CA1038599A

CA1038599A - Catalytic removal of nitrogen oxides from exhaust gases

Info

Publication number: CA1038599A
Application number: CA179,743A
Authority: CA
Inventors: James M. Maselli; Gwan Kim; Warren S. Briggs
Original assignee: WR Grace and Co
Current assignee: WR Grace and Co
Priority date: 1972-12-14
Filing date: 1973-08-28
Publication date: 1978-09-19
Also published as: JPS5635938B2; DE2360975A1; IT1001405B; FR2326973A1; GB1415757A; JPS5046586A

Abstract

AUTO EXHAUST CATALYST

Abstract of the Disclosure A catalyst that combines high activity and good stability for the conversion of nitrogen oxides to nitrogen in auto exhaust environments is described. The catalyst consists of a base metal component preferably nickel oxide combined with noble metals such as palladium, platinum, and rhodium, supported on a suitable base.

Description

BACKGROU ~ ~F ~HE IN~ENTION
This invention xelates to a method of combating air pollution and more specifically to a catalyst that is useful for removing one of the smog-forming components of the exhaust gases, nitrogen oxides. "
"Smog", as the term is generally app~lied is broadly understood to refer to a variety of phenomenon which are related to the intereaction of nitrogen oxides, hydrocarbons, and sunlight. These phenomenon include a og-like haze in the ,;
atmosphere, eye irritation, plant damage, and the like. The formation of excessive and objectionable levels of hydrocarbons in the atmosphere has been attributed to the incomplete oxidation of petroleum products, such as gasoline in internal combustion engines. Intereaction between the oxides of nltxogen and the unburned hydroaarbons that form smog requires ;;
sunlight. It has been found, for example, that when plants ~;;
are fumigated in the dark with various olefins and oxides of nitrogen at concentrations of about 1 part per million~ no damage is observed. However, when the same experiment is conduated in sunlight, typical damage patterns resuLt, ident~
¦ ical with those observed during a smog attack. In addition ¦ to plant damagej the formation of phototoxicants results in eye irritation as well.
Exhaust gases ~xom internal combustion engines have ~ ;
been demonstrated to be a primary cause of photochemical smog in heavy populated urban areas. The olefins and nitrogen oxide ;~
components of such yases have been demonstrated to be a ;-~
i principle cause of photoahemical smog.
It is well known that noble metals, suah as platinum, palladium, alone or in combinations with rhodium, ruthenium, and irridium can be used as active components for catalysts used i for the conversion of nitrogen oxides in exhaust gases to 1 ~ 2 ~

~385;9~
nitrogen. U.S. Patent 3,637,344, discusses one of these systems.
Z However, in order to be effective, a nitrogen oxide catalyst 3 must be able to function at conditions that are optimum for the above mentioned noble metals. For example, a catalyst must ` ;~
have good low temperature activityl must convert the nitrogen oxide to nitrogen and not ammonia, must be able to convert the nitroyen oxides to nitrogen under reaction conditions which vary ;~
, between net oxidizing and net reducing conditions. In addition, the catalyst must have good hydrothermal stability and perform after extended periods of time when exposed to temperatures as ~ high as 1800F.
i3 B f Description of the Invention ~Z We have found that combining nickel oxide with noble 3 metals o Group VIII results in a nitrogen oxide catalyst which Z exhibits greatly improved performance in two important areas.
The first is the thermal stability and the second is in conver- ;
sion of nitrogen oxides to nitrogen which is the desirable end product of the reaction rather than ammonia which the noble metal ;~
catalysts, when used alone, produce in substantial amounts.
Particularly good results are achieved when the nickel oxide-, noble metal catalyst systems is supported on acidic or neutral type supports in contrast to alkaline materials such as alumina I or magnesia.
Thus, in accordance with the present teachings, a process is provided for improving the high temperature conver-on of nitrogen oxides in exhaust gases of internal combustion enginqs. The gases are contacted with a catalyst containing from about 1 to 20 percent by weight of nickel oxide or a mixture of ~ nickel oxide with iron or cobalt oxides and from about 0.~001 to `~
Z 30 1 percent by weight of a noble metal distended on a suitable ', support, the percent by weight is based on the combined weight ` of the catalyst and support.
:~J ,~ ~

l ~3~
"

DESCRIPTION OF THE PREFERRED
~MBODIMENT ~;
,,, , . ~
- .
The first step in the preparation of our catalyst is 1 the selection of a suitable support. We have found that certain `~
acidic silica-aluminas such as mullite (3A12O3:2SiO2) or the , . . .
j well-known hydrocarbon cracking catalysts serve as particularIy ~t useful supports. The commonly available silica gels are also effective. However, other support materials such as any of -i the aluminas, silica-magnesias, zirconias, zirconia-silicasr i 10 zirconia-aluminas, titanlas, etc., can also be used. The , support can be in the form of nodules, pills, extrudate or ' ' ':

:-~ '' ': ;
- 20 ` ~

.. ~ ~, ., . - ~ .

: .;

.' .~ .' ' ' ~ ' "~lj : I :
-3a-:',~ . ' '~ -?' .', .

finely divided powder. The last named form is especially useful for coating monolithic catalyst ~ases for use in controlling `~
auto exhaust emissions. However, it has also been :hown 1;~
(Example 14) that the actlve components need not be distended ~i ,.. . .
on a catalyst support but can be applied directly to the monolith and still provide effective catalytic ac~vity.
The especially active and stable catalyst is made by ; combining any of the well known noble metal nitrogen oxide catalysts such as platinum, palladium, rhodium, and mixtures thereof with a base metal oxide such as nickel oxide. In some cases iron or cobalt oxides can be substituted for part of the nickel oxide. ~ ~;
After the support is selected, the ca-talyst is ~ prepared by applying the active components as solutions o 1 their salts and decomposing the salts to the metals and/or metal oxides. For purposes of simplicity, we will describe our invention using one of the preferred systems; namely, ; : `
palladium-rhodium and nickel oxide as the active nitrogen ;
oxide conversion constituents, although it is obvious that -the other c~talyst systems mentioned above can a1so be -prepared using the same general procedures.
When the support is granular, as for~example, spheres, `
extrudakes, nodules, the active components are applied by conventional techniques such as dipping, spraying, or im-pregnating the support into or with a solution of the metal `
, salts. Suitable palladium and rhodium salts include the i chlorides, nitrates, tetraamine dinitrates, etc. The nickel may be applied from a solution of the nitrates, chlorides, acetate, etc.
- 30 The composition and concentration of the catalytic-components in the catalysts depend on the type of system ;
~¦ (granular or monolithic) that is desired. When the system .

. . ~ .
,i ' ~ .

~L~38~
is a monolithic structure, the concentration o the palladium 1 ~5 and/or platinum based on the inal weight o the monolithic i :
structure ranges ~etween a~out ~. n2 to 0.5 weight percent. ...
The nickel oxide concentration amounts to about 1 to 20, and preferably 4 to 8 weight percent based on the total weight ~ of the monolithic structure~ The rhodium is present in I smaller quantities in the range of 0.001 to 0.01 weight percent. .
1 One particularly suitable method of preparing the catalyst when the support is`a monolithic structure; such as the several commercially available monoliths made up of cordierite or mullite, consists of preparing a suspension or ..
slip of the finely divided catalyst support in water or other '. suitable vehicle and applying the slip to coat the monolith to the extent o about 5 to 15% by weight based on the weight ', of the support. This can be aacomplished by dipping the mono- :
lith into a slip of aboùt 20 - 30 percent by weight solids (for exam~le 25% A12O3, silica-alumina) and removing the excess :~
coating by blowing the monolith channels free w~ith a blast of c - ~
1 air. The coated monolith is then dried between 400 and 1000F ~
:~ 20 and is ready for impregnation with a solution containing .~: ~
:, ; .. , :: .
~, the catalytically active ingredients. The solution can be prepared by dissolving the necessary amounts of nickelous nitrate, palladium tetraamine dinitrate, and rhodium chloride I in just enough water to saturate the coated monolith to j incipient wetness. After drying, the monolithic catalyst is `;'' activated by converting the noble and base metal salts to .~ i ~ metal-metal oxides by calcining in air or under a reducing .;
:, atmosphere such as with hydrogen at 1000 to 1200F.
I Another satisfactory method for distending the - ~;
.~ 30 supported catalyst over the surface of the monolithic .-.. . .
:~ structure is to impregnate the finely divided silica-alumina .~ powder with the required amounts of nickelous nitrate . ..
, ~.5 ~
:, .''',', ~ ~ .' .

~L~38S9~ ;;
. . ~
palladium, tetraamine dinitrate and rhodium chloride, dry and calci~e at 10QQ to 14Q0F. This powder is then aispersed in the proper vehicle such as water to give a slip of solids content of about 20-30 percent by weight. The monolith is dipped into the slip, the excess ~lown off, dried and heated to about 500 to 10Q0F to anchor the catalyst coating to the monolith.
' If the granular version of the catalyst is desired the salt solutions of nickel, palladium, and rhodium are `~
applied to extruded, balled, or otherwise formed shapes ;~
(approximately 4 mm in diameter) of the silica-alumina or ` silioa support material. This form of the catalyst can perform efficiently with lower levels of the active ingred-ients than is optimal or the monolithic version. The level of nickel oxide can be between 0.25 and 5 percen~. Palladium .
and rhodium concentrations of 0.001 to 0.15 percent and 0.0002 to 0.005 percent respectively can be used. ~;
¦ Our catalysts were tested using a simulated auto exhaust environment in a bench scale apparatus. The reactor employed was a one inch diameter 316 stainless steel tube (in some cases containing a Vycor* glass liner) fitted with a :, : , center line thermowell and a grid to support the catalyst.
Granular aatalysts were sized to approximately 3x4 mm and occuppied about 13 cubic centimeters of reactor volume.
Monolithic catalysts were 2.5 cm in diameter by 0.4 cm. `
cylinders. The total gas rate passing over the catalysts could be regulated to give gas hourly space velocities in the range of 40,000 to 150,000. The simulated exhaust gas ~ ;
contained 2~. CO, 250 ppm, C3H6, 1000 ppm NO, 0.2 to 1 ~ ;
percent 2' 10 percent CO2, 10 percent H2O and the balance -nitrogen. In some cases the inlet gas contained 1.33 percent CO and 0.67 percent H2 with all other gas concentrations * Trademark ;~

' remaining the same as described above. The mole percent of inlet NO converted to N2 was measured for each catalyst '-as a function of CO/O2 or [CO ~ H2] /2 ratio and temperature ' to demonstrate the catalytic performance. `
Our invention is further illustrated by the following ~
specific, but nonlimiting examples~
ExAMæLE 1 A commercially available monolith 1 inch in diameter by 1 inch long was coated with silica-alumina cracking ~' catalyst (24% A12O3, 75~ SiO2) having a surface area oE 70 ' square meters per gram to the extent of 10% by weight of the i~
monolith.
l The coated monolith was dipped inko nickelous~
nikrate solution, dried 5 hours at room temperature, heated ~
5 hours at 230 to 300F and calcined in air for 2 hours at ~
1'1'; . ~
3 1200F. The monolith was dipped again into 2.1 ml of an '~' aqueous solution containing 9.2 milligrams of palladium and ~i 0.4 milligrams of rhodium in the form of palladium tetraamine ~ ' 'dinitrate and rhodium chloride respectively. The imp~egnated ;~ ~ ;
~1 20 monolith was dried at room temperature and heated to 260F. '`
:1 . . .
Afte~ drying, the monolith was exposed to formic aaid vapors ~'' while maintained at a temperature of 190F. At the end of . ~ . . . .
~.! this time, the monolith was activated at 300F. The catalyst . :~ .. , had the following composition, 0.a94% palladium, 0.0041%
rhodiumj-5''~--3 ~'NiO, 9'.0%'siIica'-al'umina, the ~alance being '~ ~-'~ the coraierite monolith.
The catalyst was tested in bench scale equipment i operated at a gas hourly space velocity of 40,000. The feed '' composition was as follows: ' CO 2~
~1 C3H6 250 ppm :'.~' . ;:
'' ~ 7 ~ ' ;~1 .
.~ ~ >:
.. " .

... .. , ,.., : ,.: . . , , :

~ 3aS~
NO 1000 ppm

2 Q.2 to 1%

C2 10%
H O 10%

N~ Bala~ce The mole percent conversion of NO to N2 at~ained using this -~
catalyst is presented in Table 1 as a function of temperature (F) and ~CO/O2] ratio.

~ 10 Mole % Conversion of NO to N2 ;~ [CO/O2~ Inlet Ave. Bed 72 83 97 99 ;~
It i~ apparent from these data that the catalyst has excellent NOX conversLon at an average bed temperature of 1100 and 1300F, with carbon monoxide to oxygen ratios . '~
as high as 10.

~¦ EXAMPLE 2 ~ ~ 20 ` A commerciallyaVaila~le l~i x 1" cordierite monolith ;;, was coated with silica-alumina cracking catalyst having a``~

: ,, . .
~ surface area of 79 square meters per gram, to th~ extent of ;~

;; 11% by weight. The catalyst was prepared by using the same ,~
'`;l : :: :, -~ techniques described in Example 1. The composition of the ';i .~i.
5~ ' final catalyst was as follows: ~;

~ 0.094% palladium -¦ 0.0041~ rhodium -~
.. - , '.`
4.5% NiO ' ~i~
; .:. , :. i .,. 1.1% Fe203 : ;
10% silica-alumina ' j Balance cordierite monolith `~
The catalyst was evaluated using the bench scale apparatus ~`,' ~;.-i, ,, ,,.,.,.. ,. ,....... .. ,- -~B599 and the technique described in Example 1 using a gas hourly 's' space velocity of 40,0aQ. The data collected in this series of runs are set out in Table II.
TABLE II
. Mole % Conversion at NO to N2 ..

[CO/O ] Inlet Ave. Bed 2 900F 975F 1100F 1300F ;~
t 2 97 - g8 98 ~ 6 81 - 95 99 ,~
.~ 10 73 83 92 99 .;.
It is apparent from these data that satisfactory results are obtained when the nickel oxide content is reduced .~
from 5.3~i to 4.5% and the catalyst contained 1.1% Fe2O3. . : .
EX~MPLE 3 ~, :
A commeraiallya~ ble 1" x 1" cordierite monolith :~ :
was aoated with a silica-alumina cracking catalyst having a surace area of 79 square meters per gram to the extent of 10%
. . .
~i~ by weight. The catalyst was prepared using the technique .

Z described in the previous examples. The composition of this ..

: catalyst is as follows~

0-095% palladium . ~.
; 20 1 0.0041% rhodium - ~ ;

:~ 4.4% nickel oxide ,:
~ cobalt oxide .,i 9.6% silica-alumina ~ Balance cordierite monolith ~.,, :: .
:.~, The catalyst was tested in the bench scale apparatus operated .~ at a gas hourly space velocity of 40,G00. The conversion .3 of nitrogen oxide to N2 attained in this system is set out ~ in Table III.

;~, 30 ''. ','. ~' ~

. ~ . .

~ ~38~'99 TAsLE III
i Mol~ % Conversion of NO to N
. 2 . [CO/O2] . Inlet Ave. Bed ,, 900F 975F 1100F 1300F
, 2 98 ~
6 66 77 ~ - ' ,, ~ 10 60 71 83 98 ,~ ~ :
:,' It is apparent from these data that substitutions of '~
', cobalt oxide for a portion of the nickel oxide resulted in a slight decrease in conversion at average be~-temperature ,~
of 1100 but was essentially the same as the results obtained : .;
when the catalyst contained 4.5% nickel oxide and 1.1% Fe~2O3.
~-~ EXAMPLE 4 '~
J A cordierite monolith, 1" x 1" long, which had been coated with mullite ~76 m2/gm S.A.) to the extent of ' approximately 9.1~, was dipped into a nickelous nitrate ~ solution, and the excess solution was removed by blowing.
~ After drying at 200 to 300F for 2 hours in air and ,~ calcining at 1200F, the monolith was dipped again into an ;~
,~ aqueous solution containing 9.2 mg of palladium and 0.4 mg ¦ f rhodium in the form of palladium'tetraa~mine dinitrate and rhodium trichloride, respectively. After 6 hours, drying '~
, in a vacuum oven at 180F it was air calcined 2 hours at ', 1100F. The composition of this catalyst was as follows~
'0.111% palladium ~:
.;.;0.0048~ rhodium 6.66% NiO
~: :
,',8.5% mullite .,`i;
~,~Balance cordierite monolith , ~ The above catalyst was tested on the bench in a ,;,.~l, stainless steel (Type 316) reactor at approximately lQ0,000:: l ,1 30 GHSV, with the following gaseous fee.d: .~ ~ , ~::3 , ..... . . .
. .~ .
', ` '. ' ~

.. . . . .. .

~3~59~
2% CO
250 ppm C3H6 .
1000 ppm NO
, 0.2 - 1% 2 s 10% 2 10~ H2O
1~ Balance N2 - .
'! The results are presented in Table IV.
TAB~iE IV `
l, 10 Mole ~ Conversion of NO to N2 . ~
:'AVERAGE BED C/2 TEMPERATURE ( F ) 2 10 972 - 76 ` .:
1038 97 ~'` .
1 ;"

~1116 85 . .
:l1120 :11300 94 98 :: ~
'Z ~ : :
j EXAMPLE 5 :~ 20 Silica-alumina powder hav.ing a surface area of 79 m2/gm was impregnated with a nickelous nitrate solution ::
t~ slightly beyond incipient wetness, dried and air calcined ~ :`
.i; . ~ ~ .
:q 2 hours at 1200F. The resulting material consisting of 26. 6% . ;
!
~, NiO and 73. 4~ silica-alumina was reimpregnated with an aqueous ~:
:¦ solution containing palladium tetraamine dinitrate and rhodium trichloride to slightly beyond incipient wetness, dried at .,~
Z 180 to 260F and air calcined 2 hours at 1100F. The compos~
:i3 :, ition of this material was~
~ 1.15~ palladium ~ .
~, , :
~ 30 0.05% rhodium :~ 2~.~3~ NiO
:¦ Balance silica-alumina ;~ ;.
;T
. ~ i.i~ , ~ 11 ~
., . .` .. .

1~85~
A slip containing 27~ solids ~as prepared from this matexial hyhomogèn~zing with ~ater and a small amount ~0.15~ of hydroxy ethyl cellulose.
A cordierite monolit~, 1" x 1" lony, was coated with .,~ . .
the above slip to the extent of approximately 10% by weight, dried, and air calcined 2 hours at 1100F. The composition of this catalyst was as follows~
0.115% palladium 0.0Q5% rhodium 2.6% NiO
7.3% silica alumina Balance the cordierite monolith The above catalyst was tested in the same manner as ;
1 described in Example 4. The reæults obtained at approximately 100,000 GHSV are presented in Table V-A.
'ABLE V - A
Mole % Conversion of NO to N~
``:
~ C/2 -~
. :
Average Bed Temperature (F) 2 6 10 Remarks ~ 20 - - - ~ ``

3 85~ 83 67 57 ~`
;~ 950 90 83 73 ~ ~
~l 1100 95 91 88 ~ -~: .
J .
1100 - ~3 - After 100 hours continuous testing under varying~conditions including 83 hrs.
at 1400F (ave. `i~
bed) :~
: ...
.
3 1110 96 95 92 After 4 hours at , 30 1600F air `-~ calcination i~

'.1 . ~
110Q 89 78 73 After 4 hours at 1800F air ``
! calcination 12 ~ i~
' -: ~;., ~ ~38599 ` ~
In TABLE V - B are shown the performance data of the same catal~st in a stalnle~s steel tT~pe-316) reactor with a i liner at approximately lOQ,00~ GHS~, with the same feed as ;
described in Example 4, except that 2% CO was replaced by 1.33~ ~
J CO and 0.67~ H2- ~ ;
TABLE V-B
Mole ~ Conversion of NO to N2 Average Bed C2 + ~/2 Temperature (F) 2.6 4 6 10 950 93~ 79 44 38 ;~
'j 10 ' `~ 1100 95 91 79 71 These data indicate that the catalyst prepared ~?~
according to this example has the following features:
~`' 1. Superior activity for N0x control.
2. Good stability under simulated exhaust environ- ~`
ment up to 1400F.
3. Good thermal stability up to 1600F to 1800F.
~ Another catalyst having exactly the same composition :l as the above described catalyst in this example, but without nickel oxide, has also been prepared. The performance data , of this catalyst obtained in the same manner as those included `' ;~
in Table V - B are presented in Table V - C clearly illustrates ! the fact that the nickel oxide plays a significant role in the ~'~ conversion of N0x to N2 particularly under strongly reducing ;, I `: ., ~ :~
conditions. ;~

TABLE V~C
~ .- ,., ;~ Mole % Conversion of NO to N
~ - 2 1 C02~H2/02 . -Average Bed ; ;
Temperaure (F) 2.6 6 10 llOQ 85 29 21 ~;

130Q 86 - ~4 - 13 ~
~ ; .
. .. i '.~ .~:

~P3~

A cordier~te monolit~ 1" x 1`' long, which had been coated with silica-alumina hav~ng a surface area of 100 m2/gm to the extent of approximatel~ 10% was dipped into a solution containing 9.2 mg palladium, 0.4 mg rhodium, 160 mg of NiO in the form of palladium tetraamine dinitrate, rhodium trichloride, and nickelous nitrate, respect~vel~. The resulting monolith was dried at 18QF in a vacuum oven for 6 hours, and was air ;~
` calcined 2 hours at 110QF. This catalyst had the following ;~
compositions:
0.107% palladium ~-0.0046% rhodium ;,; ~:.

4.55% NiO

9.6% silica-alumina Balance the cordierite monolith `

~he above cataiys* was evaluated on the bench in a I stainless steel (Type-316) reactor with a quartz liner at -i~ approximately 100,000 GHSV, using the same feed as one described in Example 4, except that 2% CO was replaced by ~,' 20 1.33% CO and 0.67% H2. The results are given in Table VI.

, TABLE ~I ii ;

i~ Mole % Conversion of NO to N
CO H /o -Average Bed Temperature tF) 2.6 4 6 10 945 83 71 57 37 ~ ~

1092 91 89 83 75 `

1300 95 _ _ 97 , EXAMPLE 7 ;
3~
A cordierite monolith 1`' x 1" long, ~hich has ~,een coated with ~ilica-alumina having a surface area of 100 m2/gm .:'. , ": .

`':1 ` ' . -385~9 to the extent of approximately ~.3~ was set with NiO in the same manner as in Example 4. T~e resulting monolith, ~:
was redipped into an a~ueous solution containing 2.56 mg .
platinum, 1.23 mg palladium, and 0.4 mg rhodium in the form of tetraamine dinitrates for both platinum and palladium, and rhodium trichloride, respectively. After 6 hours drying in .~
a vacuum oven at 180F, it was air calcinecl 2 hours at 1100P. - ~ :
This catalyst had the following composition~
0i074% platinum :, 10 0.Q37% palladium 0.0046% rhodium .~
6.91% Nio ~:
;i 8.7% silica-alumina Balance the cordierite monolith. ... .
Another catalyst prepared by the same procedure as described above, but without rhodium had the following ;~
composition: ;~
.$ ~.077~ platinum 0.039% palladium 7.15% NiO
l 8.7~ silica-alumina .. ;
:~ Balance cordierite . The above catalystswere evaluated exactly in the :;. ;
.l same manner as described in Example 4. The results obtained ~ .
at approximately 100,000 GHSV are presented in Ta~le VII. An r`,~
~`, examination of total conversion NO to N2 on these two .
.~ catalysts at 950 to 1100F shows that the presen¢e of a .
.-.. ~ . . ...
.1 very small amount of rhodium does promote the activity of .~.' ~.;

., noble metal catalysts by improving both total conversion of NO .

'. 30 (a portion of which goes to ammonia) and the conversion of NO .~

, to N2. The results are given in Table VII. .~ ~:

:' .::.j ~ 15 ; I ~ .
:. :

~3~35~
TABLE VII
Mole ~ Conversion of NO to N2 [C~2]
C/2 ~ ~.
~verage sed Temperature (F) 2 10 Catalyst Contain~ng Rh:

1300 94 97 :
C~talyst Containing no Rh:

1100 8g 88 ~ .

: . .

A corclierite monol~th, 1'l x 1" long, which had been , coated with silica-alumina having a surface area of 100 m2/gm t to the extent of approximately 8.6% was set with NiO in the ~ same manner as in Example 4. The monolith was dipped again .~.
-~, 20 into an aqueous solution containing 9.2 mg of palladium, ~ 1.2 mg of ruthenium, 0.4 mg of rhodium in the form of 1 i:i . , palladium tetra.amine dinitrate, ruthenium red, and rhodium ~ :
trichloride, respectively. It was then dried and air . :
calcined in the usual manner. The composition o this catalyst `~
was as follows:
0.115% palladium 0.015% ruthenium 0.Q05% rhodium ~;
1 6.56% NiO .
j 30 8. a~ silica-alumina . ~ :
.~ Balance cordierite. ~
:: The above catalyst ~a~ tested at approximately 100~000 ~ .
, :~ ~`: , , GHSV in the same manner as described in Example 4. ~
: '1 , . .
~ ~ 16 ~ ..

. ; .f " `. . , ~ , ~ , . . . . . . . . . . . . . . . . .. . .

~3~5~
The results are shown in Ta~le VIII. ~- ~
: : .
TABLE V~
Mole % Conversion of NO to N2 ~ ;
C/2 ;~-Average Bed -~
Temperature (F) 2 6 10 ~965 - 75 6 1034 g6 llaO ' 97 94 90 W. R. Grace & Co. silica gel Grade-952 ~a~ impregnated ;~
with a nickelous nitrate solukion, dried and air calcined j 4 hours at 1800F. The resulting material consisting of ;
46.2% of Nio and 53.8% SiO2, having a surEace area of 109 m2/gm, was wet ball-milled and a slip containing 30% so~ids was . ~ ....
obtained.
A cordierite monolith, 1" x 1" long was coated with the a~ove slip to the extent of approximately 13.8%, dried and calcined for 3 hours at 1400F. The monolith was then dipped into an aqueous solution containing 9.2 mg of palladium, 0.4 mg of rhodium in the form of palladium tetraamine dinitrate and rhodium trichloride, respectively. After 6 hours drying at 180F ln a vacuum oven, it was air calcined 2 hours at 1100F.
The composition of this catalyst was as follows:
0.103~ palladium ~;
- O.QQ45~ rhodium 6.38~ Nio 7.42% Si2 ;
, Balance cordierite -17 ~

.~'~ ,, ~31!~5~
The above cataly~t was evaluated by the procedure .
described in Example 4. T~.e results o~tained at approximately lOQ,OQQ GHSV are presented in Table IX. ~- ;
TABLE IX
Mole % Conversion of NO to N2 CO/O '; ~ ~
Average Bed Temperature (F) 2 10 Remarks 955 94 7~
10 llO0 97 93 ~- 1292 98 97 `

After 16 hour at ~ 1400~F (average bed) .~
and [CO/O2] of 6 ~ ::

' 8ao 79 ~4 '1 ,;, , g5a 91 69 ~ ::
' lllQ 93 91 130Q 97 98 : : :
i These data clearly indicate that s~lica is a good coating .

l material for NOX catalysts and that it is stable up to at .: 20 least 1400F under~the conditions similar to an auto exhaust ` .~

l environment. . .
I ExAMpLE 10 ~;
1 A cordierite monolith, 1" x 1'l long which had been .~ :
coated with silica-alumina having a surface area of lO0 m2/gm .
followed ~y coating with NiO, was dipped int~ an aqueous solution ;.`

;~ containing 9.2 mg of palladium 0.4 mg of rhodium, and 98 mg ~ --.. , Ce2O3 in the formcof palladium tetraamine dinitrate, rhodium - -trichloride, cerous nitrate, respectively. The monolith was .
then dried 6 hours at 180F in a vacuum oven, and was air ~.1, 3Q calcined 2 hours at llOQF. T~e composition of this~catalyst ~: ~as as ~ohlo~s:

~ 18 ~ ..
;- 1 1.06~ palladium ~ ;
0.~0.46~ rhodium ;.:~
0.453% Ce23 6.44% NiO . ~:
.
7.58~ silica-alumina Balance cordierite .r~.; ; The a~ove catalyst was evaluated twice in the manner described in Example 4. The first test was on the ~resh catalyst, the second on the same catalyst after a 3-hour . ;~
air calcination at 1800F. The results obtained at ,~
approximately 100, aoo GHSV are presented in Table X. These `~
data indicate that the presen~e of a small amount of ceria thermally stabilized the catalysts. -~
TABLE X :. `;:.
M~le ~ Convers.ion of NO to N ~. .;
~ ~.... ....
CO/02 ',~
3. Average Bed , ,~
Temperature ~F) 2 10 Remarks 992 - 71 .~
, 20 1086 96 88 ~ ~"
1295 97 97 .. i :
-------- After 3 hours ...~ : :
at 1800F .: :~ :
I 9.~2 78 50 .
1108 83 71 .~
131'i~ 85 96 .:~.
3~ EXAMPLE 11 .j A cordierite monolith,.. l" x 1" long which had been ~ coated with titania (anatase, 44 ~ /gm) was dipped into an ; ~
.~ aqueous solution containing 18.4 mg o~ pa~ladium, a. 8 mg of ... ~ ~;
~i 30 rhodium, and 19.3 mg of Ce2O3 in the form of palladium ~, tetraamine dinitrate, rhodiumj.t~i~hloride, and cerous nitrate, ,t ' .~ respectively. The monolith was then dried 6 hours at 180F ~

, .

~385~
in a vacuum oven, and was air calcined 2 hours at 110~F. The .
resulting catalyst had t~e ~ollo~ing compo~1tion~
0.172% palladium 0.0Q75% rhodium 1.84% Ce23 ;~
6.52% NiO -12.8% i2 .
: : ~
salance cordierite .
I The above catalyst was tested on the bench by the :, 10 procedure descrihed in Example 4. The results obtained at approximately 100,000 G~SV are shown in Table XI.
! TABLE XI
~ Mole % Conversion of NO to N2 ;: :
:, C/2 `:
i ~verage Bed ,~Temperature (F~ 2 10 ;~ 950 88 43 ` -~ :
1092 92 83 ~ ~:
1 1300 92 g6 ;~ 20 EXAMPLE 12 A cordierite monolith 1'l x 1" long which had been ` :.
~ coated with zirconium silicate, was dipped into an aqueous `l solution containing 9.2 mg of palladium and 014 mg of rhodium 31 in the orm of palladium tetraamine dinitrate and rhodium :¦ trichloride, respectively. After drying 6 hours at 180F
. j .. . -~ ..
.~, in a vacuum oven, the monol~th ~as air calcined 2 hours at -~
.~. llQ0F~ The resulting catalyst had the following composition~
, 0 097% p~l~ladium ,~
;¦ 0.0042% rhodium -~¦ 30 6.46% NiO
I lQ.~% zirconium silicate .: ~ ., .
Balance cordierite .,,~
.. . .
~- 2a ~

:: , 1~3~35~9 The above catalyst was evaluated in the same mannex as descri~ed in Example 4. The results obtained at approx~

imately I00,0a0 GHSV are shown in Table XII.

TABLE XII

Mole % Conversion of NO to N2 ~.

Average Bed - Temperature (F3 2 10 ~ ~

- 956 91 58 .~. ;.

1305 96 97 ii ;

EXAMPLE 13 .
~1 `. . ' ' ~, , ~ . .
A commercially available monolith, 1" x 1" long, . which had been coated with ~alumina was dipped into a .¦ nickelous nitrate solution, dried and air calcined 4 hours l;
at 1800F~ The resulting monolith was dipped again :into an aqueous solution containing 9.2 mg of palladium and 0.4 mg of rhodium in the form of palladium tetraa~mine dinitrate and . ~hodium trichloride, respectively. The monolith was then dried at 220F and was air calcined 2 hours at 1100Fo The 1 composition of this catalyst was as follows~
:¦0.104~ palladium :
.,0.0045~ rhodium `

l -5.74~ NiO
:,~15~ 123 .

Balance cordierite The above catalyst ~as evaluated in the same manner .¦ as described in Example 4. The results obtained at approx~
imately l~OQOaQ G~SV are presented in Table XIII.

:i"~

.,.~ .. ~
~ ~ 21 ~
: '! `
'.. ', " ''; ^ ~

:: ,, ,, ~ ~

~3~35~
~ABLE XIII . :
Mole % Conversion of NO to N2 ~ .
CO/02 ;; .~ ,:
Average Bed .
Temperature (F~ 2 6 10 850 77 38 29 .
950 88 45 35 ~:
1100 ~6 56 43 ~`

EXAMPLE 14 :~
A cordierite monolit-h, 1" x 1" long was dipped into ` ~
a nickelous nitrate solution, and the excess solution was~ ; ;
removed by an air blast. The monolith was dried at 200 to 260F and then air calcined 2 houxs at 1400F. It was re-dipped into an aqueous solution containing 8.3 mg o~ palladium and 0.36 ~g of rhodium in the form o~ palladium tetraa~mine dinitrate and rhodium trichloride~ respectively. The `~
resulting monolith was dried at 190F in a vacuum oven, and air calcined 2 hours at 1100F. The composition of this catalyst ;`
was as follows: ~.
~ Q.108% palladium ~.. ...: .
¦ 0.0047% rhodium ` :
7.92% Nio .
~ Balance cordierite .. ~ ' .¦ This example demonstrates that the nickel oxide need not be ..
¦ distended on a support prior to application to the monolith .
in order for the catalyst to be effective.
, The above catalyst was tested in the manner ~ described in Example 4. The results obtained at approximately ~. .
lO0,000 GHSV are presented in Table XIV. i~:

;~

~. .
. ~"
.
'~
.; " : . .

~93~9~
TABLE XI~
M~le % Con~er~ton o:~ NO to ~ .;
CO/02 , Average Bed Temperature (F~ 2 10 .
952 - 57 .
1030 97 77 .i.
.i ,j ,,~ ~ .
~, 1100 95 90 ::3 .
,'~ ''~'' ~
i ,,: ; .~ ' ' ,.~ ;~

;`
. ;~
~,"
'. `~; .`:, "~

. : .

" '.~
.~ ;: ~ ~

~s ~ 23 ~
:

" ,, ~

::

Claims

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A process for improving the high temperature conversion of nitrogen oxides in exhaust gases of internal combustion engines which comprises contacting said gases with a catalyst containing from about 1 to 20 percent by weight of nickel oxide or a mixture of nickel oxide with iron or cobalt oxides, and from about 0.001 to 1 percent by weight of a noble metal distended on a suitable support , said percent by weight being based on the combined weight of the catalyst and support.

2. The process according to claim 1 wherein the catalyst is distended on an acidic support selected from the group consisting of silica-alumina, silica and mullite.

3. The process according to claim 1 wherein the support is in nodular form.

4. The process according to claim 1 wherein the catalyst is distended on a monolithic structure.

5. The process according to claim 1 wherein the noble metal is selected from the group consisting of platinum, palladium, rhodium, ruthenium, and mixtures thereof.

6. The process according to claim 1 wherein the support is in nodular form and the noble metal is present as about 0.001 to 0.1 percent by weight.

7. The process according to claim 1 wherein the catalyst is distended on a monolith and the noble metal is present as about 0.05 to 1 percent by weight.

8. The process according to claim 1 wherein the noble metal is palladium with promotional amounts of rhodium.

9. The process according to claim 1 wherein the nickel oxide is present as about 4 to 8 percent by weight.

10. The process according to claim 5 wherein platinum or palladium is present as about 0.01 to 0.10 percent by weight and rhodium is present in a concentration of 0.0004 to 0.04 percent by weight and the support is in nodular form.

11. The process according to claim 5 wherein platinum and palladium are present as about 0.05 to 0.5 percent by weight and rhodium is present in a concentration of 0.002 to 0.02 percent by weight and the catalyst is distended on a monolith.

(12) A process for catalytically reducing nitrogen oxides contained in internal combustion engine exhaust gases while minimizing residual ammonia, which comprises contacting said gases with a catalyst comprising about 5 to 15.7% by weight nickel and about 0.03 to 1% by weight palladium on active alumina, said percent by weight being based on the combined weight of catalyst and alumina.
(13) A process for catalytically reducing nitrogen oxides contained in internal combustion engine exhaust gases while minimizing residual ammonia, which comprises contacting said gases with a catalyst comprising about 2 to 15.7% by weight nickel and about 0.03 to 1% by weight platinum on active alumina, said percent by weight being based on the combined weight of catalyst and alumina.