GB2258083A

GB2258083A - Sample analysis apparatus and method.

Info

Publication number: GB2258083A
Application number: GB9116038A
Authority: GB
Inventors: Mark Graeme Dowsett; Stephen Paul Thompson; John Laurence Wilkes
Original assignee: Kratos Analytical Ltd
Current assignee: Kratos Analytical Ltd
Priority date: 1991-07-25
Filing date: 1991-07-25
Publication date: 1993-01-27
Also published as: JPH05209849A; GB9116038D0

Abstract

An apparatus for analysing the surface of a sample (1) by means of irradiation of the sample by a primary beam includes means for displaying a visual image of the surface of the sample, e.g. on the VDU of computer 15, together with means (3, 5) for varying the position of the beam on the sample (1) such that material is ejected from a series of different points over the sample dependent on the image. The dwell time of the beam on each pixel of the sample may also be varied. In an arrangement described for mass analysing regions of a semiconductor sample, secondaries from a primary beam incident on the sample are used to form an image of the sample on a computer VDU screen and a mouse is used to define areas of the image it is desired to analyse. The results of the analysis from corresponding areas of a number of identical transistors, for example, may be summed to obtain a more precise depth profile of the dopant in the region of interest. <IMAGE>

Description

SAMPLE ANALYSIS APPARATUS AND METHOD This invention relates to sample analysis apparatus and methods. The invention has particular, although not exclusive relevance to sample analysis apparatus and methods for analysing the surface of a sample.

Modern materials technology has created the need to analyse inhomogeneous and anisotropic patterned samples. Examples of such samples are fibre reinforced plastics, polycrystalline materials such as high temperature superconductors and metallic alloys, semiconductor and other electronic devices, and layered composites and coated materials. Additionally some naturally occurring samples such as ores, and animal or vegetable sections may also be inhomogeneous, and/or anisotropically patterned.

The need arises in such samples to determine both the identity, and the position in a sample of a chemical species, with a high spatial resolution, typically from 1 mm to 100 microns. The need may arise, for example, to find the correct area of a patterned sample to analyse, the analysis being confined to this area which may be an arbitrary shape whichis not known in advance.

Furthermore, in the case of patterned samples, micromachining may be required to expose the region(s) of interest, the shape of the volume which needs removal also being arbitrary and not known in advance.

Alternatively, micro-machining may be used to repair a damaged region of a sample. In either event it is necessary to determine the region within the sample in which micromachining is taking, or has taken place.

Where mass spectrometric techniques are used to analyse a sample, such techniques consume material from the sample. Thus, in the case of, for example, semiconductor devices in which there is a very small proportion of a particular dopant within selected regions of the sample, these regions being repeated over the surface of the sample, it would be useful to excite materials of interest from the sample over a number of these regions.

Known sample analysis apparatus, for example secondary ion mass spectrometry (SIMS) apparatus, are generally only capable of analysing a single rectangular or circular analysis zone of a sample.

H. Frenzel et al, in an article in SIMS, VI, 1988, pages 219 to 224, describes a SIMS instrument in which it is possible to analysise material from different parts of an analysis zone by means of a "chequer board" gating.

Such an instrument, however, only allows material to be analysed from restricted, pre-defined shapes within the sample, for example an array of contiguous squares or rectangles.

It is an object of the present invention to provide a sample analysis apparatus and method enabling the spatial identification of chemical species or other features of the sample over the sample.

According to a first aspect of the present invention, there is provided a sample analysis apparatus, comprising excitation means for producing a beam effective to eject material from a point on a sample, means for forming an image of the sample, and means for controlling the position of intersection of the beam and the sample such that material is ejected from a series of different points over the sample dependent on the image of the sample.

According to a second aspect of the present invention, there is provided a sample analysis method, the method comprising the steps of producing an excitation beam effective to eject material from a point on a sample, forming an image of the sample, and controlling the position of intersection of the beam and the sample such that material is ejected from a series of different points over the sample dependent on the image of the sample.

One embodiment of an apparatus in accordance with the invention, together with a number of embodiments of methods of use of the apparatus, will now be described, by way of example only, with reference to the accompanying figures in which: Figure 1 is a block schematic diagram of the embodiment of the apparatus; Figure 2 illustrates the manner in which pixels of the image of the sample may be addressed; Figure 3 illustrates the pixel order in a raster scanning arrangement; Figure 4 illustrates a pixel pattern enclosed within a domain; Figure 5 illustrates two further domains; Figure 6 illustrates a number of contiguous and non-contiguous sub-rasters within an image of the sample; Figure 7 illustrates an outer raster nested in an inner raster; Figure 8 illustrates an inner raster nested in a random outer raster; Figure 9 illustrates the effect of analogue gain on domain size;; Figure 10 illustrates a first method for milling a surface profile of a sample in accordance with an embodiment of the invention, and Figure 11 illustrates a second method for milling a surface profile on a sample in accordance with an embodiment of the invention.

Referring firstly to Figure 1, the apparatus includes a charged particle source, (not shown), effective to direct a beam of charged particles on to a sample 1 mounted on a motorised sample stage (not shown). The beam is directed through an X-scan unit 3 and a Y-scan unit 5, which are arranged to control the position of intersection of the charged particle beam with the sample 1 for a chosen dwell time, as will be described in detail hereafter. The X and Y scan units 3,5 may take the form of magnetic or electric charged particle optics or any combination of the two, effective to apply a time varying electric or magnetic field to the charged particle beam.

The beam scanning operation is controlled by the loading of sets of three numbers, known as '3-vectors' into three random access memories 7,9,11. Each set of three number defines the position of intersection of the beam with the sample 1 as a function of two orthogonal directions, X and Y, and the dwell time of the beam on each point of the sample. The three random access memories 7,9,11 may be integral with the electronics for the X and Y scan units 3,5, but in the particular embodiment shown, are connected to a communicating bus 13 which is, in turn, connected to a computer system 15, the bus 13 optionally having connected to it an auxilliary memory 16.

The numbers loaded into the three random access memories 7,9,11 are fetched sequentially by an appropriate electronic means (not shown), and fed into respective digital to analogue converters 17,19 for the X and Y scanning units 3,5, a timing system 21 being effective to control the dwell time. The outputs from the digital to analogue converters 17,19, are connected by respective amplifiers 23,25 to the X and Y scan units 3,5, so as to supply appropriate voltages to the scan units 3,5. The gain of the amplifiers 23,25 are controlled by respective gain units 27,29 which determine the absolute spacing between successive points of intersection of the beam with the sample 1, i.e. the pixel spacing, together with the maximum displacement of the beam of the over sample 1, i.e. the domain size.

Alignment of the charged particle beam with the sample 1 is performed by the loading of further numbers of appropriate magnitude into respective X and Y digital to analogue converters 31,33 through X and Y alignment buffers 35,37 under the control of the computer 15 through the bus 13.

The apparatus also includes a suitable signal detector means comprising a pulse counter 39, and an analogue detector 41, the signal detector means being effective to measure the signal corresponding to the charged particles which have been emitted from the sample 1 due to the impinging of the primary particle beam on the sample. A total ion or electron detector 43 is arranged to integrate the measured signal, so as to accummulate the signal from a particular analysis zone of the sample 1. Each analysis zone is determined by an analysis zone random access memory 45, together with an analysis zone pointer random access memory 47 under the control of a pixel counter 49 connected to the pixel dwell timer 21.

The apparatus also includes an optical imaging arrangement 51 effective to acquire a light optical image of the sample 1.

Thus, in use of the apparatus the position of the beam on the sample 1 is controlled by selecting a point, or series of points, defining a selected domain on an optical image of the sample 1 displayed on a visual display unit connected to the computer 15, using a computer linked pointing device, e.g. a mouse.

Feed-back to the motorised sample stage via a stage control unit 53, is used to place each selected point of the image of the sample in the middle of the field of view of the apparatus. Alternatively, a pattern or image may be loaded into an image store 55, associated with the computer 15 and a correlation between the pattern or image, and an image acquired from the sample 1, may be sought with the sample stage translating the sample in a search pattern until the correlation is found.

When the dwell time for any particular pixel has elapsed, the next set of 3-vectors is fed into the 7,9 and 11 and the cycle is repeated until the desired scanning operation has been completed.

Referring now also to Figures 2 to 11, Figure 2 illustrates the manner in which the apparatus may be used to displace the particle beam over the sample 1 by any distance up to a predetermined maximum in any direction in a plane orthogonal to the optical axis of the beam transport system. When the dwell time for a particular pixel has elapsed, the next set of 3-vectors is fed into the three random access memories 7,9,11.

Displacement of the beam over the sample 1 may be performed in an ordered sequence as illustrated in Figure 3, which illustrates the pixel order in a ten by ten line flyback raster. Generally, the beam scanner will incorporate means (not shown) for positioning the beam in an area outside the domain to be interrogated, this generally being off the sample, so as to produce blanking periods. Alternatively, the beam may be switched off for a controlled period of time at the end of each scanning period so as to produce blanking periods.

An example of a pixel pattern within a complex domain is shown in Figure 4. Sequential displacements of the beam over the sample, the displacements being of uniformly or non-uniformly sized steps in any direction, will form a pixel pattern which may be referred to as an "inner raster." Referring now also to Figure 5, this figure shows two examples of domains with interior areas excluded. Such domains may find application in, for example, the analysis of semiconductor devices. Thus, for example, an image of an interconnecting track on a semiconductor device may be enclosed in a polygon shaped domain with the rgions of the device not carrying the track excluded. The pixels within the domain only may be scanned by an ion beam for the purpose of removing all, or part, of the track without substantially affecting the surrounding regions of the device.

Another use of domains incorporating excluded interior areas, is where an intensity threshold is set for an image of the sample, such that all pixels containing an intensity above that threshold are incorporated in the domain. For example, if the interconnecting track on a semi-conductor device is made of aluminium, and an image is taken in a contrast mode differentiating aluminium from surrounding regions, then all the pixels on the track would be included in the domain. Alternatively, an intensity threshold may be set for the image such that all pixels containing an intensity below the threshold are incorporated in the domain. Thus, in the case of a semiconductor device having an interconnecting track made of aluminium, if an image is taken in a contrast mode differentiating aluminium from the surrounding regions, then all pixels off the track will be included in the domain.

A transform image may be formed from any mathematical transform of the image of the sample using the same, or different, contrast mechanisms as described above, the transform image having superimposed upon it a polygonal area so as to define the pixel set within the domain.

Alternatively, an intensity threshold may be set so that pixels above or below the intensity are included in the domain. If, in the example of a semiconductor having an aluminium track, the track is blurred or noisy, then a mathematical noise reduction technique may be applied prior to the domain selection process.

The 3-vectors corresponding to any desired inner raster may be calculated by the computer 15 or the auxiliary memory 16. As illustrated in Figure 6, these inner rasters may take the form of any number of contiguous and/or non-contiguous and/or overlapping regions, these regions being termed "subrasters". Referring now also to Figure 7, the beam alignment may be reprogrammed by means of the computer 1 or memory 16 through the X and Y alignment buffers 35,37, so that the inner raster is itself scanned laterally over the sample within the limits of an "outer raster." If the dimensions of the inner raster are greater than the displacement of the inner raster, then the outer raster may be said to be "nested" in the inner raster.It may be seen from Figure 7, that by appropriate displacement of an inner raster in this manner, the pixel density in a domain may be temporarily increased, so as to produce a high resolution image.

Where the dimensions of the inner raster are less than the amplitude of the overall displacement of the inner raster, then the inner raster may be said to be nested in the outer raster. This situation can be seen in Figure 8, in which the top left hand sub-rasters have been displaced to seven positions within the domain.

The beam scanning units 3,5 may be arranged to move the domain in the X and Y directions by electronically floating the voltages used to generate the inner raster. Thus the domains and/or the inner rasters may be aligned with appropriate features in the sample 1.

Where the apparatus is a SIMS apparatus, for example, in which a high extraction field is used to collect secondary ions from the sample 1, such beam alignment may be used to preserve the registration between the inner raster and chosen features of the sample when the extraction field polarity is reversed in order to change from positive to negative ion analysis.

The overall size of the domains and hence the absolute pixel spacing, may be varied by controlling the gain of the X and Y amplifiers 23,25 examples of different doman sizes being illustrated in Figure 9.

It will be seen that the time for which the beam resides on any pixel of the sample 1 may be individually controlled. Thus, a milling operation as illustrated in Figures 10 and 11, may be performed on the sample by the beam so as to produce a surface profile, such as a bevel. In the operation illustrated in Figure 10, the bevel is achieved by increasing the dwell time of the beam on the sample for pixels towards the right hand edge of the sample. The same result may be achieved by the process illustrated in Figure 11 in which the pixel density is increased for pixels towards the right hand side of the sample.

It will be appreciated that where the sputter rate varies across the field of view, the pixel spacing, or the pixel dwell time, or both, may alternatively be controlled such that the eroded region remains flat.

The resolution of the apparatus described herebefore, by way of example, is such that where the inner raster contains a maximum of 65536 pixels, the position of each pixel may be determined with a precision of 1 part in 65536. The dwell time is suitably determined with an accuracy of 1 part in 16,000. The analysis zone random access memory 45 incorporated in the apparatus will generally include sufficient memory to store data from up to 256 analysis zones, with a 32-bit precision for each zone.

It will be appreciated that an image of the sample may be acquired by any suitable means. These include phase and intensity contrasts produced by a light source. A visually displayable map of the sample surface may also be produced by atomic number and topographic contrast, produced by secondary electrons induced by a primary focussed electron beam (SEM). A primary electron beam may alternatively be used to produce chemical or other contrast by means of ion emission from the sample. A primary focussed ion or fast atom beam may be used to produce atomic number and topographic contrast from electrons thus ejected from the sample, or chemical or other contrast from secondary ions ejected from the sample.

The images thus formed may then be used to select the analysis zone and the shape of the current density distribution across the area of the sample irradiated by the beam.

It will be seen that a particular use of an apparatus and method in accordance with the invention is where it is required that signals from a particular analysis zone of a sample be added to subsequent signals in real time. For example a semiconductor device consisting of an array of tranistors may contain a number of regions correspondingly doped, each individual region containing too little dopant for a precise analysis of its vertical distribution in a SIMS depth profile. If similarly doped regions are identifiable via the image of the sample displayed via the usual display unit of the computer 15, sets of individual polygons may be drawn round each doped region, and the signals from the pixels in the polygons be summed in a subsequent depth profile so as to give a single analysis zone. Thus the depth distribution of the dopant in the regions of interest may be acquired with a good statistical precision, independently of the dopant distribution in the surrounding regions.

Claims

1. A sample analysis apparatus comprising: excitation means for producing a beam effective to eject material from a point on a sample, image forming means for forming an image of the sample, and means for controlling the position of intersection of the beam and the sample such that material is ejected from a series of different points over the sample dependent on the image of the sample formed by the image forming means.

2. An apparatus according to claim 1 in which the means for controlling is effective to define an analysis zone corresponding to points which are contained within a polygonal zone on the sample.

3. An apparatus according to claim 2 in which the polygonal zone is a reentrant zone.

4. An apparatus according to claim 2 or claim 3 in which at least one area within the polygonal zone is excluded from the analysis zone.

5. An apparatus according to any one of claims 2 to 4 in which the means for controlling includes a computer linked pointing device effective to define the analysis zone.

6. An apparatus according to any one of claims 2 to 5 including means effective to determine the intensity contour of the image, and means for determining said analysis zone dependent on said intensity contour.

7. An apparatus according to any one of claims 2 to 6 including algorithm generating means effective to define said analysis zone.

8. An apparatus according to any one of the preceding claims including means effective to store sets of three numbers, each set defining the position and a predetermined dwell time of the beam for a pixel corresponding to a respective point on said sample, and means for addressing any pixel in any order for the predetermined dwell time of the pixel.

9. An apparatus according to claim 8 including means for accumulating signals from different pixels.

10. An apparatus according to claim 8 including means effective to generate the sets of three numbers from the image of the sample.

11. An apparatus according to claim 8 including an processing means effective to generate the sets of three numbers independently of the image of the sample.

12. An apparatus according to any one of claims 8 to 11 including an electronic storage means configured such that the position of each pixel in an analysis zone determines the address in the store where the signal from that pixel is accumulated.

13. An apparatus according to claim 8 including displacement means which acts on the beam scanning mechanism to displace the position of each pixel by the same amount.

14. An apparatus according to claim 13 including means for varying the displacement caused by the displacement means as a function of time.

15. An apparatus according to claim 8 including means for varying the dwell time of the beam as a function of time.

16. A sample analysis method, the method comprising the steps of: producing an excitation beam effective to eject material from a point on a sample; forming an image of the sample; and controlling the position of intersection of the beam and the sample such that material is ejected from a series of different points over the sample dependent on the image of the sample.

17. A sample analysis apparatus substantially as hereinbefore described with reference to the accompanying figures.

18 A sample analysis method substantially as hereinbefore described with reference to the accompanying figures.