 13.36 16.50 8.90 9.88 10.63 10.71
These rate studies enabled the selection of conditions for preparation of partially hydrated samples. After being completely hydrated under ambient conditions overnight, portions of the sample were heated at various temperatures between 300 and 700 °C for 1 hour and then immediately placed in the glove box. Pellets of 10–20 mg were made in the glove box. The water contents were measured by mass change, assuming that the completely hydrated sample contained 22.27 wt% of water as determined by TGA. Each water content was confirmed by measuring the mass change of a portion of each sample after complete hydration under ambient conditions. All samples, except the ones with 5–7 wt% H2O, were completely rehydrated under ambient conditions. The hydration experiments confirmed within error the water content measured by dehydration. The samples that contained 5–7 wt% H2O were hydrated to 15–17 wt% H2O in ambient condition, as confirmed by TGA. The XRD patterns of all samples confirm retention of the CHA framework (Fig. 1). Calorimetry FIGURE 1. The X-ray diffraction (XRD) patterns of Ca-CHA (08432). Top pattern is for original sample, middle pattern for sample heated one hour at 700 °C, bottom for rehydrated sample.
changes were measured under ambient conditions and in a glove box. In ambient air, the sample was completely hydrated in 20 minutes. The atmosphere in the glove box is Ar purified using a 5 Å molecular sieve to have a moisture content of
