High brightness point ion sources using liquid ionic compounds

Details High brightness point ion sources using liquid ionic compounds High brightness point ion sources using liquid ionic compounds

1995-10-10

2002-01-08

Shingleton, Michael B. (Department: 2817)

Electric lamp and discharge devices: systems

Discharge device load with fluent material supply to the...

Plasma generating

C315S111810, C250S492210, C219S121480, C204S298410

Reexamination Certificate

active

06337540

ABSTRACT:

FIELD OF INVENTION
The present invention relates to focused ion beam (FIB) technology and more particularly to novel high brightness point ion sources using ionic compounds in the liquid state, including mixtures of molten salts, acids or bases. The novel high brightness point ion sources associate the desirable high brightness, ion intensities and energy distribution characteristics of conventional liquid metal ion sources (LMIS) with the possibility of significantly enlarging their spectrum of ion species. In particular, chemically reactive ions are now produced including positive and negative ions, molecular ions and even protons, which has not been made possible so far with said conventional LMIS sources. According to the present invention, any conventional focused ion beam system can be easily adapted to use such novel high brightness point ion sources.
BACKGROUND OF THE INVENTION
Till today, almost all industrial point ion source applications are based upon liquid metal ion sources (LMIS), which gave rise in the 1970-1980 period to the spectacular explosion of focused ion beam (FIB) technology. In comparison to all other types of ion sources, LMIS sources exhibit quite excellent optical qualities (brightness and low energy spread) that allow the FIB systems incorporating the same to focus ionic spots of sub-micronic sizes with high current densities in the order of several A/cm
2
that are well adapted to industrial applications. In particular, FIB systems are extensively used in the micro-electronic field, for the modification, reconfiguration, failure analysis and manufacturing of advanced semiconductor products, typically integrated circuits (ICs) and the surface analysis thereof. However, if LMIS sources are at the historical origin of the FIB technology expansion, they have inherent physical constraints which considerably limit the potential applications of FIB systems and scientific instruments derived therefrom. With LMIS sources, the ionizable source materials that can be used for ion generation are limited to a few pure metals and to some metal alloys. For reasons of reliability and of optical qualities, the most commonly used metal is gallium. Despite some recognized advantages, the use of LMIS sources in standard FIB systems exhibits some considerable limitations and inconveniences that are recited below.
First of all, there is no chemical reactivity effect associated with the collision process of sputtering (ionic bombardment) unlike in a conventional reactive ion etching process for example. This limits the removal rate of most of the most commonly used target materials to 1-4 sputtered atoms per incident ion at energies of about 30 kV. Because there is no by-product gas formation, the sputtered atoms are not evacuated during the process and a re-deposition of the sputtered materials near the attacked areas occurs. This undesired re-deposition is complicating many FIB based etching processes. In particular, if the redeposited material is of the conductive type it can create parasitical connections on the semiconductor product. Moreover, re-deposition of the sputtered material reduces the achievable sidewall angle, and thus results in aspect ratios of etched holes no greater than approximately 6:1. Etching a deep trench thus requires a lengthy process and can produce undesired damages to the hole neighboring areas.
Another consequence of this lack of chemical reactivity can be found in an analysis technique currently referred to as the SIMS (SIMS is an acronym for Secondary Ion Mass Spectrometry). This technique can be based on the use of gallium probes and has a recognized high local resolution (an important requirement of focused ion beam applications), so that such SIMS systems reach a local resolution of some tens of nanometers. However, one of the conditions for quantitative SIMS analysis is to have a high secondary ion emission yield. Because gallium doesn't have any important effect of secondary ion yield stimulation, but has a high local resolution, the benefit of small sized gallium probes is balanced by the small emission rate of the secondary ions to be analyzed. On the contrary, commercially available SIMS systems using other ions which associate a chemical effect in addition to the collision effect of sputtering in order to enhance the secondary ion yield, unfortunately have a poor local resolution. In particular, cesium (Cs), which is one of the most chemically reactive metals, considerably enhances the secondary ion yield. Industrial SIMS systems may employ classical Cs sources (generally of the surface ionization type), but the brightness of these ion sources is low and thus cannot be compared with those of Ga LMIS sources. Moreover, the violent reactivity of cesium also makes its handling very difficult. Many attempts to produce Cs LMIS sources have been conducted in research laboratories, but because of the high chemical reactivity of cesium, Cs LMIS sources have never reached the acceptable reliability level that is required by SIMS systems used in the industry.
Finally, for example in applications such as quantum device fabrication, it is difficult to reduce the creation of ion beam-induced defects on the sample by a simple reduction of the ion beam energy, because this operation involves a very important loss in terms of current density, the machining time is significantly increased, and finally results in a costly process.
An attempt to solve the problem of the violent chemical reactivity of alkaline metals has been described in the article: Lithium ion emission from a liquid metal source of LiNO3, by A. E. Bell et al, published in the International Journal of Mass Spectrometry and Ion Processes, 88 (1989), pp 59-68. These authors conducted experiments with ion sources similar to LMIS sources. They substituted the alkaline metals with chemical compounds containing these metals. In particular, they produced a Li+ ion beam with a source using a pure molten binary salt, in this case the lithium nitrate (LiNO
3
). For these experiments they used a needle coated with this molten salt which was heated as standard and they called their ion sources “liquid metal ion sources of LINO
3
”, i.e. a variant of LMIS sources. They observed the generation of gas bubbles and very fast evaporation of the said molten salt. Measuring the energy distribution of the emitted ions with a retarding potential analyzer, they found a FWHM (full width at half maximum) energy spread of 110 eV which was comprised of two peaks. They concluded that this was due to the participation of gas-phase field ionization in addition to field desorption on the apex of the needle tip. Ion sources with such a large spread in energy distribution are unsuitable for use in industrial FIB systems because the focusing of the ion beam is roughly limited by chromatic aberrations.
On the other hand, in some advanced fields of micro-analysis, such as proton microscopy, ion beam lithography (no proximity effects for small structures and very high resist sensitivity to ions), localized Rutherford back-scattering analysis (RBS) and particle induced X-ray emission analysis (PIXE) with very high spatial resolution, the generation of protons (the known lightest ion) is quite impossible to obtain with FIB systems using conventional LMIS sources. As a matter of fact, if many attempts have been done to realize focused proton beam columns, these efforts have never ended in successful industrial applications. They often used field ionization sources in gas phase. In this particular case, the ion source generally consists of a tungsten needle that is cooled with liquid helium at a temperature of a few Kelvin degrees. A flux of hydrogen atoms is generated by field-ionization at the needle tip. These ion sources have a great theoretical brightness but imply sophisticated and thus expensive equipments. In addition, they have a very poor reliability and a low total ion current.
Consequently, it would be extremely interesting, especially for SIMS analysis of integrated circuits (ICs), but also for

Affiliated with

Also associated with

Rating

High brightness point ion sources using liquid ionic compounds does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with High brightness point ion sources using liquid ionic compounds, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and High brightness point ion sources using liquid ionic compounds will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2847154