Method for etching metal or metal oxide by ozone water, method for smoothing surface of metal or metal oxide by ozone water, and patterning method using ozone water

Abstract

Provided are a method for etching a metal or metal oxide without using a reagent, etc., that affects the environment, a method for smoothing a surface of a metal or metal oxide on an atomic level, and a method for patterning on an atomic level. Etching of a metal or metal oxide, or smoothing of a surface of a metal or metal oxide is possible using ozone water in which only ozone is dissolved. Patterning can also be performed by providing a metal that does not dissolve in the ozone water as a resist on a metal or metal oxide that can be etched by ozone water in which only ozone is dissolved, and etching using the ozone water.

Claims

1. A patterning method comprising the step of patterning a surface by etching with ozone water of solely ozone dissolved in water, wherein the etching is performed after Ti is provided on a metal or metal oxide that is etchable with the ozone water at an etching temperature of 20 to 60° C., and the metal or the metal oxide is ionized either directly or via an intermediate in the ozone water at a pH of 4.3 to 4.4, and an oxidation-reduction potential of +2.07 V.

2. The patterning method according to claim 1, wherein the etching involves ultrasonic vibration and/or ultraviolet irradiation.

3. A patterning method comprising the step of patterning a surface by etching with ozone water of solely ozone dissolved in water, wherein the etching is performed after Ti is provided on a metal or metal oxide that is etchable with the ozone water at an etching temperature of 20 to 60° C., and the metal or the metal oxide is ionized either directly or via a hydroxide in the ozone water at a pH of 4.3 to 4.4, and an oxidation-reduction potential of +2.07 V.

4. The patterning method according to claim 3, wherein the etching involves ultrasonic vibration and/or ultraviolet irradiation.

5. A patterning method comprising the step of patterning a surface by etching with ozone water of solely ozone dissolved in water, wherein the etching is performed after Ti is provided on a metal or metal oxide that is etchable with the ozone water at an etching temperature of 20 to 60° C., and the metal or the metal oxide is ionized either directly or via an oxide in the ozone water at a pH of 4.3 to 4.4, and an oxidation-reduction potential of +2.07 V.

6. The patterning method according to claim 5, wherein the etching involves ultrasonic vibration and/or ultraviolet irradiation.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 represents the relationships between pH and oxidation-reduction potential for different metals.

(2) FIG. 2 is a diagram schematically representing an experimental apparatus used in the present invention.

(3) FIG. 3 is a graph representing the measurement results for the etching rate of Cr, showing the relationship between ozone water temperature and concentration versus the etching rate of Cr.

(4) FIG. 4 is a graph representing the measurement results for the etching rate of Ni, showing the relationship between ozone water temperature and concentration versus the etching rate of Ni.

(5) FIG. 5 is a graph representing the measurement results for the etching rate of Al, showing the relationship between ozone water temperature and concentration versus the etching rate of Al.

(6) FIG. 6 is a graph representing the measurement results for the etching rate of Au, showing the relationship between ozone water temperature and concentration versus the etching rate of Au.

(7) FIG. 7 is a graph representing the measurement results for the etching rate of Ti, showing the relationship between ozone water temperature and concentration versus the etching rate of Ti.

(8) FIG. 8 represents an Arrhenius plot with an inverse of temperature taken on the horizontal axis against the vertical axis representing a logarithm of the slope of the least square approximation straight lines obtained in FIGS. 3 to 7.

(9) FIG. 9 shows AMF images of a Cr sample (a) before etching, (b) after etching with a conventional etchant, and (c) after etching with ozone water.

(10) FIG. 10 shows AMF images of a Ni sample (a) before etching, (b) after etching with a conventional etchant, and (c) after etching with ozone water.

(11) FIG. 11 shows AMF images of an Al sample (a) before etching, (b) after etching with a conventional etchant, and (c) after etching with ozone water.

(12) FIG. 12 shows AMF images of an Au sample (a) before etching, (b) after etching with a conventional etchant, and (c) after etching with ozone water.

(13) FIG. 13 shows AMF images of a Ti sample (a) before etching, and (b) after etching with ozone water.

(14) FIG. 14 is a graph representing changes in surface roughness of a Cr sample after a predetermined time period of etching with ozone water.

(15) FIG. 15 shows AMF images of a Cr sample after a predetermined time period of etching with ozone water.

DESCRIPTION OF EMBODIMENTS

(16) The following describes the present invention in detail, specifically a method for etching metal or metal oxide with ozone water, a method for smoothing a surface of metal or metal oxide with ozone water, and a patterning method that uses ozone water.

(17) The ozone water of the present invention is of solely ozone dissolved in water. Conventional metal etchants use materials such as hydrochloric acid, nitric acid, sulfuric acid, hydrofluoric acid, perchloric acid, phosphoric acid, ammonium cerium(IV) nitrate, iodine, potassium iodide, tetramethylammonium hydroxide (TMAH), potassium hydroxide (KOH), hydrogen peroxide water, and acetic acid. These are environmentally harmful, and the waste fluid cannot be disposed of unless treated beforehand. This adds to the cost. Ozone water has been used for washing in the field of semiconductors as described in the foregoing Patent Literature and Non Patent Literature. However, hydrochloric acid or the like is added to dissolve the metal residues, and ozone water only has a supportive role. Further, ozone water is conventionally intended to remove impurities such as resist compounds and metal residues, and is not used for etching a bulk metal itself. Unlike the conventional use of ozone water, the ozone water of the present invention is solely used for etching a bulk metal or metal oxide itself, and for smoothing a bulk metal surface or a metal oxide surface (hereinafter, etching and smoothing are also referred to simply as “etching”, and metal and metal oxide are also referred to simply as “metal”).

(18) The ozone water of the present invention is produced with ozone generated from oxygen and dissolved in water. The water used may be any of tap water, distilled water, and ultrapure water, and is preferably ultrapure water because it does not contain impurities, and provides a stable quality. The concentration of ozone water depends on the type of the dissolved metal and temperature, and is preferably about 2 ppm or more. A concentration below 2 ppm makes the metal etching rate too slow, and is not practical. With an ozone water concentration of about 2 ppm or more, the etching rate improves in proportion to ozone water concentration.

(19) The etching temperature of ozone water is preferably 20 to 60° C., particularly preferably 40 to 60° C. An etching temperature of 20° C. or less makes the metal etching rate too slow, and is not practical. An etching temperature above 60° C. increases the etching rate, but is not preferred because the ozone decomposition accelerates. The etching rate is highly dependent on ozone concentration and temperature, and may be appropriately adjusted according to the type of etched metal, and the etch depth. In actual practice, the supplied ozone amount may be adjusted to provide the desired ozone concentration with the use of an ozone concentration monitoring device. A heater or the like may be used to provide a constant temperature for ozone water.

(20) Etching may be performed by dipping metal in ozone water, or by spraying or showering ozone water over metal. For improved etching effect, ultrasonic vibration may be applied during the etching. It is also possible to use ultraviolet irradiation from a low-pressure mercury lamp in combination. For example, water may be irradiated with vacuum ultraviolet rays of 172 nm wavelength, or titanium oxide may be irradiated with ultraviolet rays in water to reconvert oxygen into ozone in water, and increase the ozone concentration.

(21) The metal etching time may be appropriately adjusted according to the type of etched metal, etch depth, ozone water concentration, and temperature.

(22) The metal etched or the metal smoothed on the surface by the present invention is not particularly limited, as long as it is etched in the end with ozone water alone. Examples include metals that are directly ionized with ozone water, and metals that are ionized via intermediates such as hydroxide and oxide. The present inventors found that dissolving 10 mg/L or more of ozone in water stabilizes the oxidation-reduction potential at about +2.07, and stabilizes the pH at about 4.3 to 4.4, irrespective of the ozone concentration, though the pH slightly varies with temperature. The product of the reaction between metal and the solution can be predicted to a certain extent from pH and oxidation-reduction potential such as in the pH vs. oxidation-reduction potential diagrams for different metals shown in FIG. 1. The upper limit of ozone water concentration is not particularly limited, as long as ozone is dissolved in water, and may be appropriately adjusted according to such factors as the type of etched metal, and etch depth.

(23) For example, the reaction that oxidizes Cr with ozone and produces Cr.sub.2O.sub.7.sup.2− appears to become dominant at a pH near 4.3 to 4.4, and an oxidation-reduction potential near +2.07 V. It also appears that Cr.sub.2O.sub.7.sup.2− exists in a mixed state with Cr.sup.3+ during the etching because Cr.sub.2O.sub.7.sup.2− turns itself into Cr.sup.3+ in an acidic solution. Similarly, the reaction that ionizes Ni to Ni.sup.2+, and the reaction that ionizes Al to Al.sup.3+ appear to dominate. Au is believed to become Au(OH).sub.3, and a reaction that produces Au.sup.3+ and 3OH from Au(OH).sub.3 appears to take place in the aqueous solution, though to limited extent. As clearly indicated by the Arrhenius plot to be described later, Au shows an activation energy about twice as small as those of other metals, and ionizes via hydroxide unlike other metals that ionize through oxidation. The distinctly different activation energy of Au is believed to be due to the different etching mechanism, and etching appears to proceed, though the rate is slow. In contrast, Ti forms the passive Ti.sub.2O.sub.3, and etching does not appear to proceed, as clearly indicated in FIG. 1 and Comparative Examples (described later).

(24) Metals that directly ionize with ozone water, and metals that ionize via intermediates such as hydroxide and oxide in etching are not limited to Cr, Ni, Al, and Au. Non-limiting examples of other such metals include Cu, Ag, Fe, Pt, Mn, Zn, Pd, and Ir.

(25) The present invention enables etching metal by dissolving metal with ozone water in the manner described above. The present invention also enables smoothing a metal surface at an atomic level, in a manner different from the conventional approach using a conventional etchant. As used herein, “smoothing” does not mean dissolving and removing organic materials, metal residues, and other such materials on a substrate with an etchant, or smoothing a surface by polishing metal with a polisher, but means dissolving a bulk metal surface with ozone water, and smoothing the metal surface at an atomic level.

(26) The ozone water smoothing distinguishes itself over the use of conventional etchants by the very small molecule size of ozone ions as small as water molecules, and the high diffusion coefficient of ozone, providing easier entry between the metal atoms. The metal atoms projecting out of the surface have large defects, and are highly reactive. This, combined with the large surface area/volume ratio, allows these metal atoms to be quickly ground down. It is believed that ozone does not tend to depend on crystal orientation, and this appears to be a reason that enables smoothing without creating a crystal plane in the etched metal surface. Some conventional etchants dissolve metal, and form compounds with the metal. Such compounds may accumulate and form irregularities on metal surface. Ozone, on the other hand, does not form such compounds and thus does not form surface irregularities.

(27) For improved accuracy, atomic level surface smoothing is required in the field of semiconductors and other applications. The etching or smoothing rate of metal with ozone water is slower than the rates of conventional etchants, though it depends on the concentration and temperature of ozone water. However, the ozone water smoothing is effective for high accuracy smoothing and atomic level smoothing in terms of conditions such as etch depth, the extent of smoothing, and process time. For faster smoothing, the smoothing method of the present invention may be used in the final smoothing process after smoothing the surface to some extent by using conventional techniques such as polishing.

(28) With the advancement in nanotechnology, the trend for further miniaturization and higher density is expected to continue in semiconductors, and the need for finer, atomic-level circuits should arise. The atomic level etching or smoothing with the ozone water of the present invention is very useful for the production of nanodevices because the process proceeds at a rate sufficient for actual applications, and enables etching to be very finely controlled under the adjusted temperature and concentration of the ozone water.

(29) It has been required to use different etchant compositions for different metals requiring etching or smoothing, and to use a different waste fluid process suited for each etchant. However, ozone water can dissolve many different metals commonly used in the field of semiconductors, and there is no need to change etchants according to the metallic species. Further, the need for a waste fluid processing device is eliminated, and efficient etching or smoothing is possible.

(30) Metal etching with conventional etchants involves formation of a compound by a reaction of etchant and metal, and the process may require isolating the metal when reusing the metal. On the other hand, ozone water turns itself into water over time, and metal can be collected simply by drying. This makes it possible to easily collect noble metals and rare metals such as gold, and reuse such resources.

(31) Further, because ozone water decomposes into oxygen and water over time, the rinsing following the etching or smoothing process can be sufficiently carried out only with water. This simplifies the manufacturing steps of semiconductors and other products, and solves the waste issue.

(32) The present invention can etch metal or smooth metal surface with ozone water alone. The invention also allows insolubilizing or ionizing metal by adjusting the pH of ozone water. Specifically, for example, a pH adjuster may be added to ozone water to ionize metal that cannot be ionized with the pH and the oxidation-reduction potential of the ozone water alone, or to directly ionize metal, for example, such as Au, that ionizes in two steps, provided that it does not harm the environment which is the object of the present invention. Examples of environmentally unharmful pH adjusters include sodium bicarbonate, carbon dioxide gas, and ammonium carbonate. By making the pH more alkaline, the etching rate can be increased ten fold or more, though the ozone lifetime becomes shorter.

(33) While some metals are etched with ozone water, other metals, such as Ti, form a passive oxidation coating with ozone water, as described above. Such metals do not ionize and dissolve in ozone water once a passive oxidation coating is formed. By using this property, for example, ozone water etching may be performed after a metal, such as Ti, that is not etchable with ozone water is provided as a resist on a metal that is etchable with ozone water. This enables patterning at an atomic level without harming the environment.

(34) The present invention is described below in greater detail using Examples. It should be noted that the following Examples are given solely for the purpose of illustrating the present invention as a reference to the specific aspects of the invention. The following illustrative examples describing the specific aspects of the present invention are not intended to limit or restrict the scope of the invention disclosed herein.

EXAMPLES

(35) Experimental Apparatus

(36) FIG. 2 is a diagram schematically representing an experimental apparatus used in the present invention. Ozone was produced with an ozone generator 2 (ED-OG-R4, Ecodesign) by using oxygen gas 1 as raw material. The ozone was supplied into the ultrapure water in a beaker 4 through a gas bubbler 3 to produce ozone water 5. An etching experiment was conducted by dipping a sample substrate 6 in the ozone water 5 for a predetermined time period. In order to provide a uniform ozone concentration in the ozone water, the dissolved ozone concentration was monitored with a photo-absorptive dissolved ozone concentration meter (O3-3F, Kasahara Chemical Instruments, not shown) while stirring the ozone water with a stirrer 7 throughout the etching process. For the evaluation of the temperature dependence of etching, the liquid temperature was maintained constant with an external heater 8.

(37) Fabrication of Sample Substrate 6

(38) The sample substrate 6 for etching evaluation was fabricated by forming various metal films on an optically polished glass substrate (surface roughness of 1 nm or less) by vacuum vapor deposition, and patterning the surface by photolithography to provide an alternately occurring metal and glass surfaces.

(39) Etching Rate Evaluation

(40) Etching rate was evaluated by measuring the height from the glass surface to the metal surface before and after the etching. An optical interference profilometer (WYKO NT 9100A, Bruker AXS) was used for the measurement. The average of 5 measurement points taken on the substrate under each condition was used as the measured value.

(41) Post-Etching Surface Roughness Evaluation

(42) Surface roughness was compared and evaluated for metals subjected to ozone water etching, and metals etched with a common etchant. Surface roughness was evaluated after etching about 30 nm into metals by conventional etching and ozone water etching. The surface roughness of the etched metal was evaluated in terms of the root-mean-square roughness (Rq: JIS 1994) taken at 15 arbitrary regions (5 μm×5 μm) on the substrate, using an atomic force microscope (AMF; VN-8010, KEYENCE).

(43) Metal Etching Rate

Example 1

(44) A sample substrate with vapor deposited Cr was produced by using the procedures described in Fabrication of Sample Substrate 6 above. Etching was performed under varying dissolved ozone concentrations and temperatures (25° C., 40° C., and 55° C.). After the etching, the sample substrate 6 was thoroughly rinsed with ultrapure water, and dried for evaluation. The etching rate was measured by using the procedures described in Etching Rate Evaluation above. FIG. 3 is a graph representing the measurement results. In the graph, the vertical axis and the horizontal axis represent etching rate and ozone concentration, respectively, and the error bars each represent standard deviation. Solid lines are the approximation straight lines obtained by using the method of least squares. The same notation is used in FIGS. 4 to 7.

Example 2

(45) Etching rate was measured by using the same procedures used in Example 1, except that Ni was used. FIG. 4 is a graph representing the measurement results.

Example 3

(46) Etching rate was measured by using the same procedures used in Example 1, except that Al was used. FIG. 5 is a graph representing the measurement results.

Example 4

(47) Etching rate was measured by using the same procedures used in Example 1, except that Au was used. FIG. 6 is a graph representing the measurement results.

Comparative Example 1

(48) Etching rate was measured by using the same procedures used in Example 1, except that Ti was used. FIG. 7 is a graph representing the measurement results.

(49) As clearly shown in the results of Examples 1 to 4 and Comparative Example 1, the etching rates of Cr, Ni, Al, and Au increased with increase in the ozone concentration of the ozone water. It was also found that the etching rate increased with increasing ozone water temperatures. However, because the ozone lifetime becomes shorter as the ozone water temperature increases, the dissolved ozone concentration itself decreased. That is, the balance between ozone concentration and temperature was found to be important in the etching of metal with ozone water, and the effect of temperature was found to be more dominant than the dissolved ozone concentration.

(50) Ti was not etched with ozone water, and formed an oxidation coating instead. It was found that the oxidation coating formed at increasing rates with increasing ozone concentrations in ozone water, and with increasing ozone water temperatures.

(51) FIG. 8 represents an Arrhenius plot with an inverse of temperature taken on the horizontal axis against the vertical axis representing a logarithm of the slope of the least square approximation straight lines obtained in FIGS. 3 to 7. The linearity of the Arrhenius plot confirmed that the etching reaction followed the Arrhenius equation. The activation energy (the slope of the plot) was essentially the same for Al, Cr, and Ni, and the result that Au showed an activation energy that was at least two times smaller than the activation energies of these metals suggests that Al, Cr, and Ni have similar etching mechanisms, different from the etching mechanism of Au. Note that the activation energy of Ti shown in FIG. 8 was determined from the increased thickness, and is not based on etching.

(52) Post-Etching Surface Roughness of Metal

Example 5

(53) A Cr sample produced by using the procedures described in Fabrication of Sample Substrate 6 above was dipped in ozone water (temperature 25° C., ozone concentration 10.81 mg/L) for about 4 hours, and etched about 30 nm. After the etching, the sample substrate 6 was thoroughly rinsed with ultrapure water, and dried for evaluation. The surface roughness was measured by using the procedures described in Post-Etching Surface Roughness Evaluation above. FIG. 9(a) represents an AMF image at an arbitrary location of the sample before etching, and FIG. 9(c) represents an AMF image at the arbitrary location of the sample after the etching of Example 5. In FIGS. 9 to 13 and FIG. 15, Rq (nm) on the bottom left of each image represents the root-mean-square roughness at an arbitrary location of the sample.

Comparative Example 2

(54) Surface roughness was measured by using the same procedures used in Example 5, except that a mixed acid of ammonium cerium(IV) nitrate+perchloric acid+water=165 g+43 mL+water (total liquid volume=1 L) was used as the etchant, and that the dip time was changed to 30 seconds. The chemicals used in Comparative Examples 2 to 6 are all products from Kanto Kagaku. FIG. 9(b) represents an AMF image at an arbitrary location of the sample after the etching of Comparative Example 2.

Example 6

(55) Surface roughness was measured by using the same procedures used in Example 5, except that a Ni sample was used, and that the dip time was changed to about 6 hours. FIG. 10(a) represents an AMF image at an arbitrary location of the sample before etching, and FIG. 10(c) represents an AMF image at the arbitrary location of the sample after the etching of Example 6.

Comparative Example 3

(56) Surface roughness was measured by using the same procedures used in Example 6, except that a mixed acid of hydrochloric acid 85%+15% nitric acid was used as the etchant, and that the dip time was changed to 15 seconds. FIG. 10(b) represents an AMF image at an arbitrary location of the sample after the etching of Comparative Example 3.

Example 7

(57) Surface roughness was measured by using the same procedures used in Example 5, except that an Al sample was used, and that the dip time was changed to about 2.5 hours. FIG. 11(a) represents an AMF image at an arbitrary location of the sample before etching, and FIG. 11(c) represents an AMF image at the arbitrary location of the sample after the etching of Example 7.

Comparative Example 4

(58) Surface roughness was measured by using the same procedures used in Example 7, except that a mixed acid of phosphoric acid 80%+nitric acid 5%+acetic acid 10%+water 5% was used as the etchant, and that the dip time was changed to 10 seconds. FIG. 11(b) represents an AMF image at an arbitrary location of the sample after the etching of Comparative Example 4.

Example 8

(59) Surface roughness was measured by using the same procedures used in Example 5, except that an Au sample was used, and that the dip time was changed to about 60 hours. FIG. 12(a) represents an AMF image at an arbitrary location of the sample before etching, and FIG. 12(c) represents an AMF image at the arbitrary location of the sample after the etching of Example 8.

Comparative Example 5

(60) Surface roughness was measured by using the same procedures used in Example 8, except that an aqueous solution of iodine 5%+potassium iodide 10% was used as the etchant, and that the dip time was changed to 15 seconds. FIG. 12(b) represents an AMF image at an arbitrary location of the sample after the etching of Comparative Example 5.

Comparative Example 6

(61) Because ozone water does not etch the Ti sample as noted above, surface roughness was measured in the same manner as in Example 5 except that the dip time was changed to about 24 hours, and the result was not compared with the conventional etchant. FIG. 13(a) represents an AMF image at an arbitrary location of the sample before etching, and FIG. 13(b) represents an AMF image at the arbitrary location of the sample after the etching of Comparative Example 6.

(62) Table 1 below summarizes the Rq values (nm) determined from the AFM images taken at 15 arbitrary locations in Examples 5 to 8, and Comparative Examples 2 to 6.

(63) TABLE-US-00001 TABLE 1 Before After conventional After ozone water Metal etching etching etching Cr 1.32 ± 0.39 3.61 ± 1.12 1.17 ± 0.22 Ni 0.87 ± 0.37 1.71 ± 0.34 1.69 ± 0.79 Al 4.10 ± 0.49 4.29 ± 1.07 3.17 ± 0.72 Au 1.11 ± 0.28 3.86 ± 0.44 1.06 ± 0.28 Ti 1.19 ± 0.32 — 4.37 ± 2.43

(64) As is clear from Table 1, the ozone water etching of Cr, Ni, Al, and Au produced flatter surfaces than conventional etching. The surfaces were particularly smooth at an atomic level in, for example, Cr and Au with Rq=1 nm, demonstrating that the technique was desirable not only for etching but for producing an atomically smooth surface. The absence of any local crystal plane suggests that the ozone water etching is isotropic, and is not dependent on crystal orientation.

(65) The ozone water etching of Ni produced a smoother surface than conventional etching. However, the etched surface was rough compared to the roughness measured before etching. This is considered to be due to the partial formation of hydroxide, which appears to occur as the Ni in contact with ozone water at an ozone water pH of 4.3 and an oxidation-reduction potential near +2.07 V is in the vicinity of the boundary where Ni.sup.2+ and NiO(OH) are formed. However, because the ozone water pH is adjustable with addition of a pH adjuster to ozone water to the extent that it is not harmful to the environment, a pH lowering agent may be added to ozone water when dissolving Ni so that Ni directly becomes Ni.sup.2+.

(66) The Ti sample dipped in ozone water had a rough surface because of the growth of what appeared to be a porous Ti.sub.2O.sub.3 coating.

(67) Time Dependence of Surface Roughness

Example 9

(68) In order to examine the surface smoothing effect, the dependence of surface roughness on etching time was examined by using Cr samples. The samples were etched with the etchant used in Comparative Example 2 to prepare Cr samples with a coarse surface (Rq=about 5.4 nm). Each sample was etched with ozone water that had a temperature of 55° C. and a dissolved ozone concentration of 1.62 mg/L, and the smoothing process of the surface was evaluated by AFM over the course of etching. FIG. 14 is a graph representing changes in surface roughness by etching after a predetermined time period. Each point in graph represents the mean value of 5 samples. FIG. 15 represents AFM images at an arbitrary location of each sample after a predetermined time period.

(69) As can be seen, the Cr film that had an average surface roughness Rq of 5.4 nm in the initial state had an average surface roughness Rq of about 1.5 nm after 12.5 hours. As clearly shown in the AFM images of FIG. 15, the surface projections become smaller, and the surface becomes smoother over Apparently, this is the result of the sharp, smaller curvature portions of the metal surface with a higher surface area/volume ratio being etched earlier than other portions.

INDUSTRIAL APPLICABILITY

(70) The invention enables etching or smoothing metal with ozone water alone, without using an environmentally harmful etchant, and has potential use in the environmentally friendly production of products such as semiconductor devices. The invention also enables etching or smoothing metal at an atomic level. The invention is therefore useful for the atomic-level production of devices, and has potential use in the production of various metallic components and materials in a wide range of fields, including, for example, automobiles, medical equipment, electrics and electronics, solar cells, fuel cells, human body replacement parts (such as artificial joints), and semiconductors.