Process for cleaning harmful gas

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a process for cleaning a harmful gas. More particularly, it pertains to a process for cleaning a nitrogen fluoride such as nitrogen trifluoride which is used or generated in the semiconductor manufacturing industry.
With the continuous development of the semiconductor industry, there has been a steady rise in recent years in the amount of nitrogen trifluoride which is used in the dry etching of silicon and silicon oxide, as a gas for cleaning the chamber of a CVD apparatus and the like. Nitrogen trifluoride gas is sparingly soluble in water and rather stable at room temperature with little reactivity with an acid or an alkali, but it is highly toxic and exerts an adverse influence on human bodies and the environment since the maximum permissible concentration thereof in the atmosphere is reported as being 10 ppm. It is therefore, necessary to clean a nitrogen trifluoride-containing gas after being used in a semiconductor production process prior to the discharge in the atmosphere.
In spite of its being stable at ordinary temperature, nitrogen trifluoride generates dinitrogen tetrafluoride, dinitrogen difluoride, dinitrogen hexafluoride, fluorine and the like due to heat, discharge, etc. in etching and cleaning processes, and each of them must be removed as well because of its toxicity is stronger than that of nitrogen trifluoride.
2. Description of the Related Art
As the method of removing a nitrogen fluoride contained in a mixed gas, there have heretofore been proposed (1) a process wherein the gas is brought into contact with metallic silicon at 100.degree. C. or higher (Japanese Patent Application Laid-Open No. 12322/1988), (2) a process wherein the gas is brought into contact with metallic titanium at 200.degree. C. or higher (Japanese Patent Publication No. 48571/1988), (3) a process wherein the gas is brought into contact with Si, B, W, Mo, V, Se, Te, Ge or a non-oxide based compound of any of them at 200.degree. to 800.degree. C. (Japanese Patent Publication No. 48570/1988), (4) a process wherein the gas is brought into contact with a metal halogenide capable of transhalogenation with nitrogen fluoride (Japanese Patent Publication No. 48569/1988), (5) a process wherein the gas is brought into contact with the oxide of a transition metal such as Fe, Mn or Cu at 250.degree. C. or higher (Japanese Patent Application Laid-Open No. 181316/1991), (6 ) a process wherein the gas is brought into contact with activated carbon at 300.degree. to 600.degree. C. (Japanese Patent Application Laid-Open No. 237929/1987), (7) a process wherein the gas is brought into contact with a composition comprising as principal components Ni, Fe, Co or a noble metal such as Pt, Rh and Pd and at least one from alumina and silica at 200.degree. C. or higher (Japanese Patent Application Laid-Open No. 27303/1987).
Nevertheless, any of the above-proposed processes is insufficient in gas cleaning capability and besides suffers the disadvantage as described hereunder. A volatile fluoride is produced in the processes (1), (2) & (3), a halogen such as chlorine is formed in, the process (4) and a nitrogen oxide is generated in the process (5), each incurring an expense in the treatment of itself. It is necessary in the processes (2) & (3) to heat the reaction system to 300.degree. C. or higher in order that the produced fluoride may not cover the surface of the reaction agent and thus hinder the cleaning reaction. The process (6) involves the danger of explosion due to a violent reaction at an elevated temperature and the problem of byproducing carbon tetrafluoride that is relatively stable and difficult to remove. In the process (7), a harmful gas is not byproduced but a high temperature is required for attaining sufficient cleaning capability and in the case of nickel (Ni), for example, for the purpose of achieving practical performance, the reaction system needs to be heated to 400.degree. C. or higher, that is, much higher than 200 .degree. C. at which the decomposition activity is insufficient and besides, the surface of the reaction agent is covered with the fluoride with the progress of the reaction, thereby failing to assure sufficient cleaning capability.
SUMMARY OF THE INVENTION
Under such circumstances, intensive research and investigation were continued by the present inventors on the development of a process for cleaning nitrogen fluoride with high treatment capacity at a low temperature without producing a harmful gas or a gas having a possibility of causing environmental pollution while solving the disadvantages of the conventional techniques. As a result, it has been found by the present inventors that the nitrogen fluorides can be removed in extremely high efficiency at a relatively low temperature without generating a substance adversely affecting the environment in the exhaust gas after a cleaning treatment by using zirconium or a zirconium-based alloy as the cleaning agent. The present invention has been accomplished on the basis of the above-mentioned finding.
Specifically the present invention provides a process for cleaning a gas containing a nitrogen fluoride as the harmful component which comprises bringing said gas under heating into contact with a cleaning agent comprising zirconium to remove said harmful component and a process for cleaning a gas containing a nitrogen fluoride as the harmful component which comprises bringing said gas under heating into contact with a cleaning agent comprising a zirconium-based alloy to remove said harmful component.
According to the present invention, the nitrogen fluorides including nitrogen trifluoride which is contained in the air, nitrogen, argon and hydrogen and other nitrogen fluoride such as dinitrogen tetrafluoride, dinitrogen difloride, dinitrogen hexafloride can efficiently be removed.
DESCRIPTION OF PREFERRED EMBODIMENTS
As the cleaning agent according to the present invention, there is employed zirconium or a zirconium-based alloy. As the zirconium source applicable to the preparation of the cleaning agent according to the present invention, mention may be made of metallic zirconium and sponge zirconium each available on the market, which can be used as such or by crushing them into an appropriate size. The zirconium available on the market sometimes contains hafnium in an amount of about 1 to 5% by weight, but such extent of hafnium content does not exert evil influence on the cleaning capability and is not liable to produce a harmful substance such as a volatile fluoride or nitrogen oxide in the course of cleaning reaction.
The zirconium-based alloy as mentioned above is usually the alloy of zirconium and at least one metallic element selected from the group consisting of iron, copper, nickel, aluminum, magnesium, calcium, zinc, lanthanum, cerium, vanadium, molybdenum, titanium, chromium, tungsten, tantalum, cadmium, yttrium, niobium and tin. The preferable alloy among them includes that of zirconium and at least one element selected from iron, copper, nickel, aluminum, magnesium, calcium, zinc, lanthanum and cerium, since it does not generate a volatile fluoride during,the course of reaction with a nitrogen fluoride and is easily available at a relatively low cost.
The nitrogen-fluoride removing function in the present invention is attributable principally to the zirconium component, which function is corroborated by the fact that the powdery substance formed during the cleaning operation as the reaction product of the cleaning agent and a nitrogen fluoride is proved to be mainly zirconium fluoride by the analysis of the substance. Accordingly, the cleaning agent consisting of zirconium as simple substance is characterized by its high cleaning capacity.
In addition, the cleaning capacity per unit weight of the cleaning agent consisting of a zirconium-containing alloy increases with an increase in the content of zirconium in the alloy. However, the alloying metallic component other than zirconium exhibits the effect on lowering the temperature at which a nitrogen fluoride is removed, and such effect is remarkably enhanced by an appropriate content of said component in the alloy. In more detail, the zirconium as simple substance necessitates a temperature of about 300.degree. C. for the purpose of attaining practical cleaning capability, whereas the zirconium-containing alloy enables a gas containing a nitrogen fluoride in the same concentration to be treated at a temperature lower than 300.degree. C. at the same flow rate. For example, practical cleaning capability is achieved at a lower temperature of 170.degree. to 250.degree. C. by appropriately selecting the content the alloying metal other than zirconium at 40% or less by weight.
The content of zirconium in the cleaning agent according to the present invention is not specifically limited, but in the case of zirconium-containing alloy, it is usually 20% or more, desirably 50% or more, more desirably 60% or more by weight with the above-mentioned metallic element as the balance. A content thereof less than 20% by weight results in insufficient capacity of removing nitrogen fluoride and decrease in the effect on lowering the cleaning temperature by alloying and besides causes a fear of byproducing a volatile fluoride depending upon the cleaning conditions.
The zirconium-based alloy can be produced by blending zirconium and the above-mentioned at least one metallic element at a prescribed blending ratio and subsequently alloying the resultant blend through electron beam melting, argon arc melting, high frequency heating melting or resistance heating melting each under vacuum or in an atmosphere of an inert gas or the like method of melting. The alloy thus produced is crushed to 6 to 20 mesh by mechanical crushing by means of a ball mill, jaw crusher, roll mill or the like to be employed as the cleaning agent. Alternatively, it is pulverized to about 100 mesh fine powder, which is made into granules or granulated powder having a size about 1 to 5 mm or molded into pellets to be employed as the cleaning agent. Various zirconium-based alloys available on the market may be employed as it is or after crushing to a suitable size.
The cleaning agent comprising zirconium or the zirconium-based alloy according to the present invention can be used as any of fixed bed, moving bed and fluidized bed. Under ordinary circumstances, the cleaning agent is packed in a cleaning column, and the gas containing a nitrogen fluoride is passed therethrough while being brought into contact with the cleaning agent so that the nitrogen fluoride as the harmful component is removed so as to clean the gas.
The temperature at which the gas to be treated is brought into contact with the cleaning agent (cleaning temperature) is usually 100.degree. to 800.degree. C., desirably 150.degree. to 500.degree. C., more desirably 200.degree. to 350.degree. C. in case of zirconium, and 150.degree. to 300.degree. C. in case of a Zirconium -based alloy. A temperature lower than 100.degree. C. results in deterioration of the capability of removing nitrogen fluoride, whereas that high than 800.degree. C. brings about the disadvantages that stainless steel can not be used for the cleaning column, thereby lowering safety and besides increasing heating energy loss.
In the case where an atmospheric component is mixed in the gas to be treated, a high cleaning temperature is liable to cause heat release due to the reaction of the cleaning agent with oxygen and therefore, the cleaning operation is put into practice preferably at 250.degree. C. or lower.
The pressure during the cleaning operation is usually atmospheric pressure, but can be reduced pressure or raised pressure such as 1 kg/cm.sup.2 G.
There is no limitation to the flow velocity of the gas to be treated to which the cleaning process according to the present invention is applied, but in general the flow velocity is desirably lowered with increase in the concentration of the nitrogen fluoride contained in the gas to be treated.
The cleaning column is designed in accordance with the concentration of the nitrogen fluoride as the harmful gas, the amount and flow rate of the gas to be treated, etc. The space linear velocity (LV) in the column is preferably designed at 20 cm/sec or lower for a relatively low concentration of nitrogen fluoride such as 1000 ppm or less and at 5 cm/sec or lower for the concentration higher than 1000 ppm.
The length of the packed cleaning agent in the cleaning column varies depending on the flow rate of the gas to be treated, the concentration of the harmful gas and the like and can not be unequivocally specified, but is usually 50 to 500 mm, approximately from the practical viewpoint. In general, the length thereof is determined in accordance with the pressure loss through the packed bed, contact efficiency of the gas with the cleaning agent:, the concentration of the harmful gas and the like.
As mentioned hereinbefore, thereaction during the cleaning operation according to the present invention causes zirconium fluoride to be formed in the form of powder, which is discharged outside the system along with the treated gas or dropped to the bottom of the cleaning column, or allowed to remain in part in the packed zone of the cleaning agent depending on the operational conditions, whereby the pressure loss through the zone is unfavorably increased. Such increase in the pressure loss can be prevented by imparting vibration to the column continuously or intermittently by means of a vibrator or the like fixed to the column to drop the powder down to the bottom of the column. When necessary, a filter for collecting the powder in the treated gas may be installed on the downstream side of the cleaning column.
In summary, according to the process for cleaning a harmful gas of the present invention, nitrogen fluoride such as nitrogen trifluoride contained in a gas can efficiently be removed at a relatively low temperature without producing a harmful byproduct such as a nitrogen oxide. The process, therefore, exhibits an excellent effect on the cleaning of exhaust gas from a semiconductor manufacturing process or the like.
In the following the present invention will be described in more detail with reference to the non-limitative examples and comparative examples.

EXAMPLE 1
84.9 ml of sponge zirconium that was procured from the market (prepared by screening crushed product to 6 to 32 mesh, with 99.5% purity) was packed in a quartz-made cleaning column having 19 mm inside diameter and 400 mm length. Then, helium (He) gas containing 1% nitrogen trifluoride (NF.sub.3) was passed through the column at a total flow rate of 170 ml/min, that is, a space linear velocity (LV) of 1 cm/sec at 250.degree. C. under atmospheric pressure, and the column outlet gas was analyzed for NF.sub.3 by gas chromatography (lower limit of detectable range being 10 ppm). As a result, NF.sub.3 was not detected. Subsequently, the gas was passed through the column for further 3 hours'. As a result, there was not observed break through nor the formation of harmful byproduct such as NO.sub.x.
EXAMPLE 2
28.3 ml of the sponge zirconium same as that used in Example 1 was packed in a quartz-made cleaning column having 19 mm inside diameter and 400 mm length. Then, He gas containing 1% NF.sub.3 was passed through the column at a total flow rate of 85 ml/min, that is, a space linear velocity (LV) of 0.5 cm/sec at room temperature under atmospheric pressure, and after 20 minutes the column outlet gas was analyzed to determine NF.sub.3 concentration by gas chromatography (lower limit of detectable range being 10 ppm). Subsequently, the gas temperature was raised by increments of 50.degree. C., while the temperature was maintained for 10 minutes at every increment to determine NF.sub.3 concentration in the column outlet gas by gas chromatography and thereby obtain NF.sub.3 decomposition efficiency at each incremental temperature. By plotting the decomposition efficiency thus obtained against the temperature, the lower limit of the temperature at which the NF.sub.3 decomposition efficiency exceeded 90% was obtained by means of interpolation. The result was 280.degree. C.
Thereafter, 8.5 ml of the same sponge zirconium as the cleaning agent was further packed in the column. Subsequently He gas was passed through the column at a total flow rate of 500 ml/min, while the gas temperature was raised to 300.degree. C. Thereafter, He gas containing 2% NF.sub.3 was passed through the column at a total flow rate of 509 ml/min, that is, a LV of 3 cm/sec at 300.degree. C. under atmospheric pressure, while the outlet gas was monitored by means of an NF.sub.3 detector available on the market (TG-4100 TA, produced by Bionics Instruments Co., Ltd.) to determine the break through time by regarding the point at which NF.sub.3 concentration in the outlet gas reached 10 ppm as the break through point.
The cleaning capability (throughput of NF.sub.3 (L) per 1L of Zr) was obtained by calculation from the result thus obtained. The result was 711L/L. In addition in order to check byproduct formation, nitrogen monoxide (NO) and nitrogen dioxide (NO.sub.2) in the exhaust gas were measured prior to the break through by means of the detecting tubes (for separating and analyzing nitrogen oxides, lower limit of detectable range being 1 ppm for NO and 0.5 ppm for NO.sub.2, produced by Gastech Corp.). As a result, nitrogen oxide was not detected.
EXAMPLE 3
28.3 ml of cleaning agent comprising Zr-Fe alloy procured from the market (80% by weight of Zr with the balance Fe) which had been crushed and screened to 10 to 32 mesh was packed in a quartz-made cleaning column having 19 mm inside diameter and 400 mm length. Then He gas containing 1% NF.sub.3 was passed through the column at a total flow rate of 85 ml/min, that is, a space linear velocity (LV) of 0.5 cm/sec at room temperature under atmospheric pressure, and after 20 minutes the column outlet gas was analyzed to determine NF.sub.3 concentration by gas chromatography (lower limit of detectable range being 10 ppm). Subsequently, the gas temperature was raised by increments of 50.degree. C., while the temperature was maintained for 10 minutes at every increment to determine NF.sub.3 concentration in the column outlet gas by gas chromatography and thereby obtain NF.sub.3 decomposition efficiency at each incremental temperature. By plotting the decomposition efficiency thus obtained against the temperature, the lower limit of the temperature at which the NF.sub.3 decomposition efficiency exceeded 90% was obtained by means of interpolation. The result is given in Table 1.
Thereafter, 8.5 mol of the same alloy as the cleaning agent was further packed in the column. Subsequently He gas was passed through the column at a total flow rate of 500 ml/min, while the gas temperature was raised to 200.degree. C. Thereafter, He gas containing 2% NF.sub.3 was passed through the column at a total flow rate of 509 ml/min, that is, a LV of 3 cm/sec under atmospheric pressure, while the outlet gas was monitored by means of an NF.sub.3 detector available on the market (TG-4100 TA, produced by Bionics Instruments Co., Ltd.) to determine the break through time by regarding the point at which NF.sub.3 concentration in the outlet gas reached 10 ppm as the break through point.
The cleaning capability (throughput of NF.sub.3 (L) per 1L of Zr-alloy) was obtained by calculation from the result thus obtained. In addition in order to check byproduct formation, nitrogen monoxide (NO) and nitrogen dioxide (NO.sub.2) in the exhaust gas were measured prior to the break through by means of the detecting tubes (for separating and analyzing nitrogen oxides, lower limit of detectable range being 1 ppm for NO and 0.5 ppm for NO.sub.2, produced by Gastech Corp.). The results are given in Table 2.
EXAMPLE 4 TO 6
Sponge zirconium and reduced iron each procured from the market in a total amount of about 500 g were blended in different compositions (90%, 50% or 40% by weight of Zr and the balance iron) and then melted by electron beam twice repeatedly to afford Zr-Fe alloys, which were crushed in a ball mill and screened to 14 to 20 mesh to prepare various alloys each having a different composition from one another.
Subsequently, the procedure in Example 3 was repeated to carry out cleaning experiment except that each of the above-prepared alloys was used as the cleaning agent and the cleaning temperature was altered, respectively. The results are given in Table 1 and Table 2.
COMPARATIVE EXAMPLE 1
The procedure in Example 3 was repeated to carry out cleaning experiment except that there was used as the cleaning agent iron wires procured from the market and cut into a length of 5 to 10 mm. The results are given in table 1 and Table 2.
TABLE 1______________________________________ Alloy compo- Lower limit of temperature (.degree.C.) sition (wt %) attaining .gtoreq.90% decomposition Fe Zr efficiency______________________________________Example 3 20 80 195Example 4 10 90 195Example 5 50 50 245Example 6 60 40 250Comparative 100 0 325Example 1______________________________________
TABLE 2______________________________________ Cleaning Cleaning capability Nitrogen temperature (.degree.C.) (L/L) oxide______________________________________Example 3 200 594 NDExample 4 200 >100 NDExample 5 250 >100 NDExample 6 280 >100 NDComparative 300 3 NDExample 1 400 7 ND______________________________________
EXAMPLES 7 TO 15
Sponge zirconium and copper or silver each procured from the market in a total amount of about 500 g were blended in different compositions (40%, 50%, 75% or 90% by weight of Zr and the balance copper or silver) and then melted by an electron beam twice repeatedly to afford Zr-Cu or Zr-Ag alloys, which were crushed in a ball mill and screened to 14 to 20 mesh to prepare various alloys each having a different composition. Then, 28.3 ml of each of the alloys was packed in a quartz-made cleaning column having 19 mm inside diameter and 400 mm length. Then, helium (He) gas containing 1% NF.sub.3 was passed through the column at a total flow rate of 85 ml/min, that is, a space linear velocity (LV) of 0.5 cm/sec at room temperature under atmospheric pressure, and after 20 minutes the column outlet gas was analyzed to determine NF.sub.3 concentration by gas chromatography (lower limit of detectable range being 10 ppm). Subsequently, the gas temperature was raised by increments of 100.degree. C., while the temperature was maintained for 10 minutes at every increment to determine NF.sub.3 concentration in the column outlet gas by gas chromatography and thereby obtain NF.sub.3 decomposition efficiency at each incremental temperature. By plotting the decomposition efficiency thus obtained against the temperature, the lower limit of the temperature at which the NF.sub.3 decomposition efficiency exceeded 90% was obtained by means of interpolation. The result is given in Table 3.
Thereafter, 8.5 ml of each of the alloys as the cleaning agent was further packed in a column. Subsequently He gas was passed through the column at a total flow rate of 500 ml/min, while the gas temperature was raised to the temperature as shown in Table 2. Thereafter, He gas containing 2% NF.sub.3 was passed through the column at a total flow rate of 509 ml/min, that is, a LV of 3 cm/sec under atmospheric pressure, while the outlet gas was monitored by means of an NF.sub.3 detector available on the market (TG-4100 TA, produced by Bionics Instruments Co., Ltd.) to determine the break through time by regarding the point at which NF.sub.3 concentration in the outlet gas reached 10 ppm as the break through point.
The cleaning capability (throughput of NF.sub.3 (L) per 1L of Zr-alloy) was obtained by calculation from the result thus obtained. In addition in order to check byproduct formation, nitrogen monoxide (NO) and nitrogen dioxide (NO.sub.2) in the exhaust gas were measured prior to the break through by means of the detecting tubes (for separating and analyzing nitrogen oxides, lower limit of detectable range being 1 ppm for NO and 0.5 ppm for NO.sub.2, produced by Gastech Corp.). The result are given in Table 4.
COMPARATIVE EXAMPLES 2 TO 3
The procedure in Example 3 was repeated to carry out cleaning experiment except that there were used as the copper source, copper wires procured from the market and cut into a length of 5 to 10 mm. The results are given in Table 3 and Table 4.
TABLE 3______________________________________ Alloy compo- Lower limit of temperature (.degree.C.) sition (wt %) attaining .gtoreq.90% decomposition Zr Cu Ag efficiency______________________________________Example 7 40 60 0 235Example 8 50 50 0 235Example 9 75 25 0 220Example 10 90 10 0 225Example 11 40 0 60 260Example 12 50 0 50 255Example 13 75 0 25 235Example 14 90 0 10 250Example 15 75 15 10 225Comparative 0 100 0 335Example 2Comparative 0 0 100 430Example 3______________________________________
TABLE 4______________________________________ Cleaning Cleaning capability Nitrogen temperature (.degree.C.) (L/L) oxide______________________________________Example 7 280 >100 NDExample 8 250 >100 NDExample 9 230 568 NDExample 10 230 >100 NDExample 11 280 >100 NDExample 12 280 >100 NDExample 13 250 >100 NDExample 14 260 >100 NDExample 15 250 >100 NDComparative 350 19 NDExample 2Comparative 500 8 NDExample 3______________________________________
EXAMPLES 16 TO 29
Sponge zirconium and nickel, cobalt or manganese each procured from the market in a total amount of about 500 g were blended in different compositions (40%, 50%, 75% or 90% by weight of Zr and the balance Ni, Co or Mn) and then melted by an electron beam twice repeatedly to afford Zr-Ni, Zr-Co or Zr-Mn alloys, which were crushed in a ball mill and screened to 14 to 20 mesh to prepare various alloys as cleaning agent each having a different composition from one another. Then 28.3 ml of each of the alloys was packed in a quartz-made cleaning column having 19 mm inside diameter and 400 mm length. Each of the Zr-alloys as the cleaning agent was tested in the same manner as in Example 7 to determine the temperature attaining 90% decomposition efficiency, cleaning capability and nitrogen oxide concentration. The results are given in Table 5.
COMPARATIVE EXAMPLES 4 TO 6
The procedure in Example 16 was repeated to carry out cleaning experiment except that there was used as the cleaning agent, Ni, Co or Mn procured from the market and screened to 6 to 32 mesh. The results are given in Table 5 and Table 6.
TABLE 5______________________________________ Lower limit of tem- perature (.degree.C.) at- taining .gtoreq.90% Alloy composition (wt %) decomposition Zr Ni Co Mn efficiency______________________________________Example 16 40 60 0 0 240Example 17 50 50 0 0 220Example 18 75 25 0 0 205Example 19 90 10 0 0 205Example 20 40 0 60 0 260Example 21 50 0 50 0 240Example 22 75 0 25 0 225Example 23 90 0 10 0 225Example 24 40 0 0 60 265Example 25 50 0 0 50 245Example 26 75 0 0 25 245Example 27 90 0 0 10 240Example 28 75 15 10 0 210Example 29 90 5 0 5 215Comparative 0 100 0 0 290Example 4Comparative 0 0 100 0 315Example 5Comparative 0 0 0 100 340Example 6______________________________________
TABLE 6______________________________________ Cleaning Cleaning capability Nitrogen temperature (.degree.C.) (L/L) oxide______________________________________Example 16 280 >100 NDExample 17 250 >100 NDExample 18 210 504 NDExample 19 210 >100 NDExample 20 280 >100 NDExample 21 250 >100 NDExample 22 240 >100 NDExample 23 240 >100 NDExample 24 280 >100 NDExample 25 280 >100 NDExample 26 250 >100 NDExample 27 250 >100 NDExample 28 210 >100 NDExample 29 215 >100 NDComparative 300 17 NDExample 4Comparative 330 10 NDExample 5Comparative 350 11 NDExample 6______________________________________
EXAMPLES 30 TO 51
Sponge zirconium and magnesium, calcium, zinc, aluminum, lanthanum or cerium each procured from the market in a total amount of about 500g were blended in different compositions (40%, 50%, 75% or 90% by weight of Zr and the balance Mg, Ca, Zn, Al, La or Ce) and then melted by an electron beam twice repeatedly to afford Zr-Mg, Zr-Ca, Zr-Zn, Zr-Al or Zr-Ce alloys, which were crushed in a ball mill and screened to 14 to 20 mesh to prepare various alloys as cleaning agent each having a different composition from one another. Each of the cleaning agents was tested in the same manner as in Example 7 to determine the temperature attaining 90% decomposition efficiency, cleaning capability and nitrogen oxide concentration. The results are given in Table 7 and Table 8.
COMPARATIVE EXAMPLES 7 TO 11
The procedure in Example 30 was repeated to carry out cleaning experiment except that there was used as the cleaning agent, sand magnesium, granular calcium, granular zinc (1 to 2 mm size), granular aluminum (2 to 3 mm size) or granular cerium (1 to 2 mm size). The results are given in Table 7 and Table 8.
TABLE 7______________________________________ Alloy compo- Lower limit of temperature sition (wt %) (.degree.C.) attaining .gtoreq.90% de- Zr other metal composition efficiency______________________________________Example 30 40 Mg 60 225Example 31 50 Mg 50 225Example 32 75 Mg 25 190Example 33 90 Mg 10 195Example 34 40 Ca 60 230Example 35 50 Ca 50 230Example 36 75 Ca 25 200Example 37 90 Ca 10 210Example 38 40 Zn 60 245Example 39 50 Zn 50 215Example 40 75 Zn 25 205Example 41 90 Zn 10 200Example 42 40 Al 60 240Example 43 50 Al 50 240Example 44 75 Al 25 215Example 45 90 Al 10 210Example 46 40 La 60 245Example 47 50 La 50 240Example 48 75 La 25 230Example 49 90 La 10 240Example 50 40 Ce 60 235Example 51 50 Ce 50 245Example 52 75 Ce 25 215Example 53 90 Ce 10 225Example 54 75 Mg 15 Al 10 195Example 55 90 Mg 5 Zn 5 195Comparative 0 Mg 100 270Example 7Comparative 0 Ca 100 280Example 8Comparative 0 Zn 100 340Example 9Comparative 0 Al 100 465Example 10Comparative 0 La 100 320Example 11Comparative 0 Ce 100 310Example 12______________________________________
TABLE 8______________________________________ Cleaning Cleaning capability Nitrogen temperature (.degree.C.) (L/L) oxide______________________________________Example 30 250 >100 NDExample 31 250 >100 NDExample 32 200 404 NDExample 33 200 >100 NDExample 34 250 >100 NDExample 35 250 >100 NDExample 36 210 389 NDExample 37 220 >100 NDExample 38 280 >100 NDExample 39 250 >100 NDExample 40 230 381 NDExample 41 230 >100 NDExample 42 280 >100 NDExample 43 250 >100 NDExample 44 230 495 NDExample 45 230 >100 NDExample 46 260 >100 NDExample 47 260 >100 NDExample 48 250 406 NDExample 49 260 >100 NDExample 50 250 >100 NDExample 51 250 >100 NDExample 52 230 >100 NDExample 53 230 >100 NDExample 54 210 >100 NDExample 55 210 >100 NDComparative 300 16 NDExample 7Comparative 300 11 NDExample 8Comparative 400 25 NDExample 9Comparative 500 9 NDExample 10Comparative 350 12 NDExample 11Comparative 350 19 NDExample 12______________________________________
EXAMPLES 56 TO 88
Sponge zirconium and vanadium, molybdenum, titanium, chromium, tungsten, tantalum, niobium or tin each procured from the market in a total amount of about 500 g were blended in different compositions (40%, 50%, 70% or 90% by weight of Zr and the balance V, Mo, Ti, Cr, W, Ta or Nb or 50%, 70% or 90% by weight of Zr and the balance Sn)O and then melted by electron beam twice repeatedly to afford Zr-V, Zr-Mo, Zr-Ti, Zr-Cr, Zr-W, Zr-Ta, Zr-Nb or Zr-Sn alloys, which were crushed in a ball mill and screened to 14 to 20 mesh to prepare various alloys as cleaning agent each having a different composition from one another. Then 28.3 ml of each of the alloys was packed in a quartz-made cleaning column having 19 mm inside diameter and 400 mm length. Each of the Zr-alloys as the cleaning agent was tested in the same manner as in Example 7 to determine the temperature attaining 90% decomposition efficiency, cleaning capability and nitrogen oxide concentration. The results are given in Table 10.
COMPARATIVE EXAMPLES 13 TO 20
The procedure in Example 56 was repeated to carry out cleaning experiment except that there was used as the cleaning agent, V, Mo, Ti, Cr, W, Ta, Nb or Sn (1 to 2 mm in granule size) procured from the market. The results are given in Table 9 and Table 10, but the cleaning capability of Sn granule was not measured because of its failure to achieve 90% decomposition efficiency.
TABLE 9-1______________________________________Alloy compo- Lower limit of temperature (.degree.C.)sition (wt %) attaining .gtoreq.90% decompositionZr other metal efficiency______________________________________Example 56 40 V 60 235Example 57 50 V 50 205Example 58 70 V 30 170Example 59 90 V 10 175Example 60 40 Mo 60 250Example 61 50 Mo 50 190Example 62 70 Mo 30 180Example 63 90 Mo 10 180Example 64 40 Ti 60 205Example 65 50 Ti 50 195Example 66 70 Ti 30 165Example 67 90 Ti 10 170Example 68 40 Cr 60 230Example 69 50 Cr 50 235Example 70 70 Cr 30 200Example 71 90 Cr 10 205Example 72 40 W 60 195Example 73 50 W 50 185Example 74 70 W 30 180Example 75 90 W 10 185Example 76 40 Ta 60 185Example 77 50 Ta 50 180Example 78 70 Ta 30 175Example 79 90 Ta 10 190Example 80 40 Nb 60 225______________________________________
TABLE 9-2______________________________________ Alloy compo- Lower limit of temperature (.degree.C.) sition (wt %) attaining .gtoreq.90% decomposition Zr other metal efficiency______________________________________Example 81 50 Nb 50 195Example 82 70 Nb 30 175Example 83 90 Nb 10 185Example 84 50 Sn 50 200Example 85 70 Sn 30 195Example 86 90 Sn 10 200Example 87 70 V 15 Ti 15 160Example 88 90 V 5 Mo 5 175Comparative 0 V 100 415Example 13Comparative 0 Mo 100 395Example 14Comparative 0 Ti 100 285Example 15Comparative 0 Cr 100 435Example 16Comparative 0 W 100 450Example 17Comparative 0 Ta 100 425Example 18Comparative 0 Nb 100 430Example 19Comparative 0 Sn 100 --Example 20______________________________________
TABLE 10-1______________________________________ Cleaning Cleaning capability Nitrogen temperature (.degree.C.) (L/L) oxide______________________________________Example 56 250 >100 NDExample 57 220 >100 NDExample 58 180 >100 NDExample 59 180 >100 NDExample 60 280 >100 NDExample 61 200 >100 NDExample 62 190 >100 NDExample 63 190 >100 NDExample 64 220 >100 NDExample 65 200 >100 NDExample 66 180 >100 NDExample 67 180 >100 NDExample 68 250 >100 NDExample 69 250 >100 NDExample 70 230 >100 NDExample 71 230 >100 NDExample 72 220 >100 NDExample 73 200 >100 NDExample 74 190 >100 NDExample 75 190 >100 NDExample 76 210 >100 NDExample 77 200 >100 NDExample 78 200 >100 NDExample 79 200 >100 NDExample 80 250 >100 ND______________________________________
TABLE 10-2______________________________________ Cleaning Cleaning capability Nitrogen temperature (.degree.C.) (L/L) oxide______________________________________Example 81 210 >100 NDExample 82 190 >100 NDExample 83 190 >100 NDExample 84 220 >100 NDExample 85 220 >100 NDExample 86 220 >100 NDExample 87 180 >100 NDExample 88 190 >100 NDComparative 450 23 NDExample 13Comparative 450 28 NDExample 14Comparative 300 41 NDExample 15Comparative 450 10 NDExample 16Comparative 470 21 NDExample 17Comparative 450 22 NDExample 18Comparative 450 34 NDExample 19Comparative -- -- NDExample 20______________________________________

Number	Date	Country
4-298806	Nov 1992	JPX
4-325572	Dec 1992	JPX
4-332948	Dec 1992	JPX
4-332949	Dec 1992	JPX
4-355408	Dec 1992	JPX
4-355409	Dec 1992	JPX

Number	Name	Date
4339309	Pruett et al.	Jul 1982
4629611	Fan	Dec 1986
4960581	Harada et al.	Oct 1990
4983373	Withers, Jr. et al.	Jan 1991

Number	Date	Country
0365490	Apr 1990	EPX
0384802	Aug 1990	EPX
0431351	Jun 1991	DEX
4-122420	Apr 1992	JPX

Process for cleaning harmful gas

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

Priority Claims (6)

US Referenced Citations (4)

Foreign Referenced Citations (4)

Non-Patent Literature Citations (1)