Reduction of background due to backscattered electrons in energy-dispersive X-ray microanalysis

When energy-dispersive X-ray microanalysis is practised in a scanning electron microscope, most of the spectral background may come from electrons entering the detector. This background can be eliminated by deflecting magnets. Alternatively, the electrons can be blocked by a plastic film but only at the cost of suppression of the low-energy end of the X-ray spectrum.     

Evaluation of several common standards for the X-ray microanalysis of thin biological specimens

The question of the best type of standard to use for X-ray microanalysis of thin biological specimens remains unanswered. Standards embedded in an organic matrix have the advantage that they resemble biological specimens, but their composition is generally not known exactly. We compared several standards and, surprisingly, inorganic binary salts sprayed onto a supporting film were the most suitable: they corresponded closely with several other methods using organic matrices; they were easily produced; and their composition is known. Glutaraldehydeurea aminoplastic resin thin sections and thin films containing dissolved salts were problematic. The composition of the polymer appears to be variable, and the thin films did not correspond with any other standard tested. Chelex¹⁰⁰ bio-standard beads and flakes loaded with accurately determined concentrations of ions, embedded in epoxy resin and thin sectioned, tended to correspond to the results obtained with the binary salts. However, the results from some bio-standards were inexplicably aberrant. An epoxy resin standard was used for bromine, and was found to agree closely with the binary standards.     

Aminoplastic standards for quantitative X-ray microanalysis of thin sections of plastic-embedded biological material

The preparation of a set of aminoplastic standards for quantitation of sodium, magnesium, phosphorus, sulfur, chlorine, potassium, and calcium in ultrathin plastic sections of soft tissue is described. The standards are low in cost and easy to prepare and handle. They cover concentrations up to 750 mmoles/kg dry weight, and display chemical and physical properties similar to those plastic-embedded samples. The standards can be used to convert peak-to-continuum ratios obtained from the specimen to elemental concentrations. An application of these standards is shown using rat exocrine pancreas as a model. The standards can also be used to advantage for the calibration of the detector efficiency.     

Mass thickness determination by electron energy loss for quantitative X-ray microanalysis in biology

As is well known, electron energy loss spectroscopy can be used to determine the relative sample thickness in the electron microscope. This paper considers how such measurements can be applied to biological samples in order to obtain the mass thickness for quantitative X-ray microanalysis. The important quantity in estimating the mass thickness from an unknown sample is the total inelastic cross section per unit mass. Models for the cross section suggest that this quantity is constant to within ±20% for most biological compounds. This is comparable with the approximation made in the continuum method for measuring mass thickness. The linearity of the energy loss technique is established by some measurements on evaporated films and quantitation is demonstrated by measurements on thin calcium standards. A significant advantage of the method is that the energy loss spectrum can be recorded at very low dose, so that mass thickness determination can be made before even the most sensitive samples suffer damage resulting in mass loss. The energy loss measurements avoid the necessity to correct the continuum measurement for stray radiation produced in the vicinity of the sample holder. Unlike the continuum method the energy loss technique requires uniform mass thickness across the probe area, but this is not usually a problem when small probes (<100 nm diameter) are used.     

The use of thin specimens for X-ray microanalysis in biology

The quantitative X-ray microanalysis of ultrathin biological sections is exemplified by a recent study of the distribution of calcium in mineralizing cartilage and bone. The determination of calcium mass-fraction in one microarea of a vesicle within a section of rabbit epiphyseal plate cartilage is presented in detail in order to display all steps of the processing of the data. Mass fractions are obtained from an equation which is approximate but which is adequately accurate for most cases of interest, specifically for ultrathin biological sections where the analysed area consists predominantly of organic matrix. We shall examine the physical assumptions behind the computation, and the extent to which these assumptions have been verified in the analytical microscope. The purpose of this paper is to describe in detail the scheme of data collection and processing, which has now been in use for some years in the Cavendish Laboratory, for the measurement of local elemental mass fractions in thin biological specimens in instruments like the EMMA. After an outline of the physical theory we shall describe the handling of the data by means of working through an example of an actual measurement, and finally we shall add some comments, mainly precautions and reservations.     

Improved quantitative electron probe X-ray microanalysis of biological cryosections using an EDS germanium detector and a super-atmosphere thin window (SuperATW)

A new Link energy-dispersive GEM detector with SuperATW window was tested for quantitative electron probe microanalysis of low calcium and sodium concentrations ([Ca], [Na]) in intracellular compartments of cardiac myocytes. We compare Ca profiles with high count statistics and similar peak area collected under the same conditions with either a Be-windowed Si and a Ge SuperATW detector. The height of the Ca peak was increased by 7%, the full width at half-maximum height was reduced by 9% with the Ge detector. The counts statistics of the Ca Kα peak improved by 9% and the partial overlap with the K Kβ peak was better deconvoluted. We calculate [Ca] and errors of the single measurement in mitochondria of guinea-pig cardiac myocytes from spectra acquired with a Si or Ge detector. For identical analysis conditions, the [Ca] were identical; however, with the Ge SuperATW detector, the calculated error of the single measurement was only 1/2.7 of that calculated from measurements with the Si detector. We compare the peak area of identical [Na] in spectra collected with the Be-windowed Si detector and Ge SuperATW detector. The peak area was significantly higher with the SuperATW Ge than with the Si detector and Be window, whereas the continuum in the range 4–10 keV was comparable, demonstrating the improved sensitivity for low atomic elements such as Na of the Ge SuperATW detector. [Na] and errors of the single measurement in mitochondria of quiescent guinea-pig cardiac myocytes were calculated from spectra acquired with the Si or the Ge detector. The use of the Ge SuperATW detector improved the detectability limit for sodium by more than 80% and reduced the error of the single measurement by a factor of 7–8.     

Suggestions for the quantitative X-ray microanalysis of thin sections of frozen-dried and embedded biological tissues

When a microregion in a thin section of frozen-dried and embedded tissue is analysed by the conventional electron-probe X-ray continuum-normalization method, the measured quantity is in mmol of element per kg of embedded specimen. As each microregion contains an unknown amount of embedding medium, this quantity generally lies indeterminately somewhere within the wide range between mmol of element per kg of hydrated tissue and mmol of element per kg of dehydrated tissue. However, if a ‘tag’ element is incorporated in the embedding medium, the contribution of the medium to the local continuum count in each probed field should be measurable, and the X-ray data may then unambiguously yield mmol of element per kg of dehydrated tissue. This result should not be affected by shrinkage on freeze-drying or by incomplete replacement of water by embedding medium. The same X-ray data can additionally provide estimates of mmol of element per unit volume, mmol of element per kg of hydrated tissue and local dry-mass fraction. However, these estimates are subject to errors due to tissue shrinkage, incomplete replacement of water and beam damage.     

Progress in quantitative X-ray microanalysis of frozen-hydrated bulk biological samples

The analysis of bulk frozen-hydrated biological samples has developed now to a level where practical application of the technique is possible. Provided the sample is carefully coated with a conductive metal, the development of a space charge capable of causing a significant distortion of the electron diffusion volume does not seem to occur, and analytical resolution can be conveniently held to approximately 2 μm (both depth and lateral resolution). Two valid quantitative methods are available, and two methods of determining dry weight fractions are also available. An area where further research could lead to improvement in analysis of frozen-hydrated bulk samples is in the investigation of fracturing methods. If fracture planes that were flat and reproducible could be easily obtained, some of the difficulties of analysing frozen-hydrated bulk samples would be considerably reduced.     

Mass thickness determination and microanalysis of thin films in the TEM--revisited

Calibrating on a series of thicknesses of a single element we arrive at a simple method of mass thickness determination for amorphous and polycrystalline films in the transmission electron microscopy (TEM). The conditions to be fulfilled are: acceleration voltage 200 kV or higher and measuring the integrated transmitted electron intensity in a wide range of angles that contains also the Bragg reflections. Under these conditions the electron scattering depends on Z2 as predicted by the Rutherford approximation. There is no need to carry out a calibration for atomic number dependence. For multicomponent films of unknown composition the combination of electron and X-ray intensity measurement results in absolute (non-normalized) concentrations and the value of local mass thickness. Examples of quantitative analysis are presented and an extension to single crystal films is discussed.     

Application of the parametric bootstrap method to determine statistical errors in quantitative X-ray microanalysis of thin films

We applied the parametric bootstrap to the X‐ray microanalysis of Si‐Ge binary alloys, in order to assess the dependence of the Ge concentrations and the local film thickness, obtained by using previously described Monte Carlo methods, on the precision of the measured intensities. We show how it is possible by this method to determine the statistical errors associated with the quantitative analysis performed in sample regions of different composition and thickness, but by conducting only one measurement. We recommend the use of the bootstrap for a broad range of applications for quantitative microanalysis to estimate the precision of the final results and to compare the performances of different methods to each other. Finally, we exploited a test based on bootstrap confidence intervals to ascertain if, for given X‐ray intensities, different values of the estimated composition in two points of the sample are indicative of an actual lack of homogeneity.     

Calibration for quantitative X-ray microanalysis of skeletal muscle cells in culture.

Skeletal muscle in culture is used to demonstrate that the intracellular concentration of elements in tissue culture cells, especially of diffusible ions such as sodium, chlorine, potassium, and calcium can be measured by X-ray microanalysis in the scanning electron microscope. Quantitative analysis is possible by comparison with X-ray spectra of cells incubated in electrolyte solutions of known concentration. The minimum detectable concentration is approximately 2 mM.     

A sample preparation for quantitative determination of magnesium in individual lymphocytes by electron probe X-ray microanalysis

We present a sample preparation method for measuring magnesium in individual whole lymphocytes by electron probe X-ray microanalysis. We use Burkitt's lymphoma cells in culture as the test sample and compare X-ray microanalysis of individual cells with atomic absorption analysis of pooled cell populations. We determine the magnesium peak-to-local continuum X-ray intensity ratio by electron probe X-ray microanalysis and calculate a mean cell magnesium concentration of 39± 19 mmol/kg dry weight from analysis of 100 cells. We determine a mean cell magnesium concentration of 34 ±4 mmol/kg dry weight by atomic absorption analysis of pooled cells in three cell cultures. The mean cell magnesium concentrations determined by the two methods are not significantly different. We find a 10% coefficient of variation for both methods of analysis and a 30% coefficient of variation in magnesium concentration among individual cells by electron probe X-ray microanalysis. We wash cells in ammonium nitrate for microanalysis or in buffered saline glucose for atomic absorption analysis. We find cells washed in either solution have the same cell viability (85%), recovery (75%), cell volume (555 μm³) and cytology. We air dry cells on thin film supports and show by magnesium X-ray mapping that magnesium is within the cells. We conclude that: (a) our microanalysis cell preparation method preserves whole intact lymphocytes; (b) there is no systematic difference in results from the two methods of analysis; (c) electron probe X-ray microanalysis can determine the variation in magnesium concentration among individual cells.     

Experimental and theoretical determination of kAFe factors for quantitative X-ray microanalysis in the analytical electron microscope

Quantitative X-ray microanalysis in the analytical electron microscope involves the use of Cliff-Lorimer k_AB factors to relate measured X-ray intensities from elements A and B to their composition. This study has generated a wide range of these factors for both K and L X-ray lines. The values of the k factor for each element is ratioed to Fe, rather than to Si as has been common practice to date. The use of k_AFe rather than k_ASi factors reduces the uncertainty in the values due to variations in the efficiency of individual energy dispersive spectrometers, thus making the values more universally applicable. New calculations of the value of k_AFe have been made from first principles, encompassing all of the most recent values of the cross-section for X-ray ionization. Comparison of the experimental results with both the calculations and existing k factor data has been made. Close attention has been paid to minimizing the errors in both the experimental and theoretical calculations to reduce the overall error in quantification of X-ray energy dispersive spectra in the analytical electron microscope.     

X-ray microanalysis of wet biological specimens in the environmental scanning electron microscope. 1. Reduction of specimen distance under different atmospheric conditions

X-ray microanalysis of non-biological and biological specimens was carried out in the environmental scanning electron microscope (ESEM) under different conditions of specimen distance (the distance travelled by the electron probe within the specimen chamber) and chamber atmosphere. Using both water vapour and argon atmospheres, it was shown that reduction in specimen distance had no effect on atmospheric gas X-ray signal in either case. Unlike water vapour, increased levels of argon (up to 10 torr) caused a marked depression of specimen P/B ratios, with a decrease in both characteristic and background (continuum) counts. These effects in argon were not altered by reduction in specimen distance. Specimen distance was important in relation to beam skirting and elemental analysis. With an extended assembly (short specimen distance), beam skirting in a water-vapour atmosphere was much reduced – leading to enhanced element detectability in a discrete biological specimen (Anabaena cyclindrica).     

Calcium in secretory vesicles of neurohypophysial nerve endings: quantitative comparison by X-ray microanalysis of cryosectioned and freeze-substituted specimens

The calcium content of individual secretory vesicles in rat neurohypophysial nerve endings was measured by quantitative electron probe X-ray microanalysis. Directly frozen control and potassium-depolarized isolated endings were analysed using two presumably equivalent preparative techniques: (1) freeze-substitution in presence of oxalic acid followed by sectioning of resin-embedded pellets; or (2) direct cryosectioning of the frozen pellets followed by freeze-drying in the column of the microscope. In the pellets of stimulated endings, both approaches revealed an increase in the calcium content of neurosecretory vesicles. This increase was statistically more significant in the specimens prepared by cryosectioning, probably because in this case the contribution of 'dead' nerve endings could be eliminated on the basis of excessive cytoplasmic sodium and chloride. The results demonstrate that an increase in cytosolic calcium can lead to an increase in intravesicular calcium, and that when this occurs, it occurs within a subpopulation of vesicles in a given nerve ending. In addition, measured intravesicular calcium was dispersed over a wide range of concentrations, as predicted by the hypothesis of intravesicular calcium priming. When the vesicular calcium content was averaged per nerve ending, a relatively wide distribution of concentrations was again observed, indicating that some nerve endings respond more strongly to the stimulation than others.     

Quantitative measurements of Kikuchi bands in diffraction patterns of backscattered electrons using an electrostatic analyzer

Diffraction patterns of backscattered electrons can provide important crystallographic information with high spatial resolution. Recently, the dynamical theory of electron diffraction was applied to reproduce in great detail backscattering patterns observed in the scanning electron microscope (SEM). However, a fully quantitative comparison of theory and experiment requires angle-resolved measurements of the intensity and the energy of the backscattered electrons, which is difficult to realize in an SEM. This paper determines diffraction patterns of backscattered electrons using an electrostatic analyzer, operating at energies up to 40 keV with sub-eV energy resolution. Measurements are done for different measurement geometries and incoming energies. Generally a good agreement is found between theory and experiment. This spectrometer also allows us to test the influence of the energy loss of the detected electron on the backscattered electron diffraction pattern. It is found that the amplitude of the intensity variation decreases only slowly with increasing energy loss from 0 to 60 eV.     

Development and comparison of the methods for quantitative electron probe X‐ray microanalysis analysis of thin specimens and their application to biological material

In recent years, there has been a return to the use of electron probe X‐ray microanalysis for biological studies but this has occurred at a time when the Hall programme which acted as the mainstay for biological microanalysis is no longer easily available. Commercial quantitative routines rely on the Cliff‐Lorimer method that was originally developed for materials science applications. Here, the development of these two main routines for obtaining quantitative data from thin specimens is outlined and the limitations that are likely to be met when the Cliff‐Lorimer routine is applied to biological specimens is discussed. The effects of specimen preparation on element content is briefly summarized and the problems encountered when using quantitative analysis on resin‐embedded materials emphasized.     

X-ray microanalysis of organic thin sections in TEM using an UTW Si(Li) detector: comparison of quantification methods

We compared Hall's peak to continuum ratio method with a peak ratio method in order to quantify light elements (C, N, and O) in organic specimens as a model for biological thin sections. X-ray spectra were recorded by an energy dispersive X-ray spectrometer equipped with an ultra thin window detector. Spectra were processed by means of a top-hat filter adapted to peak full-width half maximum. The peak intensities were measured by multiple least square fitting to reference spectra. For most elements of biological interest, theoretical and experimental k-factors were determined. Absorption correction was found to be important for quantitation of carbon, nitrogen, and oxygen. Boron was efficiently detected; however, quantitative analysis was not possible. We conclude from our experiments that the peak ratio method is more suitable for quantitation of elemental concentrations in biological thin sections than the peak to continuum method.