US 3558878 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
- "States Patent I72] Inventor Hermann Neuhaus  References Cited La Canada Calif- UNlTED STATES PATENTS 31 P 745'427 3,008,044 ll/l96l Buchhold 250 495 [-2] Flled July 17, [968 3 I07 297 101963 W1ttry 250/49.5X pdtcmcd 3 387 132 6/1968 H r t 1 250/49 5 [731 Assignees Applied Research Laboratories, Inc. e rmanne a Sunland, Calif. Primary ExaminerWilliam F. Lindqtlist a corporation of Delaware A!t0meys-'H0ffman Stone  PROBE BY ABSTRACT: An electron microprobe including a magnetic objective lens for focusing an electron beam upon a specimen, P THE PBJIZICTWE LENS and means for cooling the lens to a temperature below about 3 Clams 4 Drawmg 20C. It is found that operating results are greatly improved 152 US. Cl 250 495 y mainta ing t s cold and the specimen warm-  Int. Cl ..H0lj 37/14, tamination of the specimen, particularly by carbonaceous GOln 23/22 materials, can be reduced by more than an order of magni-  Field of Search 250/495 tude without compromise with the operational features of the instrument.
PATENTEDJMBIQH v 7 3558.878
sum 10F2 flllihw- W 3O AT+22C I FIG. 2
3' I INVENTOR. AT -35C HERMANN NEUHAUS ATTORNEY 2O 3'0 BY W.
ELECTRON BEAM KV METHOD OF REDUCING SPECIMEN CONTAMINATION IN AN ELECTRON PROBE BY COOLING TI'IE OBJECTIVE LENS BRIEF SUMMARY This invention relates to a novel method of and apparatus for rapid chemical analysis by bombarding a specimen with electrons and spectrometrically analyzing x-rays emitted by the specimen in response to the bombardment. More particularly, the invention has special reference to instruments known as electron microprobes, in which the bombarding electron beams are usually focused to very small incremental surface areas of the specimens typically about I a to about a in diameter.
Microprobes are well-known. One that has found substantial commercial acceptance is described and claimed by D. B. Wittry in US. Pat. No. 3,l07,297, and that patent may be referred to to augment the description herein.
In manyways, electron microprobes are similar to electron microscopes, but they are basically different in that in the microprobe, the conditions that are sensed are those that occur on the same surface of the specimen upon which the electron beam impinges, whereas in the electron microscope, the desired information is gathered from the surface opposite from the impinging beam. In the microscope, the electrons are controlled and focused after they have passed through the specimen. It has long been known that in the electron microscope, certain deleterious effects thought to be due to contamination of the bombarded surface of the specimen can be reduced by mounting a cryogenic element close to the sur face where the electrons impinge. The technique has also been used in the electron microprobe butwith limited success and acceptance because of the need to provide for observation of the bombarded surface. It is difficult to arrange a cryogenic element closely adjacent to the area of the specimen to be bombarded without at the same time interfering with the operation of the instrument such as, for example, by obscuring the x-ray radiation it is desired to detect.
Briefly, according to the present invention, it has now been found that simply cooling the fairly large objective lens structure in a previously available microprobe to temperatures in the range of about C. to 70 C. overcomes to a high degree the problems heretofore presented by contamination. The size of the cryogenic element, and its closeness to the workpiece have been found to be relatively more important than its temperature. In the practice of the invention, the lens structure itself serves as the cryogenic element. It is relatively large, and quite close to the workpiece, and cooling it to the relatively moderate temperatures noted herein has been found to achieve much more effective contamination control than the use of a smaller, separate element chilled to a much lower temperature such as that of liquid nitrogen, l96 C. It has thus been found possible to decontaminate cryogenically without having to add a structural element in the narrow confines adjacent to the specimen, and without interfering in any way with the normal operation of the instrument.
' DETAILED DESCRIPTION A representative embodiment of the invention will now be described in detail in conjunction with the accompanying drawings, wherein:
FIG. I is a cross-sectional view of the objective lens and specimen holder of an electron microprobe in accordance with the invention;
FIG. 2 is a chart comparing the effects of contamination by carbonaceous materials in the practice of the invention with the effects noted in a previous practice;
FIG. 3 is a chart showing the contamination effects as a function of the temperature of the objective lens in the electron microprobe; and
FIG. 4 is a schematic diagram of a refrigeration system for cooling the objective lens in accordance with a presently preferred embodiment of the invention.
One of the basic requirements for accurate and reliable analysis in the electron microprobe is that the surface of the specimen being bombarded be pure, that is, representative only of the materials of which the bombarded region of the specimen is composed and free of contamination by materials deposited on the specimen from the atmosphere. During an analysis, the surface of the specimen should remain in a static condition; none of its constituents should be removed, not should any other constituents be added. In the usual case. however, the surface of the specimen is actually contaminated by the deposit thereon of material from the atmosphere in which it is supported. The contaminating materials of primary concern, that is, those that have the most important adverse effects on analysis, are carbonaceous. They may come from any part of the vessel in which the electron bombardment is carried out, and may often be inadvertently or unavoidably carried into the vessel by the specimen itself.
The rate of contamination has been found not to be the same for all specimens, but to vary depending on the composition of the specimen, the condition of its surface. the atmosphere in which it is maintained during the analysis, and the temperatures of other bodies near the specimen and of the specimen itself. The rate of contamination may be reduced by improving the vacuum, by providing a relatively cold, heat-absorbing body near the specimen, or by heating the specimen.
Improvement of the vacuum appears to provide only limited reduction in contamination, and to be uneconomically expensive. It has been found empirically that for a 50 percent reduction in the rate of contamination. the vacuum must be improved by about a full magnitude. Microprobes operate typically in the 10- torr. range, and an improvement to l() torr. or l0 torr. would require the use of much more expensive pumping equipment than is generally used at present. Moreover, it is desirable to reduce the rate of contamination not by a factor of two or four, but by at least an order of magnitude. It was also found that back streaming of oil vapors from the diffusion vacuum pump contributes only to a negligible extent to the contamination of the specimen, and it was concluded that the major part of the contamination originates from objects and substances that are normally within the evacuated chamber.
It was decided, therefore, to investigate the use of cryogenic surfaces in proximity to the surface of the specimen, and also to heat the specimen. Despite the difficulties of finding adequate space in which to mount them, elements of various different shapes and sizes were tried, each being mounted as closely as possible to the specimen, typically about one-eighth inch. They were cooled by liquid nitrogen to about l96 C. The best that could be achieved in the manner was a reduction in contamination effects to about onequarter of the effects noted in conventional operation. It was noted, however, that the size of the cryogenic element appeared to be an important factor in its effectiveness; the larger elements reduced the effects of contamination more than the smaller elements did.
It was then realized that in the microprobe structure already existing, a large coolable surface was present. As shown in FIG. I this surface was the lower face 10 of the objective lens 12, which extends, as shown, over the upper surface of the specimen [4 spaced about one-quarter inch above it. As shown, the lens 12 includes an energizing coil 16 contained within a copper shell 18, which, in turn, is secured within an iron housing comprised of a cover plate 20 and a cup-shaped lower member 22. An annular tube 24 surrounds the copper shell 18 in intimate thermal contact with it. Originally, ordinary tap water had been circulated through the tube 24 to remove heat generated in the coil 16 by the current used to energize it.
In accordance with the invention, a refrigerant is circulated through the tube 24 to cool the entire lens 12 to a temperature below about -20 C. The improvement in results is graphically illustrated in FIGS. 2 and 3. FIG. 2 shows the carbon Ka signal emitted from a specimen of supposedly pure nickel under bombardment by an electron beam at various different energies. The upper curve 30 indicates the signal strength noted when the objective lens 12 was maintained, in accordance with prior practice, at about +22 C. The lower curve 31 indicates the signal strength noted when the lens 12 was cooled to 35 C. The signals were plotted as a fraction of the signal emitted by a control specimen of pure carbon. The ordinate scale is logarithmic. The carbon Ka signal from the nickel specimen was smaller by a factor of 20 or more when the objective lens 12 was operated at 35 C. than at ordinary room temperature.
The chart of FIG. 3 illustrates the effect of changes in the operating temperature of the objective lens 12 over the range of about +55 C. to about 70 C. ceteris paribus. In this case. the specimen was of carbon-free iron, the bombarding electron beam was 0.1a. amp.. the accelerating voltage was l kilovolts, the beam was focused and scanned in a raster pattern to cover an area of about 8000 square microns. The excitation chamber was evacuated to about torr.
Surprisingly, it was found that in most cases mild chilling. to about C. or 40 C. provides adequate decontamination and intense chilling is seldom required. At about 30 C., the effect of contamination is reduced by a factor of 10, a full order of magnitude. A further improvement of about three times may be achieved by cooling to 70 C., but cooling to below 70 C. appears to have relatively little further effect. The improvement curve, as shown in FIG. 3, becomes fairly flat below about 60 to -70 C.
It is preferred in the practice of the invention also to heat the specimen 14 during the analysis. Otherwise, it may be cooled excessively by the objective lens I2, because it is so close to the lens, and detract from the contaminating effect of the lens 12. It is thought that the reduction of contamination is the surface nearest the face 10 of the objective lens. Heating may be controlled by securing a sensing device 28 to the specimen and controlling the energization of the heating element 27 in accordance with the output signal of the sensing device.
Refrigeration apparatus of the Carnot cycle type is generally capable of cooling the lens 12 to about C., and is fully adequate for most purposes. In some cases, however, it may be desired to use a simpler system, or to chill to temperatures lower than 40 C. In these cases, liquid nitrogen, or any other cryogenic source may be used.
FIG. 4 illustrates a refrigeration system arranged for the selective alternate use of either liquid nitrogen or a conventional refrigerant. In this arrangement, the economy of the conventional refrigerant is available for the usual run of work, and liquid nitrogen. is available when more intense cooling is desired.
As shown, the refrigeration system includes a compressor and condenser unit 40, which feeds refrigerant to the tube 24 I in the objective lens through an expansion valve 42. From the controlled by an electrical control circuit 52 'of any desired type, which is arranged to adjust the throttle valve 44 in rcsponsc'to changes in temperature sensed by a sensor 26. The sensor 26 is secured in good thermal contact with the lens l2. preferably on the face I0 thereof adjacent to the specimen I4. Fairly good temperature regulation, within about 3.1).1" C., is important when maximum stability and accuracy of analysis are desired. Otherwise, the effects of thermal expansion may change the focus and location of the electron beam on the specimen. Temperature control within this range of accuracy is easily achievable with presently available commercial equipment.
To convert from operation with conventional refrigeration apparatus to cooling with liquid nitrogen, it is only necessary to shut down the compressor 40, close its inlet valve 54, open the exhaust valve 56, which is connected at the outlet of the diffusion pump baffle 46, and open the feed valve 58. which is connected between the tank 60 of liquid nitrogen and the inlet to the tube 24. Both the expansion valve 42 and the bypass valve 50 should also be closed. Closing the expansion valve 42 prevents loss of refrigerant, which might otherwise escape into the stream of nitrogen. Closing the bypass valve 50 prevents unnecessary waste of nitrogen.
Care should be taken to warm the lens 12 before releasing the vacuum in the sample chamber beneath it to replace the specimen 14. If the lens 12 is exposed to the ambient atmosphere while it is chilled, it becomes covered with ice very rapidly. and pumping the chamber down again takes an intolerable time, because the ice constitutes a large and'persistent virtual leak in the chamber. The lens 12 typically dissipates about watts of energy in operation, and this is usually ample to warm it quickly to room temperature, or a bit above, once the refrigeration is stopped. Auxiliary heating means may be used if desired.
1. Method of operating an electron microprobe of the kind having an objective lens structure positioned closely adjacent to the region where the electron beam normally impinges on a specimen comprising the steps of focusing the electron beam on the specimen, and cooling the electron lens structure to a temperature between about 20 .C. and 70 C. thereby to reduce the adverse effects of contamination of the specimen.
2. Method according to claim 1 wherein the objective lens structure is electromagnetic and includes means for circulating a fluid in closethermal contact with it, and the step of cooling is done by passing a cold fluid through the circulating means.
3. Method according to claim 1 including also the step of simultaneously heating the specimen to avoid excessive cooling of it by the lens structure.