X-ray Fluorescence Thin-Film Sample Support Materials
Reprinted from American Laboratory, 17 (11) November 1985
By Dr. Monte J. Solazzi
THE RETENTION of liquid, powdered solid, solid, and slurry samples in disposable XRF Sample Cups® (Chemplex® Industries, Inc.) in preparation for energy- (EDXRF) and wavelength-dispersive (WDXRF) x-ray fluorescence (XRF) spectrochemical analysis has been previously described 1. This report describes the various types of sample cups available, some advancement in thin-film materials and the parameters used to assess them, and illustrates the influence of these materials on analyte-line transmittance.
Since the inception of WDXRF and EDXRF instrumentation, rapid technological developments have extended the range of analytical interest to lower atomic number elements, lower limits of detection, and diversified scientific disciplines, thus necessitating thin-film sample support materials more reproducible in gauge thickness and providing higher analyte-line transmittance.
A substance used in thin-film sample supports must exhibit a relatively high degree of resistance to chemical attack, excitation energy source embrittlement or degradation, thermal softening, and deterioration from the heat generated by excitation exposure, and possess relatively good sample retention strength. The materials must also be reasonably free of impurities and possess the necessary combination of chemical composition, density, and gauge to impart minimum absorption of both the primary (excitation energy) and secondary (analyte-line) radiation.
Until 1972, the most commonly used thin-film sample support material was 6.3 m gauge polyethylene terephthalate (Mylar®). Later, 2.5 m and 3.8 m gauges were introduced. These three gauges together effectively accommodated virtually all areas of analytical interest, but due to a change in the manufacturing process of Mylar, trace quantities of Ca, P, Sb, and Zn were detected. Converting to a different grade of Mylar reduced the concentration of these elements, and two of the gauges were substituted with 3.6 and 6.1 m gauges. In addition, three polymeric materials-polypropylene, polyimide, and polyethylene-qualified as thin-film sample supports and were found to be relatively free of potentially interfering impurities.
Initially, a suitable polypropylene was difficult to locate since a specific biaxially oriented type was required to avoid stretching. Thin-film supports must resist stretching and subsequent random alteration of gauge thickness and variations in analyte-line transmittance. X-ray polypropylene (Chemplex Industries, Inc.) was later offered in a 6.3 m gauge as a replacement for Mylar. Although 6.3 m gauge polypropylene resulted in a 5.4% decrease in analyte-line percent transmittance in comparison to the 2.5 m gauge Mylar at 12.4 Angstroms, only a slight increase in integration time was required. Polyimide (Kapton®) in 7.6 m gauge was also offered for specialized applications requiring increased sample retention strength for the more active analyte-line investigations and for vacuum applications. These thin-film sample support materials are shown in Figure 1.
The suitability of a substance for use in thin-film sample supports depends on its ability to permit analyte-line transmittance. This property is determined from the total mass attenuation coefficient of the material in combination with its area concentration for a given analyte-line, obeying Lambert's law in accordance with the following formula:
I= IO exp [- (µ/d) dt]
Where, I = incident intensity;
I0 = transmitted intensity;
µ/d = mass attenuation coefficient, cm2/g
dt = area concentration, g/cm 2.
This relationship shows the dependence of analyte-line transmittance on thin-film chemical composition, density, and thickness: (µ/d)(dt). Since the total mass attenuation coefficient of a thin-film material for a specific analyte line is based on the sum of its elemental constituent values, the chemical composition of the substance is critical and determines suitability for this application.
Rearrangement of the equation to express the percentage of incident and transmitted radiation, I/lo, as a function of mass attenuation coefficient, density, and thickness, exp [-{µ/d)(dt)], provides a visual presentation of analyte-line transmittance through the thin-film substance.
Figure 2 demonstrates the reduction in percent transmittance with increased analyte-line wavelength (decreased keV values) for 6.3 m gauge polypropylene. The effect of thin-film thickness on analyte-line transmittance is illustrated in Figure 3. The curves represent the percent analyte-line transmittance relationship for the same thin-film substance. (Mylar was arbitrarily chosen as the test material.) The displacement of the curves from one another is attributed to gauge differences. Note that absorption effects predominate in the long wavelength (low keV) region; the more energetic analyte lines tend to penetrate a thin-film substance with very little resistance as transmittance approaches 100%.
Other thin-film substances and gauges have been similarly assessed. The resultant curves differ completely from one another in individual combinations of mass attenuation coefficient, density, and gauge relationships. By constructing a series of superimposed curves encompassing a variety of thin-film substances and gauges, a rapid and effective visual means is established for selecting the most applicable thin-film material and thickness for specific analyte investigations (see Figure 4). However, other influential variables-the properties more directly associated with sample retention-are equally significant to thin-film material and gauge selection.
Degradation Resistance
A thin-film substance is also assessed by properties jointly classified as "degradation resistance." Degradation resistance as defined here represents the ability of a thin-film material to safely retain a specimen in an XRF sample cup during preparation and analysis. Degradation resistance includes resistance to chemical attack, thermal softening, embrittlement, tearing, and stretching.
The immediate chemical attack on a thin-film substance by contact with a specimen is usually obvious. However, deterioration is not always so evident: it may worsen with time or by heat induced from excitation, embrittlement from excitation energy exposure, or a combination of these and other events.
Although it is important to use the thinnest possible gauge thin-film substance to maximize analyte-line transmittance, very thin gauges tend to increase the threat of rupture under the weight of the specimen or under pressure differential created when the sample being analyzed is contained in a sealed sample cup in vacuum. A substance characterized by relatively high tensile strength is required. The tensile strength associated with most polymeric substances described is almost equal to 5000 psi. The thickness of the film then becomes the principal governing factor for sample retention strength.
A thin-film material must also resist stretching. Any changes associated with the thickness of a thin-film material are reflected by the degree of analyte-line transmittance and its influence on analytical accuracy. Any stretching of a thin-film material upon attachment to a sample cup will be reflected in the analytical data. Under vacuum, a differential in pressure between a sealed sample cup and the optics will cause the thin film to distend, creating two problems: a decrease in the distance from the excitation source to sample plane (defined by the thin-film sample support surface plane), resulting in false higher intensity measurements and analyte concentrations, and a decrease in the thickness of the thin film by stretching, resulting in an increase in analyte-line transmittance, thus implying a higher analyte concentration than actually exists.
In assessing thin-film substances, little information was available relating analyte-line transmittance and degradation resistance properties. Chemical and physical characteristics are generally expressed in terms of tensile strength, elongation, tensile modulus, tear strength, type of substance related to diversified and extended time and temperature exposures-all of which leave interpretation and evaluation to the spectroscopist for thin-film sample support applications.
A substance-screening procedure for thin-film applications that assigns a rating value combining degradation resistance and analyte-line percent transmittance properties emerged based on two criteria: failure of a single critical property or a combination of less critical individual properties. Substances were classified as good, fair, or poor (Table 1) and were related to chemical classifications as a common denominator for ease of referral and comparison instead of the conventional chemical material listings traditionally provided in chemical classifications.
Table 1
Degradation Resistance of Thin-Film Substances
Degradation Resistance Of Thin-film Substances
| Chemical | Mylar® | Poly- carbonate | Etnom-S | Poly- propylene | Poly- imide (Kapton®) | Prolene® | Ultra- polyester |
| Acid, dilute or weak | G | G | G | E | N | G | G |
| Acids, conc. | G | G | G | E | N | E | G |
| Alcohols, aliphatic | N | G | G | E | G | E | N |
| Aldehydes | U | F | F | E | E | E | U |
| Alkalies, conc. | N | N | G | E | E | E | N |
| Esters | N | N | F | G | G | G | N |
| Ethers | F | N | F | N | U | N | F |
| Hydrocarbon, aliphatic | G | N | G | G | G | G | G |
| Hydrocarbon, aromatic | F | N | G | F | F | F | F |
| Hydrocarbon, halogenated | F | N | F | N | F | N | F |
| Ketones | N | N | G | G | G | G | N |
| Oxidizing agents | F | N | F | F | N | F | F |
E=Excellent, G=Good, F=Fair, N=Not recommended, U=Unknown
NOTE: The information contained in the above illustrations is provided as a matter of information only and it is not intended to preclude actual testing of the subject material for suitability of use and applications.
Percent Transmittance Comparisons
Most polymeric materials permit 95% to 100% transmittance through all gauges for analyte lines less than 4 Angstroms (3.1 keV). With increasing analyte-line wavelength for a particular thin-film material, the need for a thinner gauge correspondingly increases because of higher degrees of absorption. In evaluating a thin-film substance, the entire range of analyte lines of interest and anticipated analyte concentrations in a specimen should be considered. Thin-film material and thickness should be selected that provide the greatest degree of transmittance, particularly for low concentration levels and long analyte-line wavelengths, together with other pertinent properties. In many instances, the analytes and concentration levels are not previously known and a general-purpose thin-film substance should be used.
Teflon prohibits 50% analyte-line transmittance at 3.1 keV and Kapton loses 50% of its analyte-line transmittance at 1.7 keV. Both substances are, however, very well-suited for use as thin-film sample supports because of their excellent degradation resistance, but are limited (Teflon in particular) to use with the more energetic analyte lines.
Mylar exhibits good properties with respect to degradation resistance and percent analyte-line transmittance, but it has the drawback of inherent detectable trace levels of impurities. This can be a problem if the same elements at similar concentrations are to be quantified.
A superficial examination of polycarbonate shows this substance to be acceptable for thin-film applications based on its analyte-line transmittance, which is very similar to that of 3.6 m gauge Mylar. However, the degradation resistance of polycarbonate makes it unsuitable for use with a broad range of chemical classifications. Its use is restricted to chemically unreactive solution specimens or powdered solid sample materials.
Polypropylene and polyethylene are similar in degradation resistance and transmittance properties. These substances are useful for the retention of many types of sample materials and can be used for thin-film sample supports encompassing the entire spectral range. A polyethylene device incorporating both a snap-on ring and thin-film sample support membrane is available from Chemplex Industries, Inc. (Figure 5). This device eliminates the need for separate attachment of a thin-film sample support material with a snap-on ring.
Polyvinylidene chloride (PVC) exhibits a sharp discontinuity in the percent transmittance correlation to analyte-line appearing at 2.8 keV; this represents the K-absorption edge for chlorine. This discontinuity and the substance's unacceptable degradation resistance, make PVC unsuitable for use in XRF sample retention.
Polystyrene exhibits poor degradation resistance, and is thus not suited for use in XRF sample retention. Other materials similarly tested and found to be unacceptable included celluloses, collodian, and nylon. Polyvinyl fluoride is currently being evaluated as a possible material for sample containment.
Gauge Thickness
Uniformity in thin-film thickness is important with respect to minimizing variations in analyte-line transmittance. The process of manufacturing involves gauge control to within approximately +/- 10% of nominal thickness. To evaluate the effect of variations in thickness on analyte-line absorption, percent transmittance deviations from a variety of gauges of the same thin-film substance were calculated for a number of analyte lines. Mylar was used since most of the test data were already available.
Results indicated a correlation between percent transmittance deviation units ( +/- %TDU) to decreasing thin-film gauge and increasing analyte-line wavelength. This condition would be most pronounced with very thin-gauge substances and long analyte-line wavelength investigations. For example, a deviation of +/- 3% TDU was determined for 2.5 µm gauge Mylar at an arbitrarily selected analyte line of 12.4 Angstroms. Translation of 36.79 +/- 3% transmittance to analyte concentration is insignificant for analytical concern.
Summary
Several parameters pertaining to the suitability of thin-film substances for retaining samples for WDXRF and EDXRF spectrochemical analysis have been described. The materials investigated were all synthetic polymers of various thicknesses, exhibiting differing degradation resistance and analyte-line transmittance. A comparison of percent analyte-line transmittances for thin-film substances and gauges were presented, intended to facilitate selection of a suitable thin-film substance and gauge.Note 1:
Mylar® is a registered trademark of E.I. DuPont de Nemours Co., Inc.
Prolene®, SpectroMembrane® and Chemplex® are registered trademarks of Chemplex Industries, Inc. Etnom is a trademark of Chemplex Industries, Inc.
Note 2:
Table 1, Degradation Resistance of Thin-Film Substances, is replaced with the most current version.
Note 3:
Figure 5. SpectroMembrane thin-film integrally formed with snap-on ring has been replaced with an alternate product of different design and configuration
Reference
1 SOLAZZI, M.J., "Disposable XRF sample cups and thin-film sample supports for x-ray fluorescence analysis," Am. Lab. 16 (11), 72- 78 (1984).
Dr. Monte J. Solazzi is President, Chemplex Industries, Inc., 2820 SW 42nd Avenue, Palm City, Fl. 34990, USA. Tel: (772) 283-2700.