Haruo Hirano

Design and development of an ultrahigh vacuum high-resolution transmission electron microscope

Abstract A 200 kV ultrahigh vacuum high-resolution transmission electron microscope (UHV-HRTEM) for in-situ surface observation has been developed. With use of metal sealings, the vacuum at the specimen position was 2.6×10 −7 Pa without any cooling traps, and less than 1.5×10 −7 Pa with the built-in cooling traps cooled by LN 2 . A cantilever-type goniometer, a specimen holder and high-resolution objective lens were newly developed for high-resolution transmission (HRTEM) and reflection (HRREM) to observe surfaces with a point-to-point resolution of 0.21 nm. The specimen tilt is up to ±20°. Crucible-type and direct-heating-type specimen holders were constructed for in-situ experiments on thin films and small particles, and on bulk crystalline surfaces, respectively. The usefulness of the microscope for surface studies is shown by images of clean surfaces of Au and Si prepared in the microscope.

Outgassing characteristics and microstructure of a ‘‘vacuum fired’’ (1050 °C) stainless steel surface

‘‘Vacuum fired’’ (1050 °C) type 304 stainless steel (SS304) surfaces were investigated with scanning electron microscopy and Auger electron spectroscopy. The grain boundaries of the surface were vague and shallow, which occurred due to elemental diffusion at high temperature in vacuum. The newly formed oxide layer of the vacuum fired surface was much thinner, and could be said to be finer in microstructure than the native layer of an ‘‘as‐received’’ surface. A vacuum fired (1050 °C) SS304 chamber was evacuated by a sputter ion pump whose vessel was pretreated by vacuum firing (1050 °C). An extremely high vacuum of 1.5×10−9 Pa (N2 equivalent pressure) was indicated by an extractor ionization gauge after a mild bake (170 °C) following an air exposure. The outgassing rate of the chamber wall at the elapsed time of one day after a mild bake (170 °C) was roughly estimated as low as 2×10−11 Pa l/s cm−2 (N2 equivalent value) by an orifice method. Vacuum firing has the effects of degassing the gas molecules solut...

Matrix calculation of pressures in high‐vacuum systems

A matrix analysis method has been applied to two high‐vacuum systems: an outgassing pipe and an electron microscope. These vacuum systems were analyzed as linear vacuum circuits with pressure sources, current sources, and conductances by using a digital computer. The pressures at six points along an outgassing pipe were calculated for two different models. A 5% difference of pressures was noted between the models. Pressures throughout an electron microscope high‐vacuum system were also successfully calculated by a computer. The pressure varied over three orders of magnitude depending on position.

Resistor network simulation method for a vacuum system in a molecular flow region

A simulation method to obtain the pressure distribution in a complex vacuum system has been proposed. The method introduces a new concept with regard to the function of each component of the vacuum system. The vacuum pump is regarded as a ‘‘vacuum resistor’’ connected at one side to the perfect vacuum. The gas source is regarded as a pressure generator connected to the vacuum. The conducting pipe is regarded as a vacuum resistor between the above elements. The vacuum sides of the pump element and the gas source element are assumed to be connected together by an imaginary route, and thus the vacuum system can be regarded as a closed vacuum circuit network. Such a vacuum circuit network may be replaced by an electric circuit network of a simulator for the vacuum system. The simulator was employed for the high vacuum system of an electron microscope in order to obtain the pressure distribution. The results obtained were in good agreement with the actually measured pressure distribution.

Two‐point pressure method for measuring the outgassing rate

The two‐point pressure method for measuring the outgassing rate of a solid material has been introduced, in which the pressures at two points in a pipe are measured. This method was applied to measuring the outgassing rates of two kinds of SUS304 plates, belt‐polished plates, and buff‐polished plates. The outgassing rates were measured under a pressure as low as that expected in the actual high‐vacuum system. Additionally, the one‐point pressure method has been introduced, whose validity was ascertained using measured pressures in the experimental setup. The outgassing rates for the same kinds of SUS304 plates were again measured by the conventional orifice method for comparison.

A cascade diffusion pump system for an electron microscope

Several DP systems have been discussed and evaluated with respect to dynamic evacuation under actual operating conditions of an electron microscope. As a result, a cascade DP system has been found to be the most reasonable for an electron microscope where the column is evacuated by the first DP backed by the second DP, and for the camera chamber by the second DP with a large volume foreline backed by a mechanical pump. A spectrum of the residual gases in an electron microscope with a cascade DP system indicates a very clean vacuum with hydrocarbon partial pressures of less than 3×10−7 Pa.

Mechanism of contamination build-up induced by fine electron probe irradiation

Abstract To clarify the transfer mechanism of contaminating molecules and the dominant source of them, carbon thin films were irradiated with a fine electron probe of 12 nm dia using three types of anti-contamination devices, individually. No contamination deposit was observed for a certain period or an incubation time after the specimen was pretreated by an electron beam shower. After the incubation time, the growth rate increased gradually with the elapsed time after the electron beam shower treatment and reached the saturated rate. The incubation time and the saturated growth rate depended largely on the type of the anti-contamination device used. It can be concluded that hydrocarbon molecules being adsorbed on the specimen surface are transferred to the irradiated area by surface diffusion. The dominant source of contaminating molecules is the residual gas molecules in the vicinity of the specimen even in a clean high vacuum.

Ar‐pumping characteristics of diode‐type sputter ion pumps with various shapes of ‘‘Ta/Ti’’ cathode pairs

Ar‐pumping characteristics of diode‐type sputter ion pumps (30 anode cells, 0.15 T of magnetic flux density at the center of the discharge area) with various shapes of ‘‘Ta/Ti’’ cathode pairs were measured by an orifice method. The Ar‐pumping speed of the pump of a ‘‘flat Ta/flat Ti’’ cathode pair was about 13 l/s in the 10−5 Pa range. On the other hand, the Ar‐pumping speeds of the pumps of a ‘‘holed Ta on flat Ti/flat Ti’’ pair and of a ‘‘slotted Ta on flat Ti/flat Ti’’ pair and that of the pump of a ‘‘slotted Ta on flat Ti/slotted Ti on flat Ti’’ pair were about 23 l/s and 24–25 l/s, respectively, in the 10−5 Pa range. The N2‐pumping speed of the pump of the slotted Ta on flat Ti/flat Ti pair was measured as about 55 l/s in the 10−5 Pa range. Therefore, the Ar‐pumping speed was as large as 42% of the corresponding N2‐pumping speed for the pump of the slotted Ta on flat Ti/flat Ti cathode pair.

A three‐point‐pressure method for measuring the gas‐flow rate through a conducting pipe

A new method for measuring the gas‐flow rate through an outgassing pipe has been introduced, in which the pressures at three different points in the pipe are measured to calculate the real gas‐flow rate. This method was applied to measuring the flow rate of nitrogen through a pipe. The gas‐flow rates calculated using three pressures were compared with those calculated using two pressures under the same condition. The method based on three pressures gave real rates which were higher than those based on two pressures, especially in lower pressure regions, thus indicating the presence of outgassing.

Outgassing characteristics of an electropolished stainless‐steel pipe with an operating extractor ionization gauge

The outgassing of a Bayard–Alpert ionization gauge (BAG) with a tungsten filament was observed to increase with the pressure in an isolation test. This was caused by the temperature rise of the hot filament, resulting from the gradual pressure rise. The outgassing of an extractor ionization gauge with an iridium filament coated with thorium oxide was much lower than that of the BAG. The net outgassing rate of an electropolished (EP)‐type 304 stainless‐steel (SS304) pipe wall was measured by the isolation method. The outgassing rate of the pipe wall at 3.0×10−6 Pa, after an ‘‘in situ’’ bakeout (150 °C, 20 h), was measured as 1.1×10−11 Pa l/s cm−2 by neglecting the outgassing and pumping effects of the operating EXG.