WO2013035082A1

WO2013035082A1 - Combined method of secondary ion mass spectroscopy and energy dispersive x-ray for quantitative chemical analysis of various solid materials and thin films without the use of specific patterns or standards

Info

Publication number: WO2013035082A1
Application number: PCT/IB2012/054657
Authority: WO
Inventors: Iouri KOUDRIAVTSEV; José Pablo René ASOMOZA Y PALACIO
Original assignee: Centro De Investigación Y De Estudios Avanzados Del Instituto Politécnico Nacional
Priority date: 2011-09-09
Filing date: 2012-09-08
Publication date: 2013-03-14
Also published as: MX2011009486A

Abstract

The present invention relates to a combined method of secondary ion mass spectroscopy (SIMS) and energy dispersed X-ray (EDX) for quantitative chemical analysis of solid materials and thin films, comprising the steps of a) installing a energy dispersed detector Si(Li) on a free port of a mass spectrometer for X-ray recording; b) placing the study sample in an ultra high vacuum (UHV) chamber of the mass spectrometer; c) activating the electron gun for the mass spectrometer to achieve the characteristic X-ray excitation of the study sample, wherein the electron gun provides a moderately focused electron beam with energies that can vary from 0 to 15keV and currents from 0 to 100 mA; d) analyzing the X-rays generated in step c) with the energy dispersed detector Si(Li), and find sensibility factors (RSF); e) calibrating the SIMS data with the sensitivity factors (RSF) obtained by EDX quantitative analysis of step d) using existing theories; and f) analyzing with SIMS any unknown material of the study sample.

Description

Combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of various solid materials and thin films without the use of specific patterns or standards Field of the invention.

The present invention comprises in general methods for analyzing materials by using electrical, electrochemical, or magnetic means to determine electric or magnetic properties, and particularly, relates a new analytical method, which combines traditional methods followed by a secondary ion mass spectrometer and a X-ray detector, a method which provides a quantitative chemical analysis of solid materials and thin films in a full dynamic range and discarding the need to use specific patterns and standards.

Background of the invention.

Different physical methods for chemical analysis of elements such as RBS, AES, XPS, EPMA, XRF, and Raman spectroscopy, have wide application in the semiconductor, metallurgy, pharmaceutical industries, among others, and are widely used in several investigations. These methods provide rapid and complete information on the chemical and phase composition, lateral and depth distribution of major elements, donors, and contaminants. They also provide necessary information on the chemical composition of new materials, and are also used in the control process for failure analysis and improvement of existing technology.

Among these methods, Secondary Ion Mass Spectroscopy (SIMS) has clear advantages such as a high sensitivity, which reaches up to one atom per 10¹¹ atoms of the matrix, a high lateral resolution, and in depth record resolution that allows effective analysis of all elements from hydrogen to uranium. It can also determine the isotope ratio through SIMS. So, the SIMS method is the most used in advanced materials research and different structures based on ultra pure crystalline semiconductors. From a physical standpoint, the SIMS method is based on the erosion (Ion Sputtering) of surface of a sample under ultrahigh vacuum condition and the mass separation of ions ejected.

Meanwhile, Raman spectroscopy is a technique used in chemistry and matter physics to study low-frequency modes such as vibrating, rotating, and others. This technique relies on the phenomena of inelastic scattering or Raman scattering of monochromatic light, usually from a laser in the visible light range, the near infrared, or near ultraviolet range. The photon interacts with laser light or other excitations in the system, causing the energy from the laser photons to undergo an upward or downward displacement. The energy shift provides information about the photon modes in the system. Infrared spectroscopy provides similar information, but complementary.

In view of the above, an example of a prior art for a method for chemical analysis by spectroscopy on the surface of a solid material is found in the US patent US4857730, which discloses and apparatus comprising a chamber for UHV analysis, which contains the sample to be analyzed, which is connected with a handle located on the exterior of said compartment, a close sample analyzer, and an electron emitting source that emits an electron beam, characterized in that it comprises, between the electron beam and the sample consisting of bulky solid material, an X-photon micro-source placed as close as possible to the bulky sample. Another example is represented in the US patent US7906759, which describes an economic mass spectrometer capable of obtaining structural information of a substance with an improved efficiency, wherein the time required for analysis and identification of the substance has been reduced and the identification accuracy is also improved. More specifically, this invention provides a system for tandem mass spectrometer in which the sample is ionized in the desired polarity; the fragments of the obtained ions by dissociating the ion are analyzed within the first or second section of the mass spectrometer, the polarity of the second mass spectrometer is determined based on the result of the analysis, and mass spectroscopy is performed. This invention also protects the method for mass spectroscopy.

The US patent US6008490 describes a method of measuring and analyzing a mass spectrum in which an ion is created under atmospheric pressure, or a pressure that is close there to, wherein the ion is introduced to create spectrometry masses and the resulting mass spectrum is processed and analyzed. The method comprises the steps of determining an index indicating how many times the masses differ between the selected mass of said resulting mass spectrum and other masses, according to the mass differences plurals between a quasi-molecular ion and a plurality of adduction type ions stored in a adduction ion storage previously established, and the estimation based on the index values determined for the selected masses, the ion provides a maximum value of said index indicating the difference in mass as a quasi-molecular ion.

Finally, the US patent US6492639 describes a method for identifying the characteristics deviation of a sample of reference by comparison of the mass spectrometric patterns of a sample with the reference. This method comprises the steps of determining a total mass spectrum of at least one random reference sample to a reference, to form a reduced mass spectrum associated with the related space tolerance of this total mass spectrum of the random sample of reference, determining a spectrum reduced mass of the sample to be compared with the random reference sample, and compare the reduced mass spectrum of the sample with the random reference sample to determine if the total reference spectrum is within the related space of the tolerance.

From the above documents on the state of the art we can infer that current methods for measuring and analyzing the physical and chemical properties of solid materials only allow qualitative results when using mass spectrometry, provided that during the result analysis process is implemented the use of standards of comparison. In contrast, when X-ray detectors are deployed, the results obtained reflect not only qualitative but also quantitative data of the sample studied, which means that current inventions in the prior art do not include a method of analysis on spectroscopy that allows determining from a single analysis the quantitative and qualitative characteristics of a test sample. Therefore, the Combined Method of Secondary Ion Mass Spectroscopy (SIMS) and Energy Dispersive X-ray (EDX) for quantitative chemical analysis of solid materials and thin films of the present invention solve these needs.

Objectives of the invention.

An objective of the present invention is to provide a method of analyzing materials, wherein a specific configuration of a secondary ion mass spectroscopy allows determining the quantitative and qualitative properties of a sample of solid materials and thin films under study.

Another objective of the present invention is to provide a combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of solid materials and thin films, wherein an X-ray detector is installed on a secondary ion mass spectrometer for the analysis of the sample.

Another objective of the invention is to provide a combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of solid materials and thin films, which allows the quantification in a full dynamic range without standards of solid materials and thin films under study.

Another objective of the invention is to provide a combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of solid materials and thin films, wherein the data obtained from a quantitative analysis with EDX were used for calibrating the SIMS data.

Another objective of the invention is to provide a combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of solid materials and thin films, wherein EDX provides information to the main elements, while SIMS is used both for dopants and pollution analysis and for three-dimensional distribution.

Another objective of the invention is to provide a combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of solid materials and thin films, wherein the SIMS and EDX methods work independently, but both analyze the sample without moving its position.

Another objective of the invention is to provide a combined method of secondary ion mass spectroscopy and energy dispersive X-ray for quantitative chemical analysis of solid materials and thin films, wherein the nondestructive character of EDX allows starting any analysis with this method.

The objectives of the present invention referred to above, and others not yet mentioned, will become apparent from the description of the present invention and the figures with an illustrative but not-limiting character are disclosed below. Brief description of the figures.

Figure 1. Shows a graph of results in semi-logarithmic scale of relative sensitivity factors

(RSF) for positive secondary ions of different elements ejected from a Si surface bombarded by primary ions of oxygen as ionization potential function of the element. Figure 2. Shows a graph of results in semi-logarithmic scale of relative sensitivity factors

(RSF) to negative secondary ions of different elements, ejected from a Si surface bombarded by primary ions of Cesium as function of the electron affinity of the element.

Figure 3. Shows a schematic diagram of a mass spectrometer type Cameca IMS-6F.

Figure 4. Shows a schematic diagram of the EDX technique performed with an electron gun, which is normally used for compensation of the surface charge that appears in the process of ion bombardment of several materials with high resistance and dielectric. Figure 5. Shows an EDX graph of a spectrum of Ti alloy with the elements identified in the material according to Example 1.

Figure 6. Shows a graph of the relative sensitivity factors (RSF) of the main elements found with EDX in the Ti alloy according to Example 1.

Figure 7. Shows the fragment Z/q 4-36 of SIMS mass spectrum with the elements detected in the Ti alloy according to Example 1 .

Figure 8. Shows the fragment Z/q 30-90 of SIMS mass spectrum with the elements detected in the Ti alloy according to Example 1.

Figure 9. Shows the fragment Z/q 80-160 of SIMS mass spectrum with the elements detected in the Ti alloy according to Example 1.

Figure 10. Shows the fragment Z/q 170-220 of SIMS mass spectrum of the elements detected in the Ti alloy according to Example 1.

Figure 11. Shows a graph of the SIMS depth profiles of elements detected with atomic numbers 1 to 48 in the Ti alloy according to Example 1 .

Figure 12. Shows a graph of the SIMS depth profiles of elements detected with atomic numbers 48 to 1 15 in the Ti alloy according to Example 1 .

Figure 13. Shows a graph of the concentration range of elements, which can be determined by the method of the invention (combination SIMS/EDX) according to Example 1.

Figure 14. Shows the certificate of the composition of Ti alloy according to Example 1.

Figure 15. Shows a representation of the experimental sample of the SiGe thin film used in example 2.

Figure 16. Shows the EDX spectrum obtained with 10 keV of sample N183 (SiGe) of figure 15 according to example 2.

Figure 17. Shows the EDX spectrum obtained with 5 keV of sample N183 (SiGe) of figure 15 according to example 2.

Figure 18. Shows the EDX spectrum obtained with 3 keV of sample N183 (SiGe) of figure 15 according to example 2.

Figure 19. Shows the SIMS depth profiles of secondary positive ion (+) monitoring obtained for

SiGe thin film (sample N183 of figure 15) for example 2.

Figure 20. Shows the SIMS depth profiles of secondary negative ion (-) monitoring obtained for

SiGe thin film (sample N183 of figure 15) for example 2.

Figure 21. Shows a "final" depth profile obtained for the SiGe thin film according to the present invention for the case of example 2.

Detailed description of the invention.

A SIMS instrument has several modes of operation:

1. Microscope-microprobe mode, used in the study of the lateral distribution of elements of interest;

2. Profilometer mode, used in the analysis of in depth samples; and

3. Spectrometer mode, used in the elemental analysis of samples.

The combination of modes 1 and 2 provides a three dimension (3-D) real analysis of any element or stable isotope in a solid structure.

The current of ejected "secondary" ions in SIMS is a function of the "primary" ions current j₀, and the concentration in the element matrix C_x analyzed as shown as follows:

/^■± _ . v . . P^± . Tr^±

x J o ¹ tot ¹ x ^{l r} x (1 ) wherein Y_tot is the total coefficient for obtaining target atoms;

P* is the probability of the analyzed element to form positive or negative secondary ions; and

T_r is the transmission coefficient of the instrument for a determined element.

After starting SIMS applications in the 60s of the last century, it was found that the coefficient for obtaining secondary ions (SIY) of any element, defined as the number of ions emitted by a primary ion l_x/j₀ of the equation (1 ), has a complex dependence of primary ions and the analyzed matrix, the vacuum conditions of the chamber where the sample is place, and, generally, the chemical composition of the analyzed surface. Any variation in the surface composition causes a change in the coefficient of obtaining of secondary ions called matrix effect, which does not allow the quantification of experimental results of SIMS. The conclusion is therefore that SIMS is generally a qualitative method. In the practice, the SIMS measurement quantification is performed with an additional measurement of a standard, acquired under the same experimental conditions. The preparation of standards is done frequently by ion implantation, increasing the effort, and the cost of the analysis. A coefficient called Relative Sensitivity Factor (RSF) can be obtained from the standards, where the concentration of the element of interest in other samples can be calculated if it measured under the same experimental conditions and the RFS previously calculated and the ratio of currents are multiplied. In the last 50 years there have been development different theories to explain the formation of secondary ions, such as the "quantum" or "statistical" models. According to these models, the probability of positive and negative ionization of the particles ejected depends on their ionization potential (I), the electron affinit (A), and the work function of the surface (W) as shown below:

wherein: ^ are constants used as adjustment parameters, and

/ and A are known values; however, the work function of the surface and η, are obtained theoretically.

Figures 1 and 2 show that for different materials exist a good semi-quantitative correlation between the ionization probability (equation 2) and experimental data. These figures show the relative sensitivity factors (RSF) to positive and negative secondary ions of different elements (ejected from a Si surface bombarded by primary ions of Cesium and Oxygen respectively), as function of the ionization potential (figure 1 ) and the electronic affinity (figure 2) of the element. It can be seen that quantitative analysis can be made with good accuracy if η, and the coefficients of the exponential functions in equation (2) are found by a lineal approximation in figures 1 and 2. We can speak of a precise analysis of factor 2 for the most of elements, which seems a reliable value if they studying element concentrations below 10¹⁸ atoms/cm³. For comparative analysis of similar samples, the relative error does not exceed ±20% (SIMS experimental error). The quantification method described using standards works well when studying elements whose concentrations are less than 1 % atomic; below this concentration level, the SIY of the analyzed element is proportional to its concentration. However, for the main elements this dependence becomes nonlinear and unpredictable, so that any SIMS composition analysis of solid solutions; namely, semiconductors, alloys, and multi-phase samples become a very serious problem. The common technique comprises preparing and measuring a set of solid solutions or composites of known compositions, preferably from 5 to 9 samples, thereafter the calibration curve is constructed for each element of interest. Therefore, this analysis demands much effort and its costs increase dramatically. So, other analytical techniques are preferred to obtain the information for the main elements in complex structures.

Among other techniques, energy dispersed spectroscopy (EDX) also known as electron probe microanalysis (EPMA) is widely used for the chemical analysis of different materials. In this method, the characteristic X-rays resulting from the excitation of the sample by electron bombardment are analyzed with a special semiconductor detector Si(Li). Direct quantification based on existing theories is easily possible for EDX with an experimental error of only 1 %. Furthermore, EDX technique is quite sensitive and nondestructive, and it reaches a limit of detection for most of the elements of 0.01 % atomic. It is therefore, usually acceptable as a method for the analysis of most of the main elements and dopants in some cases. Furthermore, a new generation of semiconductors of energy dispersed has been recently developed that can detect any element starting with Beryllium (Be), so that this technique provides a quantitative analysis for any modern compound and solid solutions, such as lll-V y ll-VI, ternary and quaternary compounds, Si_xGei_-x, Α'Β"Ό^ν2 and A"B^IVC^VI2 compounds, various glasses, Si_xC_YN_1-X, Si_xOi_-x, different metal alloys, and many other materials.

In this vein, the EDX method virtually is the ideal complement to SIMS for the elemental analysis of any solid material and most of the thin films. The method of the present invention implements a combination of both techniques in a single ultra high vacuum (UHV) system, wherein EDX provides information to the main elements, while SIMS is used for dopants and pollution analysis to determine the dimensional distribution.

Figure 3, shows that the mass spectrometer IMS-6F Cameca 10 is provided with two ion sources, one cesium (Cs⁺) ion gun (1 ) and duoplasmatron (2) (0₂ ⁺, Ar⁺). It also has an electron gun (EG) (3), which is used for charge compensation during SIMS analysis of highly resistive or dielectric samples. This EG gun (3) provides a moderately focused electron beam with energies that can vary from 0 to 15 keV, and currents of 0-100 μΑ. Said electron gun (3) can be effectively used for the excitation of characteristic X-rays for most of the elements at most of the interest materials. Also, the equipment has a free port (4) to install any ion gun, which, for purposes of this invention, is used to install a energy dispersed detector Si(Li) (5) for recording characteristic X-rays. Therefore, this combination electron gun (EG) (3) - energy dispersed detector Si(Li) (5) allows therefore, to perform the EDX analysis on the same system for mass spectrometry. According to the present invention, both methods work independently, but both analyze the sample without moving its position; that is, in the same vacuum chamber where SIMS technique can be used in the standard mode, while EDX technique can be performed as shown in figure 4, at the time that the ion beam (6) that cause the primary erosion on the sample (7) is arrested. By the EDX nondestructive nature, any analysis may start with the method of the invention. For the SIMS analysis of any unknown material, the data obtained from an EDX quantitative analysis can be used to calibrate the SIMS data. Because an EDX analysis can be performed for any element with a concentration from 0.01-100% atomic, the elements with a concentration between 0.01-1 % atomic will be used as SIMS calibration by using existing theories (see equation 2); i.e., coefficients ηί and the exponential coefficients in equation 2 will be found for this experimental system, and after the calibration, the SIMS data for concentrations below 1 % atomic can be accurately quantified owed to the exponential dependence between SIY, the ionization potential, and electron affinity of the elements of interest.

In view of the above and for purposes of the present invention, the method of the invention comprises the following steps: a) Install a energy dispersed detector Si(Li) on a free port of a mass spectrometer for recording X-rays;

b) Place the study sample in a ultra high vacuum (UHV) chamber of the mass spectrometer; c) Activate the electron gun of mass spectrometer to obtain the excitation of characteristic X- ray of the study sample, wherein said electron gun provides a moderately focused electron beam with energies that can vary from 0 to 15 keV and currents of 0-100 μΑ;

d) Analyze the X-rays generated in step c) above with the energy dispersed detector Si(Li), and search sensitivity factors (RSF) to the SIMS technique;

e) Calibrate with the sensitivity factors (RSF) obtained with EDX quantitative analysis of stage d), the SIMS data using existing theories; and

f) Analyze with SIMS any unknown material of the study sample.

Note that the installation of EDX can be performed on any type of mass spectrometer either magnetic sector or quadrupole or time of flight.

A clear example of the application of these methods combination according to the present invention is the analysis of volcanic glass or obsidians. Obsidians are a mixture of oxides such as Si0₂, Al₂0₃, Fe₂0₃, NaO, KO, MnO, H₂0, TiO, ZrO, etc. The study of these samples is a popular object of study in archaeology and geochemistry for many reasons. The SIMS method is not used in these cases because the composition of this type of samples is very complicated, unlike XRF and EPMA, which are the methods traditionally used for quantitative analysis. The XPS (or EPMA) method provides quantitative information of different elements, information that can be used for the SIMS quantification of those elements and more; for example, for all those elements with concentrations below 1 % (Cr, Co, Ni, Nb, Mo, Ag, Cd, Cs, Ce, La, Ta, Pb, etc.), and also for light elements (B, Be, Li, F, C, N, F), which cannot be analyzed with XRF (EPMA). The mentioned earlier calibration provides a unique opportunity for SIMS analysis of the hydration phenomenon in obsidians. The general idea of this analysis is to use the penetration of hydrogen in obsidians as a reference to determining their age. Unfortunately, the diffusion of hydrogen strongly depends on the chemical composition of the obsidian, a fact that has not been studied yet and currently there is no theory that enables the use of this method for comparing different archaeological artifacts found in different places. It is known that SIMS can solve this problem by the systematic investigation of the H diffusion, which is accompanied by the redistribution of other chemical elements in obsidian.

It must be stressed then that the method of the present invention opens a wide range of unique opportunities for analysis, because during a single experiment on any element and in most solid analysis it is possible to perform a quantitative analysis of the main elements, dopants, and contaminants in a concentration range from 100% atomic up to 10^"7 atomic percentage.

In addition, the electron gun used in conjunction with the electro-optical system of the mass spectrometers allows the variation of energy of the primary electrons between 0 to 15 keV, meaning that it can strongly vary the electron penetration length and, so, the volume analyzed owed to a high transparency of secondary X-ray compared to electrons. So, one can perform low-energy X-ray spectroscopy (LEXES), a method effectively used for the analysis of implanted samples with a superficial union. Note that this analysis is nondestructive for the sample analyzed.

A great practical interest represents the lateral distribution of dopants and contaminants on the thin films surface or interfaces. In this sense, SIMS analysis can be performed on the lateral distribution with a resolution of about 1 micrometer. Unfortunately, owed to the matrix effects, SIMS analysis of lateral distributions and three-dimensional analyzes are limited to an analysis of the isotope ratio and shows some inhomogeneities without a real estimate on the phase of locally observable areas. In practice, matrix effects are responsible for a false image, since any local change of obtaining of secondary ions (up to an order in magnitude) can be misinterpreted as a change of elemental concentration. As can be seen in what is described here, the new combined SIMS-EDX method of the present invention provides a unique opportunity to resolve this issue.

As will be seen later, with the method of the present invention it is possible to make a precise quantitative analysis of the composition of different materials without having to use standard patterns, both in volume and thin films.

EDX and SIMS methods by themselves do not allow the performance of the quantitative determinations described for the method of the present invention, because independently they have the following disadvantages: a) Analytical EDX technique.

This technique is accomplished by bombardment of surfaces of a solid material with accelerated electrons with energy between 1 keV up to 30 keV, which is the characteristic X-ray emission of the material. The analysis of these X-ray of certain energy allows determining the chemical composition of the analyzed sample because each atom that constitutes the material emits X- ray of a certain characteristic energy.

Analytical features:

a) All the elements from Carbon to Uranium can be analyzed, while the elements that cannot be analyzed with this technique are H, He, Li, Be, and B;

b) It has a detection limit for elements analyzed up to 0.01 % atomic, significantly depending on the sensitivity of the method of the atomic number Z;

c) The minimum volume (area of X-ray excitation) analyzed depends on the energy of primary electrons and the atomic density of the sample, and ranges between 0.2 up to 10 microns of approximation in relation to the aforementioned energies (0.5 to 30 keV);

d) It is a truly quantitative method, since it is possible to calculate the composition of the sample from the beginning with an experimental fail of 1 %. When the scanning is performed with the used of patterns, the method provides a quantification with an accuracy of <1 %; and

e) It is a non-destructive technique. b) SIMS analytical technique.

This technique is performed by bombarding the surface of the study sample with ions accelerated with energy between 0.5 keV up to 20 keV resulting in a destruction of the atomic structure of the material and the emission of surface atoms (secondary) with several electric charges: positive, negative, or neutral. An analysis of the atomic mass of the secondary particles positively or negatively charges, provides comprehensive information on the composition of the material analyzed. Continued erosion of the material with primary ions does not allow studying the composition of the material in three dimensions.

Analytical features:

a) All the elements of the periodic table can be analyzed;

b) Shows a sensitivity of up to 10^"7 % atomic in most elements;

c) Shows a better lateral resolution of one micron and depth of up to 1 nm;

d) It is a destructive method;

e) The quantitative analysis is only possible by using special patterns; the pattern for each element of interest must be prepared and made of the same material being analyzed;

f) The quantitative analysis is performed for elements with a concentration below 1 % atomic.

Between 10^"7 % atomic to 1 % atomic, the ratio of secondary ion current (/) and the concentration of the analyzed element (C) is a linear function according with the following: C_x= constant*!* / I_m (3) wherein /, is the ion intensity of the analyzed element (x) and the matrix element (m), the constant is called Relative Sensitivity Factor (RSF);

g) The analysis is performed with an experimental failure of ± 20%; and

h) When the concentration of the element of interest exceeds 1 % atomic, a non-linear dependence appears between RSF and the concentration of the actual element during the analysis. This dependence called "matrix effect" is currently unknown and depends on various factors, such as the system analysis, the primary ions used, etc. Quantification here is a very complex job that involves measuring a series of special patterns of known composition and the development of a "calibration curve". The analysis shows an experimental failure of ± 50%.

EDX provides the concentration of the main elements in different materials with perfect accuracy (<1 %), while the analysis of low concentration elements performed by SIMS is done with an accuracy of ± 20%. However, when such techniques are joined in a single device according to the present invention, the same area, or volume is analyzed and the calibration of the SIMS technique is possible done with the results obtained using EDX (standardless analysis).

By the present invention; for example, it is possible to perform the analysis of materials such as: a) Semiconductors and dielectric compounds (at volume and thick films);

b) Ternary and quaternary lll-V and ll-VI formed with AI-ln-Ga-As-Sb-P elements;

c) Nitrites-Ill formed with AI-Ga-ln-N elements;

d) SiyGe1-y and SiyC1-y semiconductors;

e) SiOyN1 -y oxynitrides (a new generation of MOS insulators and devices); and

f) GICS solar cells (formed with Cu-ln-Ga-Se-S elements).

As shown, by the method of the invention it is possible to perform a quantitative analysis of the composition of different materials at volume and at thin and thick films.

For all the above materials and the like, one of the modes in which the analysis is performed according to the present invention comprises, for example, determining the material composition and/or compound with EDX technique, and a subsequent analysis of contaminants and impurities with SIMS technique, including a three-dimensional, lateral (X-Y), and in-depth analysis. As already mentioned, when performed separately each of these methods (SIMS and EDX) provides limited information on the composition of the study materials.

For the analysis of semiconductor materials and dielectric compounds prepared in form of thin films by the method of the invention, all the above materials can be analyzed. Regarding the EDX technique, the primary electrons energy should have a low energy; for example, 5 keV, varying between 0.5 up to 5 keV, depending on the elements of interest and film thickness. Subsequently, SIMS technique is traditionally applied as described here.

For the analysis of implanted semiconductors with a low implantation energy of 10 keV (shallow junction) and new generation devices based on Si, the implantation of doping elements such as B, P, As, Sb, etc. with low energy results in the formation of a very thin profile of this element in the matrix with a typical thickness of less than 5 nm. Traditionally, the implanted semiconductors are analyzed with SIMS technique (a depth profile) where by definition the implanted element is at a concentration range of 10¹⁴ up to 10²¹ atoms/cm³. However, a bombardment of the surface of these materials with ions of different nature (in analysis process by SIMS technique) results in the formation of a composite film on the surface (matrix elements plus implanted ions). As a result, the transient effect appears in the measurement process of the depth profiles, not allowing an accurate analysis of the first depth nanometers (1 -3 nm), which depends of the energy of the primary ions. The result of this transient effect is a great failure in the definition of the implantation dose performed by SIMS technique.

In contrast, when such materials are analyzed by the method of the present invention, the concentration of the implanted elements with a dose higher than 10¹³ ions/cm² and with a low energy of 10 keV at the maximum implanted, the atomic percentages are reached. This concentration can be obtained by EDX technique of low energy primary electrons (<5 keV), and with a high accuracy (<1 % failure). Subsequently, this experimental dose can be applied to update the results obtained with SIMS (a depth profile) for the first 2-3 nanometers of the material profile. It is noteworthy that according to the present invention, SIMS and EDX analysis are performed in the same analysis chamber, on the same area of analysis, and practically simultaneously (the analysis time average is 20 min), which provides results in a short time, limits the handling of the analyzed material sample, and produces more reliable results.

On the other hand, EDX technique is traditionally used for the chemical analysis of materials with a highly complex composition (containing at least 10 elements), such as glass (including natural glass), ceramic materials, metal alloys, etc. However, and as mentioned above, only the main elements of these materials can be analyzed with this technique.

In contrast, the analysis of doping elements and/or contaminants (with a concentration less than 0.1 % atomic), their depth distribution at the material (thin films, and diffusion study, etc.) and on the surface of such materials by the method of the present invention is done with SIMS technique, performing the calibration of SIMS with the results produced by EDX. According to the present invention, this calibration provides sensibility factors for all the elements of the periodic table (owed to an exponential dependence between the ionization of atoms ejected and its ionization potential). Among other advantages, the application of SIMS for the analysis of a wide variety of materials according to the method of the present invention offers the opportunity to quantitatively analyze light elements and their isotopes, such as H, D, Li, C, N, O.

Moreover, the method of the present invention permits to study and analyze materials such as rare earth elements (REE) at volcanic glasses (geochemistry and geochronology), hydrogen at obsidians for dating obsidian by hydration (archaeology and geology), compositions made of glass (glass industry), elaborate ceramic compositions (ceramic industry), and compositions of metal alloys (metallurgy and special materials).

Preferred embodiment of the invention.

In the present invention, the electro-optical system of a CAMECA electron gun is modified to include a traceability function for the electron gun (initially, CAMECA electron gun does not track). The same tracking signal is applied to a CRT monitor, in which the electrical signal for each pixel is modulated by the experimental intensities of characteristic X-rays of the analyzed element. Therefore, it is possible to perform the analysis of lateral distribution by EDX (or a real EPMA method). The electron gun can be focused in about 1 micrometer, which corresponds to the lateral resolution during the SIMS analysis. Again, the nondestructive EDX analysis of the interest surface provides quantitative information about the lateral distribution of the interest elements. Moreover, the SIMS analysis of the same surface area is performed. A comparison of SIMS data with EDX data provides the necessary information to improve existing models of SIMS and obtaining a real quantification of the SIMS data.

The following examples are given to illustrate the present invention without intending to limit their scope. Example 1. Chemical analysis of a certified Ti alloy by the method of the present invention.

Stage I.

We performed an analysis of the main elements of the material to analyze (concentration between 0.1 % to 100%) by EDX technique, directly performing the quantification without using patterns (standardless).

Stage II.

An analysis of all elements of the material was performed by the SIMS technique, accomplishing the quantification in two stages as follows:

a) Having analyzed the main elements with EDX, we searched for the sensitivity factors (RSF) to SIMS of the analyzed elements (tabulated values), which were then plotted as a function of the ionization potential (Ip) for positive ions, and as electron affinity function for negative ions. The dependence RSF - Ip (RSF - Ae) was plotted on logarithmic scales forming a straight line, which provides RSF values for other interest elements; and

b) With the RSF detected, we quantified other elements, which were observed in the mass spectrum of SIMS.

Figure 5 shows the EDX spectrum resulting from the Ti alloy with the identified elements.

According to figure 5, there is a signal of Al that is relatively high because of the surface polish of the study sample with a paste of corundum (Al₂0₃), and two non-marked signals in the spectrum corresponding to W and Ni, materials commonly found on this kind of samples.

Table 1 shows the concentration of the main elements found in the Ti alloy analyzed by EDX.

Table 1

'Concentration is overestimated by corundum of Al used to polish the surface.

Figure 6 shows the sensitivity factors found with EDX as a function of the ionization potential of the elements. Regarding the density of the Ti alloy, it was estimated as the equal density of pure Ti (5.66e22 atoms/cm³). All calculations presented here, and those presented below were performed in % atomic. Figures 7 to 10 show the fragments of SIMS mass spectrum of the elements detected in the Ti alloy.

As can be seen, not mentioned elements in the certificate of figure 14, such as B, Na, K, Ca, and In, were identified in the mass spectrum. The spectrum shows no Hydrogen, but it was possible to detect it; nor the mass range between 1 u.m.a. to 10 u.m.a., and between 160 u.m.a. to 180 u.m.a. is shown because there were no signals in this range, except for H.

The depth profiles of the elements detected in the Ti alloy were obtained by SIMS (figures 1 1 and 12), obtaining their concentration (atoms/cm³) (see table 2) by the sensitivity factors previously found. Here, Oxygen was measured as isotope ¹⁸0, because of the material erosion by ¹⁶0⁺ oxygen ions.

Table 2 (% atomic)

Failure

Element Certificate EDX SIMS

(%)

H 0.072% n/d 0.07% <3%

B n/d n/d 3E-4% n/d

C 0.044% n/d 0.050% <15%

N 0.028% n/d 0.034% <20%

O 0.21 % n/d 0.40% <75%

Na n/d n/d 4E-4% n/d

Al^** 2.72% 9.2%^* 2.72%^* n/d

Si^** 0.12% 0.1 % 0.1 % <15%

K n/d n/d 3E-6% n/d

Ca n/d n/d 6E-5% n/d

Ti^** 87.4% 83.2%^* 87.4% n/d

4.2% 4.5 4.5% <7%

Cr n/d 0.027% n/d

Sum: 0.03%

Mn n/d 0.01 1 % n/d Table 2 (Cont.)

* Failure by surface contamination that occurs by polished with corundum.

** Detected elements with EDX, and which were used for SIMS calibration.

n/d Not determined. As shown, by the method of the present invention it is possible to perform a quantitative chemical analysis without using patterns in the entire range of concentrations, for example from 100% up to 1 E-6% atomic. Also, measurements performed with EDX provide a quantification of the main elements (0.1 -100% atomic) with an accuracy of ±10% relatives, while the low concentration elements detected in the material are analyzed with SIMS (after data calibration of EDX) with an accuracy of factor 2 (or ± 50% relatives) (see figure 13).

Example 2. SiGe thin film (500nm) analysis grown on silicon using the method of the invention.

A sample of SiGe thin film with the characteristics shown in figure 15 was analyzed by the method of the invention.

Figures 16, 17, and 18 show the EDX spectrum obtained for the analyzed SiGe material, obtained by primary electron energy of 10 keV, 5 keV, and 3 keV. Table 3 shows the composition of the SiGe film defined by EDX with different electron energies.

Table 3

Element 10 keV 5 keV 3 keV Reference

Si 51 .77 48.77 36.29 35.1

Ge 48.23 51.23 63.71 64.9 As shown, with the EDX system installed on the SIMS computer according to the present invention and an electron energy of 3 keV it is possible to perform a truly quantitative analysis of a SiGe thin film of 0.5 micron without using patterns, with a failure of < 2%.

Regarding the depth profile obtained by SIMS (ion intensity as a function of erosion time) for a SiGe thin film sample from N183 sample (see figures 19 and 20), only the main elements are shown, which were measured on the first 100nm of the SiGe film. The result corrections obtaining by SIMS for the SiGe thin film (sample marked as N183) were performed by calculating the SIMS concentrations in standard mode using the following sensitivity factor (RSF), which were obtained for the implanted patterns:

For Si+: 5e22, Ge+:1 .5e23 (relative to Si), Si+: 1 .2e22 y Ge+:4.4e22 (relative to Ge). For Si-: 5e22, Ge-:1.5e23 (relative to Si), Si-: 1 .2e22 y Ge-:4.4e22 (relative to Ge).

Table 4 shows the results of SIMS with reference to Si implanted in matrix Ge and to Ge implanted in matrix Si, performing calculations for two different regimes: measuring positive and negative ions. The third method (SEM/EDX, AES) provides the correct composition of the film and is shown as reference.

As shown, the SIMS measurement technique of the main elements in the SiGe film provides results with an 18-26% failure for positive ions and 31 -37% for negative ions, if the calculations are made with sensitivity factors obtained with implanted patterns¹. Also, because the SIMS technique is not quantitative, the quantification of experimental data from the beginning without patterns is impossible².

Table 4

Figure 21 shows the "final" depth profile for the analyzed material by the method of the invention, showing the concentration of the relevant elements as function of depth in the material. The concentration of the main elements (Si, Ge) was calculated with EDX analysis data, while the doping elements and contaminants (H, C, N, O) were recalculated as standard SIMS implanted patterns.

In view of the above, the present invention provides through the SIMS technique a quantification of the main elements of the material, and for the other existing elements, even in a low concentration, with a failure of ±20%¹, while the EDX technique provides the quantification of said elements with a failure of less than 2% atomic.

In conclusion, the present invention provides a truly quantitative analysis of the elements that conform film materials where the low concentration elements (< 0.1 % atomic) are measured with SIMS with a precision of ±20% failure, while the main elements are measured with EDX with an accuracy of 2% failure, all simultaneously and without the use of standards and patterns.

References.

1 . R.G. Wilson, F.A. Stevie, C.W. Magee, Secondary Ion Mass Spectrometry, A Practical Handbook for Depth Profiling and Bulk Impurity Analysis, Wiley, 1989, p. 543.

2. Benninghoven, F. G. Rudenauer, H. W. Werner, in Secondary Ion Mass Spectrometry Basic Concepts, Instrumental Aspects, Applications and Trends, Wiley: New York, 1987, p. 310.

Claims

Claims.

1. A combined method of secondary ion mass spectroscopy (SIMS) and energy dispersed X- ray (EDX) for quantitative chemical analysis of solid materials and thin films, characterized because comprising the steps of:

a) Install a energy disperse detector Si(Li) on a free port of a mass spectrometer for recording X-ray;

b) Place the study sample in an ultra high vacuum chamber of the mass spectrometer; c) Activate the electron gun of the mass spectrometer to obtain the characteristic X-ray excitation of the study sample, wherein this electron gun provides a moderately focused electron beam with energies that vary from 0 to 15 keV and currents of 0 to 100 μΑ; d) Analyze the X-rays generated in step (c) with the energy dispersed detector Si(Li), and search the sensitivity factors (RSF) for the SIMS technique;

e) Calibrate the sensitivity factors (RSF) obtained from the qualitative analysis with EDX from stage (d) with the SIMS data using existing theories; and

f) Analyze with SIMS any unknown material of the study sample.

2. The combined method of secondary ion mass spectroscopy and energy dispersed X-ray energy of claim 1 , characterized because the mass spectrometer which is installed the energy dispersed detector, it is provided with two ion sources: a cesium (Cs⁺) ion gun and a duoplasmatron (0₂ ⁺, Ar⁺), and has an electron gun that is used for load balancing during the SIMS analysis of highly resistive or dielectric samples.

3. The combined method of secondary ion mass spectroscopy and energy dispersed X-ray energy of claim 1 , characterized because the mass spectrometer which is installed the energy dispersed detector can be of the type of magnetic sector, quadrupole, or time of flight.

4. The combined method of secondary ion mass spectroscopy and energy dispersed X-ray energy of claim 1 , characterized because the sensitivity factors (RSF) obtained from the EDX quantitative analysis of step (d) is plotted as an ionization potential function (Ip) for positive ions, and as electron affinity for negative ions of the analyzed elements so that the dependency RSF - Ip and RSF - Ae with semi-logarithmic scales form a straight line that gives RSF values for other elements of interest in the study sample.

5. The combined method of secondary ion mass spectroscopy and energy dispersed X-ray energy of claim 1 , characterized because during step (d) it is possible to perform the lateral distribution analysis by EDX or a EPMA method when the electron gun focuses at 1 micrometer, which corresponds to the lateral resolution during the SIMS analysis, which gives a quantitative information about the lateral distribution of the elements of interest on the study sample.

6. The combined method of secondary ion mass spectroscopy and energy dispersed X-ray energy of claim 1 , characterized because the EDX analysis can be performed for any element at a concentration of 0.01 to 100% atomic, from which the elements at a concentration between 0.01 to 1 % atomic will be used as SIMS calibration.

The combined method of secondary ion mass spectroscopy and energy dispersed X-ray energy of claim 1 , characterized because the calibration of SIMS of step e) using sensitivity factors (RSF) is accomplished through the following equation:

wherein, ηί and the exponential coefficients of this equation are found for this experimental regime, and after the calibration of the SIMS data for the corresponding concentrations below 1 % atomic can be accurately quantified due of the exponential dependence between SIY, the ionization potential, and the electron affinity of the elements of interest in the study sample.