Exhaust gas conversion process and catalyst

Page 1 — bibliographic record

Page 2

with the aid of a suitable binder, e.g., 5-20 weight-percent EXAMPLES 1-3 of a gelatinous, acid-peptized alumina. Other conven tional binders such as clays and the like may also be Three unsupported catalysts of the present invention utilized. The granular composites so obtained are then were prepared by coprecipitation of Cu(OH)2 and dried and calcined as described hereinafter.

The coprecipitation technique however is not essential. Fe(OH) Fully equivalent results have been obtained by mixing the with sodium hydroxide from a solution of the nitrate salts. powdered metal oxides, mulling with peptized alumina or The coprecipitation was carried out by metering a stream other suitable binder, extruding, drying and calcining.

Other compounds, such as the metal carbonates, oxalates, IO of the salt solution and a stream of NaOH solution into the inlet of a centrifugal pump. By this means a very rapid nitrates, acetates, or similar salts whose anions can be mixing of the solutions at a constant pH of about 7.5 was readily decomposed during calcination, can also be used readily accomplished. The gelatinous precipitate was in preparing the catalyst by the co-mulling technique.

While highly active catalysts can be prepared by the filtered, dried at 220 F., and washed free of sodium above techniques, the resulting essentially undiluted com (0.01%) by a stream of deionized water flowing upward posites generally have an undesirably high density. Low through the granular solid. The washed product was dried density composites of substantially equal activity may be and pulverized through a 100-mesh screen. The powder prepared for example by grinding the comulled or copre was then thoroughly mulled with enough gelatinous acid cipitated metal compounds to micron or sub-micron size peptized alumina to give 10 percent by weight of alumina and then coextruding or copelleting the mixture with rela 20 in the final product. The mulled mix was extruded tively large amounts of alumina or other inert matrices through a 46' die, dried, and calcined for 3 hours at such as clays, silica gel, silica-alumina, and the like. 1100, 1400 and 1800° F. (separate portions). The Precalcined composites of iron oxide and copper oxide finished catalysts contained about 10% AlO3, 45% CuO may similarly be powdered and coextruded or copelleted and 45% Fe2O3 by weight.

The three catalysts were then activity tested for NO with such inert matrices. By either of these methods, the 25 conversion, final composite is again dried and calcined. using a synthetic exhaust gas composed as Suitable low density catalysts may also be prepared by follows:

conventional impregnation methods wherein a porous, in Mole percent ert support such as Carborundum, alpha alumina, delta CO--------------------------------- 1.0 alumina, mullite, aluminum phosphate or the like, in suit 30 C3H6 H2 --------------------------------- 0.33 able granular or monolithic form, is impregnated either ------------------------------- 0.10 sequentially or simultaneously with an aqueous solution NO -------------------------------- 0.08 or solutions of the nitrates or other soluble and heat-de HaO -------------------------------- 10.0 composable compounds of copper and iron, followed by 8.2 (O2) ---------------------------- s (0.3) draining, drying and calcining of the impregnated support. - - - - - - - - - - - -- more m- - - - - - - - - - - - - - - 13.0

The metal content of the impregnated catalysts may range between about 1 and 20 percent, preferably 4-10 percent 100.00 of copper as CuO, and between about 1 and 20 percent, preferably 4-10 percent of iron as Fe2O3. Both in the The test procedure consisted in passing the feed gas supported and unsupported catalyst compositions the 40 through the catalyst bed at a gaseous hourly space velocity weight ratio of Cu/Fe, calculated as CuO and FeO, is of 23,000, measuring NO conversion at about 1000 F. between about 1/5 and 5/1, preferably between about (which generally gives 100% conversion), then at suc 4/6 and 6/4. It is found that a mole excess of CuO over cessively lower temperatures so as to bracket the 50% the stoichiometric ratio in the spinel, CuFeO4, is desir conversion temperature and obtain temperature coeffi able to achieve maximum NO conversion activity. Al 45 cients. From this the 50% conversion temperatures were though the metal contents and ratios have been expressed calculated, based on the first-order rate equation. The re herein in terms of oxides, it is not to be concluded that Sults were as follows:

those oxide forms are the active species in the final cata lyst. It would appear in fact that the lower valent oxides,

CuO and Fe0, or even the respective free metals, may 50 TABLE be at least in part the active species.

The dried catalyst composites, prepared by any of the Temp., F. above methods, are then calcined for about 0.5 to 12 Calcination

Conversion hours or more at temperatures ranging between 1200 and Catalyst temp., F. of NO 2400 F., preferably about 1600-2000 F. The calcining 55 A--- 1,100 11,014 is normally carried out in air (static or flowing), but may B-. 1,400 536 be effected in any suitably inert atmosphere. Conventional C--- 1,800 488 techniques may be utilized, e.g., heating in a rotary 30% conversion. forced-air furnace, passing hot gases through a static bed of the catalyst, etc. Preferably the heat-up to calcination 60 temperature should be somewhat gradual over a period of It will be seen that the catalyst calcined at 1100 F. was e.g., /2 to 1 hour. Time and calcination temperature markedly less active than the other two, and the catalyst should be controlled and correlated so as to achieve a calcined at 1800" F. was substantially more active than substantial, but preferably incomplete, formation of a crys the 1400 F. calcined composition. talline spinel phase, as determined by X-ray diffraction 65 analysis. EXAMPLES 4-15

Following the calcination, the catalysts may if desired be subjected to reduction with hydrogen or carbon monox Twelve additional catalysts containing varying propor ide at e.g., 700-1000 F., although this procedure nor 70 tions of CuO and Fe2O3 were prepared by the same gen mally offers no real advantage over placing the calcined eral methods described in Examples 1-3, except that the catalyst directly on-stream for in-situ reduction by the highest temperature calcining operation was carried out under somewhat less severe conditions (utilizing a dif exhaust gases. ferent furnace), resulting in the formation of spinel phases The following examples are cited to illustrate the in which were not so sharply crystalline as those formed in vention, but are not to be construed as limiting in scope. 75 Example 3 (catalyst C). Upon activity testing these cata

Page 3

lysts as described in Examples 1-3, the following results I claim:

were obtained: 1. A method for the conversion of nitrogen oxides in TABLE 2 engine exhaust gases, which comprises contacting said ex VP haust gases at elevated temperatures with a catalyst in

Catalystit (Ex.

composition, 10 wt. wt. which the active component is essentially an intimately ASSE) percent St. TEp., admixed, calcined composite of copper oxide and iron ox

C&R. ide in which the weight ratio of Cu/Fe is between about - - - - 6/4 and 4/6, as CuO/Fe0, said composite having been g 8 SS 5 calcined at temperatures between about 1600 and 2000 67 33 1.66 542 F. for a sufficient time to form a spinel phase.

: E. : 2. A method as defined in claim 1 wherein the weight 50 50 1866 330 ratio of Cu/Fe in said composite is about 1/1, as CuO/

y 1,946,964. 2/1934 Cobb as a post as a ma - a - w w 423-594X

It will be seen that the 1800 F. calcined CuO-Fe2O3 3.299; 4:2: Sile et al - re- a or 121: mixtures were substantially more active than either CuO 2370443 2A1945 Blefeld ? um am m mwar as a423 594 X or FeO- alone. Also, the8 1800 F. calcined composites 20 3,544,264

were significantly more active than the corresponding com- 3,476,508 11/1969 Kearby et al. 423-213 posites calcined at 1400 F. In the case of pure Fe0, the activities were reversed for the 1400 and 1800 calcina

tions. A rather sharp maximum in activity appears to GEORGE. O. PETERS, Primary Exami occur at the 50/50 proportion of CuO and FeO (cata- 25 , Primary Examiner lyst I). 8. 4. U.S. C. X.R. The following claims and their obvious equivalents are 252-474 intended to define the true scope of the invention.