CA1060474A - Catalyst for hydration of nitriles - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
Solid catalyst of reduced copper magnesium sili-cate type is improved for its use in catalytic hydration of acrylonitrile to make acrylamide by calcining the catalyst particles for longer catalyst life in the reactor and slows the rate of decay of catalytic activity so the average prod-uctivity over the life of the catalyst is improved.

Description

~0~04'74 The invention relates to improvements in the cata-l~tic hydration o~ nitriles with water to produce amides by a continuous process us~ng certain solid heterogeneouQ cata-lysts ~or the reaction, More particularly, the invention provides an improvement in solid catalyst used ~or such re-action, havlng improved physical strength which improves the length o~ use~ul catalyst life with consequent increase Or the amount Or product that can be produced with a given amount Or the catalyst.
The particular type Or catalyst to which the inven-tion relates was in use be~ore it was improved by the present invention, This particular type o~ catalyst, called copper-magnesium silicate catalyst, was prepared by precipitation copper and magnesium ~rom aqueous solution by introducing alkall metal silicate and alkali metal carbonate into a solu-tlon o~ copper and magnesium salts. The copper-magnesium carbonate silicate precipitate wa~ separated by ~iltration ~rom the mother liquor, washed, dried and pressed or e~truded to ~orm pellets, beads, or the like which were ~urther dried and then shipped, stored and charged to the reactor in dry pellet ~orm. Treatment with a reducing gas was de~erred until ~ust be~ore the catalyst use. Catalysts o~ this type may pre~erably also contain in minor proportions, e.g. .01

2~ to about 5 percent by weight, o~ one or several promoters ; ~uch as compounds o~ barium, zinc, cadmium, chromium, molyb-i dsnum, tungsten, vanadium, uranium, titanium, thorium or the like. Catalyst~ with such promoters are prepared in some ; ~nstances by precipitating the promoter together with the insoluble copper and magne~ium compounds ~rom a similar copper magnesium solution which ~urther contains a corre-~ spondingly small amount o~ a dissolved salt o~ the ~elected - promoter metal in addltion to the copper and magnesium salts.

4'74 Rererence i9 made to the German Patent No. 869,o52 ror more detailed de~cription Or the preparation Or catalysts o~ the type de~cribed. Belgium Patentq No. 813,973 and No~ 813,974 described typical hydration Or acrylonitrile using such cat-alysts, A typical catalyst o~ this type is available com-mercially from Badische Anilin ~ Soda Fabrik AG, Ludwigsha~en, West Germany under the tradename BASF Cataly~t R3-11, also called BTS Catalyst~ It contains approximately 30 percent by weight Or copper combined in compounds Or copper disper-sed throughout the magnesium silicate matr~. This catalystis supplied in cylindrical pellets about 1/8 inch long

3/16 inch diameter, For its partlcular use in the hydration Or nitrile the catalyst i8 activated by redu¢tion with hydrogen or other errective reducing agents, ~rererably the reduction la car-ried out with hydrogen at 180 to 230C, and prererably with the catalyst already placed ln the rixed-bed reactor and also prererably ~ust prlor to use Or the catalyst ror the contin-UOU8 hydration process, Arter the catalyst has been reduced the reduced catalyst is kept away rrom o~ygen to avoid o~i-dation o~ catalytic copper surrace areas and consequent 1088 Or catalytic activity.
Combined copper in the catalyst is conveniently reduced to elemental copper by ~irst charging the cataly~t pellets to the bed Or a ri~ed bed catalytic reactor which is to be used ror the hydration proce~s, Then a stream Or re-ducing gas is red through the catalyst bed at temperature in the range from 180-230C, until there i~ no ~urther re-duction to elem~ntal copper at the selected reducing con-- 30 dition~, The reducing gas is prererably hydrogen, but other reducing gas such as carbon mono~ide or the l~e could be used. It ia prererred to rirst heat the catalyst with not nitrogen to about 120 to 160C, and then gradually add ~ ~ k - 2 -- . ; - . , - ~ . - .
. .

~Of~0~74 hydrogen to the nitrogen stream until the bed temperature is raised to about 180-230C. m e reaction ~or reduction o~ copper with hydrogen is e~othermic and the bed temper-ature is maintained at the selected temperature by regulating the gas ~eed rate and ad~u3ting the concentration o~ reduc-ing gas in the gas reed. Depending upon the particular re-ducing conditions selected, the reduction step as described usually is completed in about 8 to 24 hours. A~ter cooling the catalyst bed, the reactor is ready ~or introduction Or a nitrile-in-water reactant solution a~ the ~eed stream to commence the continuous hydration reaction. ~uring the treatment with reducing gas to reduce the copper, the cata-lyst is sensitive to exces~ive temperature and it is nec-essary to ¢are~ully regulate the bed temperature durlng re-1~ ductlon to avold deactlvatlon by overheatlng the catalyst,We pre~er to avoid heating the catalyst bed beyond an upper limlt o~ about 230-250C. to pre3erve the catalyst dur-ing the reducing reaction. Deactivation, or at least a con-siderable 1088 o~ activity can be expected as temperatures about 250C. are approached or exceeded during portions o~, or all o~ the reduction step.
For the hydration o~ acrylonitrile with water the catalyst obtained commercially and reduced as described, was round to produce very hiBh yield Or acrylamide at conversion rate~ comparable with thoss obtained with a good reduced copper chromium oxide oatalyst, notwithstanding the consider-ably lower proportion o~ copper in this catalyst. Thi8 cata-lyst was clearly superior to a good copper chromium o~ide catalyst on the basis o~ higher conversion per hour per pound 3~ Or copper in the catalyst and on the basis o~ longer catalyst li~e, that is, a slower rate o~ decay o~ the catalytic ac-tivity as the hydration process was carried out continuously.
A principal disadvantage with this catalyst was that the -- 3 ~

()4'74 catalyst particles did not have as much physical strength as would have been desired. This was particularly so after the catalyst had been reduced and washed with the liquid reactants The pellets were apt to fracture and cr~ble causing blockage and channeling in the fixed catalyst bed with --consequent loss of circulation of reactants to much of the effective catalytic surface area in the catalyst bed, This was found to be especially severe in large industrial size catalyst beds containing several tons of the catalyst Furthermore, solid polymer was apt to form at regions of lost circulation in the catalyst bed causing those regions to solidify Such events would drastically reduce the effective activity of the catalyst bed Thus, because of the inadequate physical strength of the catalyst particles, much of the advantage of the long catalyst life would be lost by irreversible physical deterioration of the catalyst bed structure during use.
The present invention provides in a process for catalytic hydration of acrylonitrile to produce acrylamide by flowing a reactant solution com-prising acrylonitrile in water through a fixed bet of copper magnesium silicate type catalyst after reduction of the catalyst, the improvement wherein said catalyst has been treated prior to the hydration reaction by calcining the catalyst in a non-reactive atmosphere at calcining temperature in the range from 300 to 500C. for time from about 1 to 24 hours sufficient to substantially decrease the rate of decay of catalytic activity during use of the reduced catalyst in the defined hydration process.
According to the invention, catalyst particles of the copper magnesium silicate type described herein are treated by calcining the catalyst particles at high temperature in an essentially, non-reactive atmosphere, The calcining can be carried out either b0fore or after the catalyst has been reduced. It was expected that the high temperature needed for effective calcining to improve particle strength ~300-500C.) would cause physical deterioration of the particles, either by fracture and crumbling of the particles or by sintering of the active copper, or would

4 _ ! ~ , . , ' . ' . , , , ' ' . . . ' 10~;0474 otherwise cause deactivation of the catalytic copper, thus spoiling the catalyst for subsequent use, Deactivation of the same catalyst had in fact been reported during reductions carried out at excessively high temperature, Also, destructive physical deterioration of the catalyst had been observed during attempts at high temperature oxidation for ' _ 4a -.. ..
,... .. .. : . . . .

regeneration of the spent catalyst. It is found however, that calcining the catalyst before it is used in the hydration reaction, either before or after the reduction of the catalyst, at calcining temperatures in the range from about 300C. to about 500C. for periods in the range from about one hour up to about 48 hours, does not cause as much physical deterioration or deactivation of the catalyst during such calcining step as might have been expected.
The calcining does in fact physically stengthen the catalyst.
; More important, such calcining is found to increase the effective catalyst life as the catalyst is used in a fixed bed reactor for the hydration of acrylonitrile with water.
It had been anticipated that high temperature calcining before the reduction step would cause chemical changes of the copper compounds during calcining, and it was expected that such chemical change probably would detract substantially from the excellent activity and catalyst life when the calcined catalyst was reduced and used for the catalytic hydration reaction. As , was expected, the calcining does in most instances reduce the initial activity of the catalyst, but this can be kept to a tolerable extent. However, the rate of decay of catalyst activity ; ..
~3 of the catalyst in a fixed bed-reactor during the continuous ,~ hydration reaction is so much improved by the calcining treatment, that the increase of effective catalyst life by slower decay of activity will more than compensate for the reduced initial activi- -ty. More acrylamide will be produced during the extended useful life of the calcined catalyst than would be produced with the higher initial activity but faster decay of the uncalcined catalyst.
The process o~ precalcining the catalyst before the initial reduction step is distinguished from the step of preheat~ng t~e catalyst to reduction temperature that

- 5 , ; ~ , - - ~ , - . -was u~ed to begin the prior nrt reduction ~t~n ~ cribe(~
above. In both methods the cataly~t i~ heated prior to the initial reduction, and both may be carried out with the cat-alyst already placed in the reactor, but in the prior art process the catalyst was preheated only to about 140-180C.
to help initiate the reduction reaction. In the prior art pro~ess the temperature was carerully regulated during re-duction, including the preheating step, to avoid overheating the cataly~t. In the calcining proces~ the catalyst is heated in a non-reactive atmo~phere, either be~ore or arter the reducing step, to a calcining temperature that is higher as contrasted with tho~e temperatures which had been previ-ously ~ound ~uitable ~or the reducing ~tep.
When the calcining proces~ i3 carried out be~ore rcduction o~ the catal~t, we re~er to that process as pre-¢alcining, as distinguished ~rom post-reduction calcining.
Precalcinlng is carried out in a non-reactive, e.g. non-reducing, atmo~phere and the high temperature ~or the pre-calcining treatment is applied to the catalyQt ~or a period lone enough to e~ect ~ubstantial strengthening o~ the cata-~; lyst during ~uch period o~ high temperature calcining. An atmosphere o~ heated air or other non-reducing atmosphere can be used to surround and heat the cataly~t during the precalcining step. If air or any other atmosphere contain-ing an oxidizing agent is u~ed during precalcining, one ~hould care~ully purge the cataly~t bed with inert gas, e.g.
~ ~ .
~ ; nitrogen, be~ore introducing the reducing gas, e.g. hydro-:: `
~; gen, to avoid any danger Or explo~ion, It is pre~erred, a~ter precalcining,~to cool the catalyst at least to a lower temperature that i9 suitable ~or the reduction step be~ore beginning the reduction ~tep, It i~ u~ually mo~t convenient, but not necessary ; in all in~tances, to calcine the catalyst after it has been . - ~ - . . . . ~ . . .
: ,: . . ~ - . , 10~0474 placed in the catalytic reactor. Alternatively, the cata-lyst can be precalcined el~ewhere under the temperature and other conditions prescribed herein ~or calcining; ~or ex-ample, the catalyst particles might be precalcined as a final step in the manuracture Or the catalyst. me precalcined catalyst prior to reduction is stable, not sub~ect to de-activatlon by exposure to air, and can be handled, stored and shipped, the same as had been done in the past berore reduction o~ the uncalcined catalyst.
Arter the catalyst has been precalcined, the ac-tivation by reduction and the use Or the reduced catalyst ror hydration are carr-ed out es~entially the same as de-. .
scribed ~bove ror the activation o~ an untreated catalyst, Examxle 1 A sample taken rrOm a bat¢h Or BASF Catalyst R3-11, as received, i9 tested ~or crush strength, A sinele cylin-dri¢al pellet o~ the catalyst is crushed by application Or mechanical ~orce in a recording prsssure press, The crush-ing rorce is applied at opposed sides along the length Or the cylindrioal pellet. Crushing rorce applied is divided by length Or the pellet to obtain a comparative crush strength value in units Or pounds rorce per inch length. Nominal di-ameter Or all Or the pellets tested is 3/16 inch. The same test ror crush stren8th is used for all the examples herein.
Crush strength o~ the dry pellets rrom one batch (Batch II) as received was 126 + 27 lb./in. Wetting the pellet~ with water i9 found to reduce the crush strength and not all Or the original strength is restored when the wet pellet is ~; ~ dried, Notioeable variation in crush strength was round ~rom one batch to another. -Los~ Or crush strength also re-~ults from several kind9 of treatment o~ the catalyst, par-tioularly aqueous wash or soak treatments at elevated tem-perature, ~t`~4 h?ark 7 . . . . . . . . . . ..

~ 4t~
From another batch (Batch IV) Or the 8ASF R3-11 A catalyst ~he dry crush ~trength Or sample pellets, tested as received, was 95 + 21 lb./in. Pellet~ ~rom this batch cru~h-ed in the wet state at 56 + 9 lb./in. arter being wet ror one hour and 40 + 7 lb./in. arter 14 days wetting. Pellets rrom this batch whlch were wet and then dried in air at 110C., crushed dry at 61 + 9 lb,/in. Pellets wet and then dried and then wet again ror one hour, crushed wet at 28 + 6 lb./in.
Other pellets ~rom this same batch were calcined, as re-ceived, at 40ooc. ror 16 hours. Another sample rrom thesame batch was calcined at 450C. ror 16 hours. Arter cal-cining, the pellets which were calcined at 400C. crushed dry at 168 + 41 lb~/in. and crushed wet, arter being wet ror one hour, at 88 + 25 lb./in. The pellets calcined at 450C.
crushed dry at 155 + 44 lb./in. and crushed wet, a~ter being wet ror ono hour, at 91 + 12 lb./in.
Sin¢e the eatalyst ln use is wet and in reduced state, ~urther measurements Or the catalyst strength were carried out arter reduction and washing with water ror two days. It was round that catalyst Or the same lot (Batch II) which was reduced wlthout prior treatment (see E~ample 2 ror reduction procedures) and then washed with deaerated and deionized water at 75C. ror two days cru~hed wet at 18 * 2 lb./in. Furtherm~re, arter about 4 months operation ~ .
25 the catalyst was round to have a wet crush strength o~ only about 10 lbs./in. with many pellets 90 badly ~ragmented or 80 ~o~t to the touch that the strength o~ those pellets could not be measured. The same ¢atalyst calcined at 400C.
or 16 hours, reduced and washed ror two days crushed wet 30 at 30 + 6 lb./in. Likewise, ¢atalyst ¢al¢ined at 450C.
reduced and washed ~or two days ¢rushed wet at 36 + 6 lb./in.
Cal¢ination a~ter redu¢tion was also investigated.
Batch IV catalyst was reduced, then calcined at 350C. ~or 17 ~ ~ ~ac/~ mc~ ~k - 8 hours and washed for two days. The wet crush strength was ;
39 + 10 lb.tin. Continued washing fOT one month reduced the wet strength to 30 + 7 lb./in. In exactly the same manner, if calcina-tion after reduction is carried out at 400 &., the wet strength is 54 + 17 lb./in. after two days washing and 37 + 8 lb./in. after one months washing.
The results show that there is a severe loss of strength when the catalyst is reduced. Furthermore, the catalyst continues to lose strength with time in use. It is therefore of considerable -~
im~ortance in industrial sized deep beds intended for continuous operating for a period of about a year or more to increase the initial strength of the catalyst so that the catalyst will retain strength sufficient for use in the reactors over a longer period.
These tests established that increase of both dry and wet crus-h strengths would be obtained by calcining the catalyst ~ut the e~ects of calcining on the catalytic activity in the hydration reaction were still uncertain. Further tests under `
conditions simulating actual operating conditions were conducted to examine catalytic activity in use of the calcined catalyst in a fixed bed catalytic reactor for the hydration o acrylonitrile.
EX~lé 2 A. In an oven, 90.7 gms. of BASF Catalyst R3-11 from ~ -Batch II as received is heated to 450C. and calcined at that temperature for 45 hours and then gradually cooled. The calcined catalyst is charged to a laboratory fixed bed con-tinuous reactor. The reactor is purged with nitrogen to re-~sve all of the air. The next step is to reduce the catalyst and this is done by f.irst bringing the bed temperature to 180 C. and then beginning gradual addition of hydrogen into a stream of nitrogen flowing through the bed. At first, the Trademark .~ 9 ,.~ .

)474 concentration of hydrogen in the nitrogen stream is about 1.5 per-cent and this is gradually increased ~sneeded to maintain the bed temperature in the range from 180 to 220C. as the exothermic re-duction of copper proceeds. The total reduction time is about 6 - 7 hours. When there is no further reduction the bed is gradually cooled to room temperature after which the flow of deaerated and deionized (DA-DI~ water at 75C. is begun and continued for a period of about 40 hours to remove soluble inorganics. Then the reaction is begun by introducing a 7 percent solution of deaerated acrylonitrile in DA-DI water as the reactant stream to the reactor.
: -During most of the continuous hydration the flow rate of the re-actant stream is regulated to maintain about 85 to 90 percent con-version but at periodic intervals the flow rate is varied to ob- -tain several diferent conversion rates for short times for test purposes, The reaction temperature is maintained at about 70C.
~.i As the activity o~ the catalyst declines the feed rate is de-creased to maintain the conversion rate in the selected operating range, The selectivity of conversion to acrylonitrile is near 100 percent throughout the continuous process. The reactor, tem-2a perature is maintained at about 70 C, throughout the operation by a constant temperature oil bath in ~hich the reactor is sub-mesged, The reactor is run continuously for several months.
The catalyst activity at the beginning of the reaction is indicat-ed by an initial productivity rate of 112 lbs. AMD produced per 1~ . . .
I000 lbs. catalyst per hour, At the observed activity decay rate, ; productivity declines to 92 at the end of three months, to 78at the end o~ six month~, and to 68 at the end of nine months.
All o the productivity rates expressed herein are the values o~tained by measurements made at a selected conver-3~ s~on o~ 60 percent conversion per pass during the short test per~ds, T~us, the reported productivity rates for all .
10 .`

,.. , . .. , - .. ~ .... . ...... .. .~ . .

l()~U474 o~ the hydration reaction~ that were run at the 3ame reac-tion temperature are directly comparable ~or indication o~
the relative catalytic activitie~.
B. A control run i9 operated in another identical reactor u~ing other catalyst rrom the same batch which i~
prepared without calcining. The cataly~t a~ received i9 charged to the reactor and washed with water at 75C. for 2 - 3 days and then dried with air at 100C. berore reduction and the catalyst i9 reduced by the same proce~ described ror reduction Or the previouQ catalyst. The control cata-lyst i9 then u~ed in the continuous hydration reaction the same as in the previously de~cribed reaction. At the be-ginning Or the hydration at 70C. the freshly reduced cat-alyst has an initial productlvity rate o~ 132. At the ob-served dec~y rate, the prodùctivit~ rate at the end Or three , months continuous running will have declined to 98 and thento 77 àt six months and to 64 at nine months.
C, Another reactor is prepared by loading the re-actor with 63 gm. catalyst ~rom the same Lot II, reducing the catalyst with hydrogen in a nitrogen stream a~ described above and then calcining the catalyst by gradually increas-ing the catalyst temperature to the calcining temperature Or 350c. by rlowing hot nitrogen (with trace a~ount Or hydro-gen) through the bed. The catalyst bed temperature is held at,350C. ror three hours and then gradually cooled by reducing the temperature Or the flowing nitrogen. The catalyst bed i~ then washed ror 2 - 3 dayY with water rlow-ing through the bed at 75C. and then the reactor i8 charged with a reed ~tream o~ 7 percent acrylamide aqueous solution at 70C. to begin the continuou9 catalytic hydration process.
The ~eed rate i~ adjusted to obtain B5-90 percent conversion per pass. Initial actiYity Or the catalyst is indicated by an initial productivit~ rate Or 111, At the end Or three 1()~(~474 months on stream the productivity declines to 93, decreas-itlg to 80 at ~ix months to 70 at nine month~.
In the three continuous hydration reaction~ de-scribed above, the control cataly~t has higher initial ac-tivity but su~rers a more precipitou~ decline o~ activity.
At the end o~ 9i~ months all three cataly3ts have roughly equivalent activities and at the end o~ nine months the a¢tivity Or the control catalyst has rallen below the ac-tiv1tieq o~ the calcined catalysts. The avera~e productiv-ity o~ each o~ the three catalysts over three, si~ and ninemonth periods was calculated and is tabulated below. Over the nine month period the control ¢atalyst has produced only slightly more acrylamide than the calcined catalyst. As the production continues beyond nine months, the average produc-tivity o~ the calcined catalyst will exceed that o~ the un-cal¢ined catalyst.
__ . . .
Average Productivity at 70C. over period o~
3 months 6 months 9 months ¦
Example 2a. 101 93 86 _ _ Zb, 113 100 90 2c. 102 94 ~7 By operating the continuous reaction at 85C. in-tead Or 70C. the reaction is accelerated and the cata-lyst decay is also accelerated. mus, the contrast o~ be-havior of a calcined catalyst and a control catalyst i8 more marked within a given period.
D. A conbrol catalyst i8 prepared and used the same a~ described in 2(b) above e~cept the contlnuous hy-dration reaction temperature is maintained at ~5C. insteado~ 70C. Initial produ¢tivity is 248, declining to 160 at the end Or throe month9, 118 at si~ months and 94 at nine months.

4'74 E A post-reduction calcined catalyst iQ prspared qnd used ~he same as in 2(c) above except the continuou~ hy-dration reaction temperature is maintained at 85C. instead o~ 70C. The initial productivity i~ 215, declining to 174 at the end Or three monthsJ to 146 at the end o~ si~ months and 126 at the end o~ nine months, The avera~e productivity Or the two catalyst~ over --three, siz and nine month periods is tabulated below. At the end Or si~ months the calcined catalyst has produced more acrylamide than the control, and 3till more at the end of nine months.
. . _ . . ~
Average Productivity at 85C over period Or 3 months 6 months 9 months ._ .
15Example 2d. 198 167 147 2e. 193 176 162 Example 3 For commercial production Or acrylamide by cata-lyt~c hydration o~ acrylonitrile, it has been round advan-20 tageou~ to treat the BASF Catalyst R3-11 with an aqueou~
solution o~ a soluble sul~ate, such as an alkali metal sul-rate, in order to rix barium in the catalyst. Thi8 pre-vents leaching Or barium rrom the ~olid catalyst into the --liquid product stream as the catalytic reaction is carried out A. A laboratory size, rl ed-bed reactor is charged with 100 gm. o~ BASF Catalyst R3-11 rrom another batch des-ignated Batch IV, as recelved. Berore any heat treatment, the catalyst bed i9 soaked ~or about 20 - 24 hours in a 5 percent ~a2S04 aqueous solution at 75C. The sul~ate solu-tion is drained o~, and the catalyst bed i9 then washed with water circulated throug~ the bed at 75C. ror 2 - 3 days to remove sul~ate. After waQhing, the bed i~ dried by ~lowing h~k - 13 -, .

4'74 air heated to 110C. through the bed until dry. The tem-perature o~ the bed i~ then raised to 400C. b~ rlowin~
heated air through the bed, and the catalyst bed temper-ature i8 held at 400C. ror 21 hours to calcine the catalyst.
The cataly~t bed i9 then cooled and purged with nitrogen and reduced with hydrogen, cooled and then ~tarted on stream with an aqueous 7 percent acrylonitrile reactant solution, all by procedures essentially the same as those descr1bed in E~-ample 2 e~cept the wash is omitted arter the reduction step.
The continuous hydration reaction is carried out at 70C., 85-90 percent conversion.
B. A control reactor is prepared by preparing and reducing the catalyst bed the same, using catalyst from the same batch, e~cept the calcining step is omitted and the ¢ataly~t bed is charged with 65 gms. instead Or 100 gms.
cataly~t because the rea¢tor size is not the s~me. The hy-dration reaation i9 carried out the same in the ¢ontrol reactor as in the test reactor.
C. Still another test rea¢tor i8 prepared by rirst calcining 115 gm8. Or catalyst rrom the same bat¢h at 400C.
ror 17 hours, cooling the bed and then redu¢ing, cooling to about 75C. and then soaking the reduced catalyst with de-aerated ~ percent Na2S04 solution overnight at 75C. and then wQshing with deaerated, deionised water ror 2 - 3 days.
2~ me reactor i9 then operated in the h~dration reaction as described above, at 70C., 85-90 percent converslon.
Productivity initially and at the end Or three, .
~' 8i~ and nine month periods at the observed decay rate are tabulated below. The calculated average productivity ror each rea¢tion over three month, 8i~ month and nine month periods are also tabulated for each reaction. At the end of 8i~ month~, both Or the calcined catalysts have pro-duced more acrylamide, at the average productivity ror that - - -: , .. . .

1()60474 period, than the control catalyst has produced. At the end Or three ~onths, the activity o~ the control catalyst, though initially higher, ha~ declined to activity below that o~ either o~ the calcined catalysts.

1~ ' .:

` ~' ~ 30 - - . -: .. ~; " , .

1()6Q474 , C.7 ~a 0~ O) O
~ ~ ~0 1~~
;~ '~ ~ ., . .~ ~ ~ ~ :' 00 _ ., ~ 00 ~ ~

, .~ ~ U~ ~

--. . . : ' ~: ~:

.

- 16 ~

04'74 Example 4 E~periment~ were carried out to investigate the interaction Or po~t-reduction calcination time and temper-ature and the resulting er~ect on catalyst strength, ac-tivity and lire, Batch IV cataly~t was used in all cases. me catalyst was reduced in the reactor as described in the previous e~amples and then calcined in the reactor by M ow-ing hot nitrogen (with a trace Or hydrogen) through the bed Or reduced catalyst. The calcining temperature and time held at that temperature is shown in the Table below ror each run. Arter calclning and cooling, the catalyst was washed by rlowing Or DA-DI water ror two days through the bed, In the case o~ the control, calcination was not car-1~ ried out. Wet crush strength o~ catalyst samples was mea-sured, The reactors were operated as described in the previous e~amp1es ~or the h~dration reactions, e~cept the reaction temperature was maintained at 800C. in all re-actors. The results are shown in the attached ~able. This e~ample demonstrates that post-reduction calclnation at lower temperatures but ~or longer times is errective ~or giving increased crush strength and catalyst li~e while minimizing the 1088 Or initial activity.
Optimum conditions ror post-reduction calcination appear to be calcining in the temperature range Or 300-350C.
or one to two day~.

' :, . . .

106(~4~74 P~ ~ ~ o 0 h~
. P~ '~
~E~
o c: ~.~o _~

0 a~ h ~3 O `O :J I
e~ ~0~ ~ ~: ~ _ ~ -:~ 0 0 ,C
~ J ~ a ~ ~ ~o cr) - 1 :
,,i ~ ~ ~a ~ ~ .. I
0 C:~ ~ 0 j ~ I ~ N 'I ~

~ r ~ h b~ ¦ p~ v ~ 2 o~ ~ , IY; ~

h ~ ~: :~ ~ O ~ r-0 :~ h ~ +1 +1 +1 +1 +1 +1 0 . . _ . .. . , U ~ '~0~
~: .. o ~ ~ r- ~ ~ . .
~o ~ ~ a~
: C~ o ~ ~o~
:: ~ ~0 ~ ~, h ~ ~ o ~ o o o O O
O~d ~ ~ t ~t -10~04'~4 Calcining o~ the catalyst e~ther be~ore or a~ter t~.e rcduction ~tep i~ round to cause Qome le99 Or initial catalyst activity but it also improve~ the rate Or decay o~
~; cataly~t activity. Thererore, the calcined catalysts remain actlve oYer Q longsr period Or use in the reactor than the use~ul period ror the uncalcined catalyst. When the cal-cining i9 carried out arter the cataly~t has been reduced lt is neces~ary to protect the catalyst ~rom oxidizing agents during calcining and 90 an atmosphere Or heated ni-trogen or other inert gas surrounds the catalyst during thecalcining operation, A sma}l amount of hydrogen (e.g. 1%,) may be added to purge any transient o~idizing agents.
It is observed that milder calcining condition~, i~e. lower temperatures, cause less 1088 Or initial catalyst activity while more severe calcining conditions oause more improv0ment Or the physlcal strength, hen¢e longer catalyst lire, While it has not yet been determined ~ust which com-bination8 o~ the several variable calcining conditions will produce the optimum calcined catalyst, there iB enough e~-perimental data to demonstrate the improvements obtainedand to derine certain general ranges Or calcining conditions ror improving the catalyst. It i8 al~o noticed that the ~; copper magnesium silicate catalysts Or the type treated by ~;~ the invention may tend to vary in small degree rrom one lot to another with respect to the values Or several properties arrected by a similar treatment, such as initial activity, crush strength, et , 80 that precise productivity values and rate Or decay Or the catalytic activlty may be round to vary in some instances to small degree depending on the particular properties of any catalyst lot. While precise quantitative values are not always reproducible, a general improvement Or the total quantity Or product obtained with a catalyst lot treated by the invention is obtained by .

10~()474 :

improvement Or the rate Or decay Or the catalytlc activity qs the treated catalyst is used ror the hydration Or nitriles :~
in a rixed bed reactor and by the e~pected longer catalyst li~e due to increase o~ catalyst strength by calcining.

, .

::
,, :
, ,:
; .
~ - 20 -.

Claims (4)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. In a process for catalytic hydration of acrylonitrile to produce acrylamide by flowing a reactant solution comprising acrylonitrile in water through a fixed bed of copper magnesium silicate type catalyst after reduction of the catalyst, the improvement wherein said catalyst has been treated prior to the hydration reaction by calcining the catalyst in a non-reactive atmosphere at calcining temperature in the range from 300° to 500°C. for time from about 1 to 24 hours sufficient to substantially decrease the rate of decay of catalytic activity during use of the reduced catalyst in the de-fined hydration process.

2. An improved process defined by claim 1 wherein the defined catalyst has been treated by calcining, as defined, before the catalyst is reduced.

3. An improved process defined by claim 1 wherein the defined catalyst has been treated by calcining, as defined, after the catalyst is reduced.

4. An improved process defined by claim 1 wherein the defined catalyst is additionally treated by washing the catalyst with an aqueous solution of sodium sulfate and rinsed clear of sulfate before the catalyst is used in the hydration reaction.