Verified Syntheses of Zeolitic Materials

2nd Revised Edition

Determination of the elemental compositor of zeolitic materials

Walter Zamechek
UOP Research Center, Des Plain es, IL 60017-5016, USA

1. Introduction

The determination of the bulk elemental composition of zeolites is of importance in many aspects of zeolite synthesis, characterization and applications. This information is used to verify the synthesis formulations, the bulk silica/alumina ratio, the cation(s) concentration, degree of ion exchange, and the detection of contaminant elements (impurities, poisons). The elements of interest can be grouped into two broad categories, metals and non-metals. The latter commonly include sulfur, chlorine, carbon and nitrogen. Generally the results are reported on volatile free basis, thus the Loss on Ignition (LOI) at a specified temperature and time is also determined. The concentration of water is determined if the request is for the results to be based on anhydrous basis and volatile components other than water are present. The information herein is a brief review of the techniques and their characteristics as used for the determination of bulk elemental compositions of zeolites.

2. Determination of metals

The most common techniques for the determination of compositional metals are Inductively Coupled Plasma Emission Spectroscopy (ICP), Atomic Absorption Spectroscopy (AAS) and X-Ray Fluorescence Spectroscopy (XRF). Because these methods offer the benefit of reduced interferences and matrix effects, and have improved accuracy, precision and speed, the use of "classical wet chemistry" methods (for example, gravimetric silica, titrimetric or spectrophotometric alumina) has been greatly diminished.

ICP is probably the most widely used technique for the determination of the elemental composition of zeolites. It offers the capability for the simultaneous (or sequential) determination, with good sensitivity and precision, of most compositional matrix metals of interest, for example silicon, aluminum, phosphorous, titanium, and many others. Relative standard deviations of 1% can be routinely obtained for the major and minor compositional metals. In general, the sensitivity of ICP is better than that possible with conventional flame AAS for most refractory-like metals and phosphorus. However, flame AAS has somewhat better sensitivity for Group IA elements, including sodium and potassium, and relatively similar sensitivity for many metals of interest, such as calcium, magnesium and iron. Because the cesium emission lines are above 800 nm, conventional ICP instrumentation is incapable of its determination and AAS has to be utilized.

Wavelength dispersive X-Ray Huorescence (XRF) is also used for the determination of the elemental composition of zeolitic materials. As compared to ICP/AAS, the benefits of XRF include the ability to determine some non-metals, conceptually simpler sample preparation and improved precision. Relative standard deviations of -0.1 to 0.2% are possible for samples introduced as glass discs. The disadvantages include poor sensitivity for light elements and sensitivity to changes in the matrix composition. This means that in many cases XRF can not perform the complete characterization. ICP/AAS has to be employed for some of the determinations, for example, lithium and low concentrations of sodium. Also, especially for best accuracy, changes in the matrix compositional elements require the use of matrix matched calibration standards and the use of mathematical models. Thus XRF has its greatest impact in a controlled, manufacturing environment, while in an R & D environment ICP/AAS is often the technique of choice.

3. Sample decomposition for ICP and AAS

Both conventional ICP and AAS require that the sample be introduced as a liquid, thus decomposition is necessary prior to analysis and similar preparation schemes apply for both techniques. There are two main approaches for sample dissolution in order to determine major and

minor compositional elements (including silicon). The sample can be solubilized by a fusion with lithium tetraborate (or a similar flux) followed by dissolution of the flux in dilute hydrochloric acid, or it can be digested with acid in a beaker.

a. Beaker digestion for ICP or AAS analysis

Although this approach is used widely and can produce good data, it has several areas of concern. The acid digestion may not completely solubiize binders, such as clay, increasing the severity of the decomposition conditions (e.g., higher temperature) can generally ensure complete dissolution, but such a modified digestion may require elevated pressures in a Teflon lined bomb. The digestion requires the addition of hydrofluoric acid in order to solubilize the silicon, and because of the volatility of the resulting fluorides, special care must be taken to avoid the loss of silicon.

A more serious problem is the fact that fluoride will attack a conventional ICP sample transport system that contains quartz/glass components, thus necessitating the complexation of the fluoride. This can be accomplished utilizing boric acid, or a commercial complexing agent. In the authorÔs experience these solutions mostly extend torch life, they do not completely eliminate attack on the quartz and thus accurate determination of silicon is not possible. Hydrofluoric acid resistant sample transport systems are available from instrument manufacturers. We have found these to be inadequate because the typical alumina sample delivery torch tube is attacked by fluoride thus contributing to increased aluminum and silicon (impurity in the alumina) backgrounds. Replacing the alumina tube with sapphire eliminates this problem, unless very high concentrations of fluoride are present. [11 A commercial sapphire torch tube is currently available from at least one manufacturer.

Typical AAS sample introduction systems are silica free and the fluoride containing solutions can be analyzed, but the precision for silicon and aluminum is generally inadequate. If the silicon is of no interest, these issues, except for completeness of dissolution, become irrelevant. In fact, for the determination of minor and trace metals, it is often desirable to volatilize the silica during sample decomposition, thus reducing the total dissolved solids.

b. Dissolution by Fusion for ICP or AAS analysis

It is the authorÔs experience that sample dissolution by fusion results in better precision and accuracy than possible by a beaker digestion. Although other fluxes can be used, the author fuses the sample, in a platinum crucible, using a 15:1 ratio of lithium tetraborate to sample. The fusion can be accomplished using automated fusion devices, via a muffle furnace or air boosted natural gas burners. The resulting melt is allowed to cool to room temperature and is then dissolved with dilute (5-10%) hydrochloric acid. In case of the automated devices, the hot melt is poured directly into the dilute acid. This is preferable as it enables faster and less troublesome dissolution. In order to facilitate the pouring of the hot melt, a small amount of cesium iodide (same weight as the sample) is weighed into the platinum crucibles together with the flux and the sample material.

4. Sample preparation for X-ray fluorescence analysis

There are two main approaches utilized for sample preparation in order to achieve the best precision for major compositional elements. The sample is fused akin to the fusion described above. The resulting hot melt is poured into a platinum mold and forms a glass disc that is introduced to the XRF. While a number of fluxes can be used, lithium tetraborate is often the flux of choice for zeolitic materials. in order to reduce the sensitivity of XRF to the matrix composition, lanthanum oxide (20%) can be added to the lithium tetraborate flux. A number of approaches can be taken in order to prepare calibration standards; however, the fusion based sample preparation approach enables the addition of elements of interest (usually as oxides) directly into the platinum crucible prior to the fusion.

Pressing pulverized samples into pellets is also a viable approach. The drawback is that this technique is likely to have poorer precision than the fusion and be more sensitive to changes in the matrix composition. However, it is satisfactory for many applications and it enables the

determination of non-metals, such as halogens and sulfur, that would be volatilized during the fusion. Often, if the desired precision can be achieved, the samples can be introduced to the XRF simply as ground powders, without pressing The author routinely determines sulfur and chlorine with a detection limit of 100 ug/g.

5. Determination of non-metals

a. Determination of carbon, hydrogen and nitrogen (CHN)

The determination of CHN is generally requested because these elements are present in an organic amine template used during synthesis or because the zeolite adsorbs these during a specific application. In either case conventional CHN analyzers are frequently used for this determination. Alternative approaches include the determination of nitrogen by classical Kjeldahl digestions and the determination of carbon by combustion analyzers specific for only carbon. The hydrogen concentrations from a CHN will include hydrogen from water and hydrocarbons. Karl-Fischer based determination of water, at specified temperature, can be used to differentiate the two sources of hydrogen.

b. Determination of sulfur

The determination of sulfur is generally accomplished via a combustion type analyzer, often together with carbon. Commercial equipment exists for single or multi-element analysis of C, H, N, and S. There are several alternative non-oxidative techniques; these include XRF with a detection limit of about 100 ug/g, or specialized reduction-distillation of the sulfur as hydrogen sulfide followed by calorimetric detection via methylene blue.

c. Determination of chlorine

The determination of chlorine can be accomplished via a variety of leaching or closed vessel (microwave) digestions followed by silver nitrate titrations of the liberated chloride. Alternatively, leaching (which may include full dissolution by addition of hydrofluoric acid) followed by Ion Chromatography has been found very useful as it enables the detection of chloride as well as nitrate, sulfate and ammonium. Also, XRF can be used, with a detection limit of about 100 ug/g chlorine.

d. Determination of fluorine and chlorine via pyrohydrolysis [2]

A specialized technique has been implemented for the determination of fluorine. The sample is pyrolyzed in a nickel combustion tube in a steam atmosphere at 10000C. The steam/analyte eluent is scrubbed and the fluoride concentration is then determined via Ion Chromatography or Ion Selective Electrode. Chlorine is volatilized together with the fluorine and can also be determined.

6. Summary

All the techniques described herein have their strengths and weaknesses and areas of overlap. The selection is dictated by specific sensitivity and precision requirements and by instrument availability. Each approach described above can be modified to address specific requirements. For example, a graphite furnace and hydride AAS can be used with great sensitivity for the determination of trace contaminants such as arsenic or mercury.

7. References

[1] Walter Zamechek, Robert G. Pankhurst, in Proceedings of the International ICP Winter Conference, San Juan, Puerto Rico, 1980, Heyden Press, p 121
[2] Analytical set-up based on Anal. Chem. 32 (1960) 118