SECONDARY ION MASS SPECTROSCOPIC METHOD, MASS SPECTROMETER AND USES THEREOF

Abstract

In a secondary ion mass spectroscopic (SIMS) method, and a mass spectrometer for implementing the method, for depth-profiling analysis of alkali metals in a sample which comprises an insulating material or is an insulator. The sample is irradiated by an ion beam as an analysis beam for desorption of secondary ions from the uppermost layers, such that the surface of the sample is removed with the same or a further ion beam. The ion beam used for removal of the sample surface comprises essentially gas clusters or consists of gas clusters.

Claims

1.-10. (canceled)

11. In a secondary ion mass spectroscopic (SIMS) method for depth-profiling analysis of alkali metals in a sample comprising an insulating material, the sample being irradiated by an ion beam as an analysis beam for desorption of secondary ions from the uppermost layers of the sample and the surface of the sample being removed with the same or a further ion beam, the improvement wherein the ion beam used for removal of the sample surface essentially comprises or consists of gas clusters.

12. The method according to claim 11, wherein the ion beam used for removal of the sample surface essentially comprises or consists of gas clusters of oxygen molecules or oxygen-containing molecules or both oxygen molecules and oxygen-containing molecules.

13. The method according to claim 12, wherein the ion beam used for removal of the sample surface comprises or consists of 40% of clusters of oxygen molecules or oxygen-containing molecules or both oxygen molecules and oxygen-containing molecules.

14. The method according to claim 12, wherein the ion beam used for removal of the sample surface comprises or consists of 60% of clusters of oxygen molecules or oxygen-containing molecules or both oxygen molecules and oxygen-containing molecules.

15. The method according to claim 12, wherein the ion beam used for removal of the sample surface comprises or consists of 80% of clusters of oxygen molecules or oxygen-containing molecules or both oxygen molecules and oxygen-containing molecules.

16. The method according to claim 12, wherein the ion beam used for removal of the sample surface comprises or consists of 90% of clusters of oxygen molecules or oxygen-containing molecules or both oxygen molecules and oxygen-containing molecules.

17. The method according to claim 11, wherein the gas clusters are 80% gas clusters with 100 to 5,000 oxygen molecules or oxygen-containing molecules per cluster molecule or both oxygen molecules and oxygen-containing molecules.

18. The method according to claim 17, wherein the gas clusters are 80% gas clusters with 500 to 2,000 oxygen molecules or oxygen-containing molecules per cluster molecule.

19. The method according to claim 11, wherein the gas clusters of the ion beam used for removal of the sample surface impinge with an energy of between 3 keV and 50 keV, respectively including or excluding the range limits, on the surface of the sample.

20. The method according to claim 19, wherein the gas clusters of the ion beam used for removal of the sample surface impinge with an energy of between 5 keV and 25 keV, respectively including or excluding the range limits, on the surface of the sample.

21. The method according to claim 19, wherein the gas clusters of the ion beam used for removal of the sample surface impinge with an energy of between 10 keV and 20 keV, respectively including or excluding the range limits, on the surface of the sample.

22. The method according to claim 11, wherein said SIMS method is a time of flight (ToF) SIMS method.

23. A mass spectrometer for carrying out the method according to claim 11, comprising an ion source for producing the gas cluster ion beam used for removal of the sample surface.

24. A mass spectrometer according to claim 23, wherein the sample comprises an insulating material.

25. The method according to claim 11, comprising the step of examining the depth profile of alkali metals in a sample which comprises or consists of an insulating material.

26. The method according to claim 25, wherein the sample comprises alkali metals.

27. The method according to claim 26, wherein the sample comprises alkali metals 1 ppm.

28. The method according to claim 11, wherein the sample comprises alkali metals 10.sup.16*1/cm.sup.3.

29. The method according to claim 11, wherein the sample comprises alkali metals 10.sup.19*1/cm.sup.3.

30. The method according to claim 11, wherein the sample additionally comprises SiO.

31. The method according to claim 11, wherein the sample comprises a glass.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG. 1 shows a measurement of the Na.sup.+ secondary ion intensity as a function of the depth by means of ToF-SIMS on an approx. 1 m thick SiO.sub.2 layer on a silicon substrate with a removal with O.sub.2, Cs and O.sub.2 gas clusters.

[0025] FIG. 2 shows a measurement of the Na concentration as a function of the depth by means of ToF-SIMiS on an approx. 1 m thick SiO.sub.2 layer on a silicon substrate with a removal with O.sub.2, Cs and O.sub.2 gas clusters.

[0026] FIG. 3 shows the cluster size distribution in an oxygen gas cluster ion beam.

[0027] FIG. 4 shows a measurement of the Na.sup.+ secondary ion intensity as a function of the depth by means of ToF-SIMS on an approx. 1 m thick SiO.sub.2 layer on a silicon substrate for oxygen- and argon gas cluster removal.

[0028] FIG. 5 shows measurements of the Na concentration profile for differently treated SiO.sub.2 samples.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] The preferred embodiments of the present invention will now be described with reference to FIGS. 1-5 of the drawings. Identical elements in the various figures are designated with the same reference numerals.

[0030] FIG. 1 shows a depth profile recorded by a ToF-SIMS device. The Na.sup.+ secondary ion intensity as a function of the removed depth is illustrated. It concerns a logarithmic intensity scale in which the number of recorded secondary ions per data point is indicated. As sample, an approx. 1 m thick SiO.sub.2 layer which is applied on a silicon substrate was used. In this 1 m thick SiO.sub.2 layer, sodium with an energy of 120 keV with a surface dose of 5.Math.10.sup.14 ions/cm.sup.2 had been implanted. The depth distribution of such an implantation profile is basically known. The present measurement serves for demonstrating the excellent properties of oxygen gas cluster ions as sputter ions relative to a conventional depth profiling by means of oxygen ions O.sub.2.sup.+ or caesium ions Cs.sup.+.

[0031] In a first measurement, the sample was analysed in a ToF-SIMS mass spectrometer ToF.SIMS 5 of ION-TOF GmbH with a dual-beam method using oxygen ions O.sub.2.sup.+. A bismuth liquid metal ion source with the primary ion species Bi.sub.1.sup.+, an energy of 30 keV and an analysis surface of 100100 m.sup.2 served as pulsed analysis beam. The analysis current was 4.6 pA at a pulse frequency of 10 kHz. The measuring time per data point was 1.6 s. The removal was effected on a surface of 300300 m.sup.2 with an O.sub.2.sup.+ ion beam with an energy of 2 keV and an ion current of 255 nA. As is common in the state of the art, a low energy electron source with an electron energy of 20 eV was used for compensation of the surface charge by the oxygen ion beam. The associated measurement is illustrated in FIG. 1 as a dotted curve.

[0032] As was to be expected, the measured depth profile (dotted curve) is extremely distorted by electromigration. Shortly after the beginning of the depth profile, the sodium intensity falls to extremely low values. Only after some time of the depth profiling, i.e. apparent at a greater depth, does the intensity then increase again slightly. At the interface between the SiO.sub.2 layer and the silicon substrate, the signal then increases extremely sharply and then falls again rapidly in the silicon to very low values. Therefore a typical signal progression for electromigration is produced since the sodium atoms react to the surface charge and move away from the surface. Only when reaching the interface does the electrical conductivity of the sample increase again and the surface charge and hence also the electromigration disappears. The sodium signal hence increases again greatly and the result is an excess of sodium signals because of accumulation of migrated sodium ions in this zone of the sample. The depth distribution was therefore very greatly falsified by this measurement.

[0033] In the case of a second measurement which is illustrated by a broken-lined curve in FIG. 1, a caesium ion beam was used, with identical analysis conditions, for the sputtering of the sample surface between the individual analyses. The energy of the ion beam was 2 keV with a beam current of 145 nA. As was to be expected, the electromigration is significantly reduced. The sodium intensity profile corresponds approximately to the expected depth distribution which occurs on insulators, such as SiO.sub.2, with implanted sodium ions. Nevertheless slight electromigration takes place and leads to a small excess of the signal at the interface to the silicon substrate. However, a serious disadvantage of this method is the low sensitivity of the measurement. By introducing caesium, the positive secondary ion yields which are essential for sodium are significantly reduced. At the maximum of the distribution, the intensity is below 2,000 ions/data point.

[0034] The continuous line in FIG. 1 was measured with a method according to the invention and a mass spectrometer equipped according to the invention. For the removal of surface layers, large oxygen gas clusters with essentially 1,400 molecules/ion and an energy of 20 keV were used for depth profiling with the same analysis parameters. The oxygen gas clusters were produced by a nozzle with an inlet pressure of approx. 35 bar. The gas cluster beam propagating in the vacuum was ionised with an electron beam and subsequently accelerated to 20 keV beam energy. The typical size distribution of the oxygen gas cluster ions is illustrated in FIG. 3. The size distribution is very wide and begins at approx. 100 O.sub.2 molecules and reaches up to approx. 3,800 O.sub.2 molecules. The maximum is at 1,400 O.sub.2 molecules per cluster ion. For the removal, the gas cluster beam was scanned over a surface of 300300 m.sup.2 with a beam current of 4.9 nA.

[0035] The continuous line in FIG. 1 corresponds precisely to the implantation profile which is to be expected and known. In particular on the curve progression at the interface between the SiO.sub.2 insulator layer and the silicon substrate at a depth of 1,000 nm, the extremely low electromigration occurring with this measuring method is shown. The intensity at this interface is below the highest intensity in the depth profile by approx. three orders of magnitude.

[0036] It should be assumed as operating mechanism that, by the use of the oxygen gas clusters as sputter ions, the surface of the sample is very well oxidised, which leads to a significantly increased positive secondary ion yield for sodium during desorption by the Bi analysis beam. Consequently, very low detection limits for sodium in the insulating material result therefrom. The minimum concentration which can be detected is below the peak concentration of the implantation profile by more than four orders of magnitude. The special interaction between cluster projectile and surface in conjunction with a higher sputter yield lead to almost complete elimination of the electromigration of alkali metals in the insulators.

[0037] FIG. 2 shows the measuring results of FIG. 1 with the three different sputter ion beams after a calibration to the implantation dose on a logarithmic concentration scale in atoms/cm.sup.3. Whilst the peak concentration in the measurement with oxygen gas clusters is determined at 3.Math.10.sup.19 atoms/cm.sup.3 (continuous line), the electromigration with the O.sub.2 sputter beam (dotted line) leads to completely false results. At the maximum of the implantation profile, a concentration reduced by more than 2 orders of magnitude is measured and, at the interface to the silicon substrate, the concentration increases to approx. 10.sup.20 atoms/cm.sup.3. The concentration profile under caesium sputter removal (broken-line) shows a merely slightly lower peak concentration in comparison with the oxygen gas cluster ions. However, as a result of the electromigration, the concentration at the interface is still above 2.Math.10.sup.19 atoms/cm.sup.3. Reduction in the ion yield leads to a poorer detection limit of approx. 2.Math.10.sup.16 atoms/cm.sup.3, whilst, with O.sub.2 gas cluster ions, detection limits below 2.Math.10.sup.15 atoms/cm.sup.3 are achieved.

[0038] As a measure for the strength of the electromigration, the proportion of the Na dose in the SiO.sub.2/Si interface can be determined. The proportion of the dose in the hatched region in FIG. 2 was determined for all three removal conditions:

TABLE-US-00001 a. O.sub.2 sputter ion beam (dotted curve) .sup.94% b. Cs sputter ion beam (broken-line curve) 9.2% c. O.sub.2 gas cluster ion beam (continuous curve) 0.01%

[0039] It becomes very clear herefrom that the electromigration under oxygen gas cluster removal is practically negligible and the concentration profile can be determined quantitatively.

[0040] In FIG. 4, the depth profile with removal by oxygen gas clusters (see FIG. 1) is compared with the removal by an argon gas cluster beam. The analysis- and sputtering parameters are the same for both measurements. FIG. 4 shows the depth profile for the silicon matrix signal Si.sup.+ and also for Na.sup.+ for both different gas cluster beams. The oxygen gas cluster beam obtains complete oxidation of the SiO.sub.2 during removal and also leads to complete oxidation of the silicon substrate. This is shown on the Si.sup.+ profile under oxygen gas cluster bombardment (continuous line). Under argon gas cluster bombardment (broken-line), the oxidation is slightly reduced and the signal intensity of the Si.sup.+ is slightly lowered. In the silicon substrate, the signal then breaks down greatly since, because of the lack of oxygen, the positive ion yield is reduced by several orders of magnitude. The Na.sup.+ profile under both bombardment conditions is very similar. The detection limits under oxygen gas cluster removal are lower approximately by a factor 2.5. This example shows that the alkali concentrations in the SiO.sub.2 can be determined correctly with both types of gas cluster beams. The concentration profile in the silicon substrate can however be measured only with the oxygen gas cluster beam since the latter ensures complete oxidation and hence constant ionisation.

[0041] FIG. 5 shows further depth profiles which were measured on different SiO.sub.2 layers. In microelectronics, the mobile ions, such as e.g. sodium and potassium, play a very important role since they rapidly diffuse as impurities in components due to the applied voltages in the dielectric layers and as a result the properties of a component can be considerably impaired. Therefore for optimisation of the reliability and lifespan of a component, a sensitive and quantitative detection method for alkali metals is required. The SiO.sub.2 insulator layers in the test samples used have a thickness of 200 nm and are applied respectively on a silicon substrate. In the case of these samples, sodium was introduced specifically by electromigration into the SiO.sub.2 insulator layer. The samples differ by the length and the direction of the migration process. Thereafter, the samples were coated with gold. The thickness of the coating is approx. 40 nm. The curves illustrated in FIG. 5 show the sodium distribution in the individual samples. The measurements were effected in the dual-beamdual-beam method. A bismuth-liquid metal ion source with the primary ion species Bi.sub.1.sup.+, an energy of 30 keV and analysis surface of 100100 m.sup.2 served as pulsed analysis beam. For removal of the surface layers for depth profiling, large oxygen gas clusters with essentially 1,400 molecules/ion and an energy of 20 keV were used according to the invention. The removal was effected on a surface of 300300 m.sup.2.

[0042] The dash/dot curve shows the distribution of the sodium ions after introduction of sodium into the SiO.sub.2 layer was approx. half-finished. The dotted curve represents the sodium distribution after complete conclusion of the migration. The broken-line curve shows distribution after half of the migration in the reverse direction. The continuous curve shows the sodium distribution after the sodium had been removed again from the layer. The results show that the method according to the invention enables quantitative analysis of these samples.

[0043] There has thus been shown and described a novel secondary ion mass spectroscopic method, mass spectrometer and uses thereof which fulfills all the objects and advantages sought therefor. Many changes, modifications, variations and other uses and applications of the subject invention will, however, become apparent to those skilled in the art after considering this specification and the accompanying drawings which disclose the preferred embodiments thereof. All such changes, modifications, variations and other uses and applications which do not depart from the spirit and scope of the invention are deemed to be covered by the invention, which is to be limited only by the claims which follow.