Method for Making Acrylic Acid From Acrolein

In the process of making acrylic acid the most widely used processes are those which oxidize propylene or acrolein to acrylic acid. Many different combinations of metal oxides have been used as catalysts. Most of these contain molybdenum oxide as the principal component. Some of the catalysts are effective in oxidizing the propylene directly to acrylic acid although this is frequently a function of the conditions under which the catalyst is used; others oxidize acrolein to the acid. In either case, any acrolein not converted, or acrolein made in the process of oxidizing propylene to acrylic acid, may be recycled to the feed stream and subsequently oxidized to form the acrylic acid.

In addition to molybdenum oxide, the catalysts of the prior art contain many other metals (usually as their oxides) which promote the catalytic effect of the molybdenum. The transition metals of Group VIII of the periodic chart, including iron, cobalt and nickel, have been employed in many such catalysts. Others selected from various groups of metals of the periodic chart have been employed. Thus, for example, titanium, vanadium, chromium, tungsten and manganese in groups IVB, VB, VIB and VIIB are known to be useful. From group IIA such metals as magnesium, calcium, strontium and barium, from groups IB and IIB copper, silver, zinc and cadmium; also from groups IA and VI sodium, potassium, antimony and bismuth; all have been taught as catalytic promotor components. Phosphorous has been employed, usually added as phosphoric acid or as metal phosphates, as an essential part of many of the prior art catalysts. Some of the early catalysts employed molybdenum oxide in combination with 1 to 2 other metals, e.g., U.S. Pat. No. 2,881,212 which employed several metals as phosphomolybdates. In more recent years, catalysts containing 3-6 or as many as 8-10 different metals have been disclosed.

The art known to the inventors as being closest to that of the present invention is found in U.S. Pat. Nos. 3,775,474; 3,833,649; 3,886,092 which teach the use of Mo, V, Cr, Cu, and W in various combinations. None of these teach the use of tantalum, titanium, or niobium as taught by the present invention, nor does the instant catalyst composition contain the tungsten taught by the above three U.S. patents. Another U.S. Pat. No. 3,865,873, employs tantalum together with molybdenum and phosphorus, but contains none of the other components of the catalyst of this invention. Inventors are also aware of British Pat. No. 1,488,889 which employs oxides of Mo, V, Ti and a fourth component selected from among Cu, Co, Cr, and/or Mn, as catalyst for oxidizing an unsaturated olefin to the corresponding acid.

The efficiency of the catalysts made from any particular combination of metals apparently is affected by the manner in which they are made, and whether or not they are supported or in pelleted form. The porosity and the surface area of either the pellet or the support are important to the performance of the catalyst and will determine to some extent the amount of catalytic material employed on the support. It is extremely important in the preparation of the catalyst to obtain uniform distribution of the various oxides contained therein, otherwise the effect of the combination may be lost. The molybdenum and the various promotor metals are added as their soluble salts, usually in acid solution. They are sometimes mixed together in the same solution, but because of possible problems with premature precipitation which would cause non-uniformity in the finished catalyst, the metals are most frequently made up in separate solutions which are in turn added together under the proper controlled conditions. Generally, the final pH of the solutions is slightly on the acid side, at about pH 6-6.5. The solvent is then evaporated from the solution of the catalyst components in the presence of a support in order to burden it, when a support is used. If not supported, the components are dried and pelleted. Most of the techniques associated with the manufacture of the catalyst are well known to the prior art. The particular techniques used to make the catalyst of the present invention will be enumerated and exemplified herein.

The parameters of feed composition, flow rate, temperature and pressure are well known to the prior art. Thus the aldehyde is usually present in the feed stream from about 1-10 volume %, the remainder being molecular oxygen (0.8 to 21 vol. %) and inert gas. Steam is frequently used as the inert gas although since air is usually employed as the source of oxygen, nitrogen is most often present as an inert gas component. Temperatures employed are within the range of 200.degree.-400.degree. C. and a pressure of from about 1-10 atm. is commonly used. Contact times are usually on the order of 0.4 to about 15 seconds depending upon the temperature employed and the efficiency of the particular catalyst.

The present invention is the vapor phase oxidation of acrolein to acrylic acid over a catalyst consisting essentially of molybdenum, vanadium, chromium, copper and at least one of tantalum, titanium and niobium. Silicon may also be employed. All of these elements are present in the form of their oxides and the catalyst is preferably employed on an inert support. The operable ratio of the atoms of the above elements in the catalyst of this invention is

where M is tantalum, titanium, niobium or mixtures thereof. A preferred catalyst is one containing the above atomic ratios but wherein M is Ti and is employed at a ratio of about 1-3. A more preferred catalyst is one having a composition of

it being understood that silicon is optional in both operable and preferred catalysts.

The tantalum pentoxide which is employed as a component is insoluble and difficult to incorporate into the catalyst mixture and as a result is generally slurried in a solution of the other components. A particularly preferred method of making a solution of catalyst components is to employ the tantalum pentoxide as a colloidal suspension. The commercially available oxide powder is first run through a colloid mill prior to mixing with the silica (when silica is employed) and then added to the remaining components as described in Examples 1 and 16. The use of the finely divided colloid produces a superior catalyst.

As previously mentioned, it is known in the art that the amount of catalyst used will depend, at least in part, upon the porosity and surface area of the particular inert support employed. In the present invention the total amount of catalytic oxides burdened on the support is from about 13% to about 30% by weight based on the combined weight of catalyst and support. A preferred burden is from about 18% to about 23%. The preferred support is alumina and the surface area of the preferred support should be not more than 2 m.sup.2 /g with a porosity of 35-65%, 90% of the pores being in the diameter range of 50-1500 microns.

In a representative preparation of the catalyst of the present invention, the first solution was prepared by heating 1400 mls of distilled water to which was added 172.7 gms of ammonium molybdate, 43.9 gms of ammonium meta-vanadate, and 6.0 gms of ammonium dichromate. A second solution was prepared by adding 43.9 gms of cupric nitrate to 75 mls of distilled water which had been acidified with 3 mls of concentrated nitric acid. A third solution was prepared by adding 27.5 gms of tantalum pentoxide to 28.3 mls of Ludox L.S. (a 30% colloidal solution of silica).

The second solution was added drop-wise to the first solution with stirring and heating. Upon completion of this addition, the third solution was added to the other two with continued heat and stirring. The carrier support (Norton S.A. 5205, 1/4-inch spheres of alumina) was preheated in an oven at 150.degree. C. and then added to the composite solution above with continued heating and stirring to remove the excess water. When most of the water had been removed, the carrier was placed in an oven at 150.degree. C. for 1 hour to dry it. The dried catalyst was then calcined for a period of about 6.5 hours, the temperature being gradually increased from about 200.degree. to 400.degree. C. over a period of about 1.5 hours. The temperature was then maintained at 400.degree. C. for the remainder of the time. The finished catalyst was removed and cooled prior to use. The catalyst prepared as above contained about 19.2 wt. % of the metal oxides on the catalyst support, and the atomic ratios of the metals were

In the manner of Example 1, another catalyst was prepared in which titanium was substituted for tantalum. After the first two solutions had been combined as in Example 1, 7.3 g of titanium dioxide was added to the combination solution while heating and stirring was continued. The remaining steps of making the catalyst were carried out as above. The catalyst contained 19.2 wt. % of the metal oxides on the support and had the following atomic ratio of metals:

The catalyst of Example 1 was placed in a stainless steel reactor made of a 1-inch tube 10 feet long within a concentric pipe which contained a heat exchange fluid for temperature control. The feed to the reactor contained 5.6 mole % acrolein, 30 mole % nitrogen (as diluent) and oxygen; the oxygen to acrolein mole ratio being 1.48. Temperature in the reactor was 300.degree. C. Contact time was 2.9 seconds. This feed stream when passed through the reactor over the above catalyst resulted in a conversion of 99% of the acrolein and gave a selectivity to acrylic acid of 93.4% with 4.4% going to carbon oxides. In like manner, catalysts of varying compositions were tested. The mole % acrolein in the feed of each of the following Examples was in the range of from about 4.5 to about 5.5%. The results are shown in Table I.

In the manner of Example 1A other catalysts were prepared containing various amounts of titanium. These were tested as in Example 2 and the results given in Table II. All components are the same as in Example 1A except titanium, the atomic ratio of which is given in the column labeled (x).

That titanium oxide in the compositions of the invention is equivalent to or better than the oxides of tantalum or niobium can be seen by comparing the examples in Tables I and II. Of particular interest are Examples 3, 5 and 8 in Table I and G and E in Table II.

In order to show that each of vanadium, chromium, copper and tantalum or niobium is necessary to assure the good performance of the catalyst of the present invention, catalyst compostions substantially identical to that of Example 5, save one of the above components, were run in the manner of Example 2. Results are shown in Table III wherein catalysts outside the scope of the present invention were employed.

As can be seen from Table III, omitting any one of vanadium, chromium, copper and tantalum will give both lowered conversions and selectivities as well as high carbon losses.

In a preferred preparation of the catalyst of the present invention, a first solution was prepared by heating 6000 ml of distilled water to which was added 1427.6 g of ammonium molybdate, 362.8 g of ammonium metavanadate, and 49.0 g of ammonium dichromate. A second solution was prepared by adding 363 g of cupric nitrate to 150 ml of distilled water which had been acidified with 20 ml of concentrated nitric acid. A third solution was prepared by adding 226.4 g colloidal tantalum pentoxides, less than 1 micron in size, to 500 ml water.

The second solution was added drop-wise to the first solution with stirring and heating. Upon the completion of this addition, the third solution was added to the other two with continued heat and stirring. The carrier support (Norton S.A. 5205, 1/4-inch spheres of alumina) was preheated in an oven at 150.degree. C. and then added to the composite solution above with continued heating and stirring to remove the excess water. When most of the water had been removed, the carrier was placed in an oven at 150.degree. C. for 1 hour to dry it. The dried catalyst was then calcined for a period of about 6.5 hours, the temperature being gradually increased from about 200.degree. to 400.degree. C. over a period of about 1.5 hours. The temperature was then maintained at 400.degree. C. for the remainder of the time. The finished catalyst was removed and cooled prior to use. The catalyst prepared as above contained about 22.3 wt. % of the metal oxides on the catalyst support, and the atomic ratios of the metals were

The above catalyst, when employed to oxidize acrolein in the manner of Example 2, gave a 99.8% conversion of acrolein and a selectivity of 96.5% to acrylic acid.

A catalyst for the oxidation of propylene to acrolein disclosed by the present inventors in an application filed in the U.S. Patent and Trademark Office, Dec. 15, 1975, (Ser. No. 640,616 now U.S. Pat. No. 4,049,577) was employed in conjunction with a catalyst of the present invention in sequential operation in the following manner:

A metal oxide composition containing atomic metal ratios of Mo.sub.15 Co.sub.6.7 Fe.sub.1.35 Bi.sub.1.3 K.sub.0.011 formed into pellets 1/4-inch diameter and 1/4-inch thick was placed into a first column, stainless steel tubular reactor 12 ft. tall and 1.25-inch diameter. A conduit connected this first column with a second column 19 ft. long.times.1.5-inch I.D. of the same stainless steel into which was packed the supported catalyst of Example 16. Each column was heated by means of a heat exchange fluid flowing through an outer concentric tube. To the first column was fed a gas stream containing 5.3 mole percent propylene, 9.4 mole percent oxygen, and balance nitrogen at a pressure of 28.7 psig and a flow rate of 12.5 lb/hr to obtain a contact time of 2.19 seconds. The feed was preheated to a temperature of 325.degree. C. and the first reaction column was maintained at 331.degree. C. The exit stream was passed through the conduit to the second column which was maintained at 279.degree. C. Contact time in the second column was 4.87 seconds. The exit gases from the second column were quenched and the non-condensible gases were recycled to the first column. Conversion of propylene was 98.3% and yield to acrylic acid was 89%.

Each of the catalysts employed in the preceding 17 examples was adhered to a spherical alumina support of about 1/4-inch diameter. In Table IV the support employed for each catalyst and the loading (weight % catalyst) employed based on total weight of catalyst and support) is given. Supports A, B and C had different porosities and surface areas as follows:

A=49-55% porosity; 90% of the pores in 50-1500 microns rangs; 0.005-0.5 m.sup.2 /g surface area.

B=60.2% porosity; 90% of the pores in 50-1500 microns range; 0.024 m.sup.2 /g surface area.

C=38-42% porosity; 90% of the pores in 50-1500 microns range; 0.005-0.04 m.sup.2 /g surface area.