CA1278306C - Olefin epoxidation in a polar medium - Google Patents

Abstract

OLEFIN EPOXIDATION IN A POLAR MEDIUM

ABSTRACT OF THE DISCLOSURE

A hydroperoxide charge stock (t-butyl hydroperox-ide or t-amyl hydroperoxide) is reacted with a C3 to C20 olefin charge stock in liquid phase in a reaction zone in the presence of a catalytically effective amount of a solu-ble molybdenum catalyst to form a product olefin epoxide corresponding to the olefin charge stock and a product alcohol corresponding to the hydroperoxide charge (t-butyl alcohol or t-amyl alcohol), which process is improved in accordance with the present invention by maintaining a reac-tion medium composed of more than 60 wt.% of polar components (hydroperoxide charge stock, product alcohol and product epoxide) in the reaction zone by charging to the reaction zone at least about a 30 wt.% solution of the hydroperoxide charge stock in the corresponding product alcohol and charg-ing said olefin charge stock to said reaction zone in an amount relative to the amount of said charged solution of charged hydroperoxide in product alcohol sufficient to pro-vide a ratio of from about 0.5 to about 2 moles of charged olefin per mole of charged hydroperoxide.
The preferred olefin charge stock is propylene and the preferred hydroperoxide charge stock is t-butyl hydro-peroxide. The corresponding epoxide in this situation is propylene oxide and the corresponding product alcohol is t-butyl alcohol.

Description

~27~33~6 OLEFIN EPOXIDATION IN A POLAR MEDIUM
(D~80,313-C1) BACKGROUND OF THE INVENTION

Field of the Invention This invention relates to the molybdenum-catalyzed epoxidation of C3 to C20 olefins with tertiary butyl hydroperoxide or tertiary amyl hydroperoxide in liquid phase in a polar reaction medium.

Prior Art The epoxldation o~ olefins to give various oxide compounds has long been an area of study by those skilled in the art. It is well known that the reac~ivities of the various olefins differs wlth the number of substituents on the carbon atoms involved in the double bond. Ethylene ' A

~7~33~6 itself has the lowest relative rate of epoxidation, with propylene and other alpha olefins being the next slowest.
Compounds of the formula R2C=CR2 where R simply represents alkyl or other substituents may ~e epoxidized fastest.
S Of course, the production of ethylene oxide from ethylene has long b en known to be accomplished by reaction with molecular oxygen over a silver catalyst. Numerous patents have issued on various silver-catalyzed processes for the production of ethylene oxide.
Unfortunately, ~he silver catalyst route is poor for olefins other than ethylene. For a long time the com-mercial production of propylene oxide could only be accom-plished via the cumbersome chlorohydrin process.
Another commercial process for the manufacture of substituted oxides from alpha olefins such as propylene was not discovered until U. S. Patent 3,351,635 taught that an organic oxide compound could be made by reacting an olefini-cally unsaturated compound with an organic hydroperoxide in the presence of a molybdenum, tungsten, titanium, columbium, tantalum, rhenium, selenium, chromium, zirconium, tellurium or uranium catalyst. U. S. Patent 3,350,422 teaches a simi-lar process using a soluble vanadium ca~alyst. ~olybdenum is the preferred catalyst. A substantial excess of olefin relative to the hydroperoxide is taught as the normal pro~
25 cedure for the reaction. See also U. S. Patent 3,526,645 -which teaches the slow addition of organic hydroperoxide to an excess of olefin as preferred.
However, even though this work was racognized as extremely important in the development of a commer~ial ~2~7~3~

propylene oxide process that did not depend on the chloro-hydrin route, it has been recognized that the molybdenum process has a number of problems. For example, large quan-tities of the alcohol corresponding to the peroxide used were formed; if t-butyl hydroperoxide was used as a co-reactant, then a use or market for t-butyl alcohol is required. With propylene, various undesirable propylene dimers, sometimes called hexenes, are formed. Besides being unde~irable in tha~ propylene is consumed, problems are caused in separating the desired propylene oxide from the product mix. In addition, the molybdenum catalyst may not be stable or the recovery of the catalyst for recycle may be poor.
A number of other me~hods for the production of alkylene oxides from epoxidizing olefins (particularly propylene) have been proposed. U. S. Patent 3,666,777 to Sargenti reveals a process for epoxidizing propylene using a molybdenum-containing epoxidation catalyst solution prepared by heating molybdenum powder with a stream containing unre-acted tertiary butyl hydroperoxide used in the epoxidationprocess as the oxidizing agent and polyhydric compounds.
The polyhydric compounds are to have a molecular weight from 200 to 300 and are to be formed a~ a by-product in the epoxi-dation process. A process for preparing propylene oxide by direct oxidation of propylene with an organic hydroperoxide in the presence of a catalyst (such as molybdenum or vanad-ium) is described in British Patent 1,338,015 to Atlantic-Richfield. The improvemen~ therein resides in the inclusion of a free radical inhibitor in the reaction mixtuxe ~o help eliminate the formation of C5 to C7 hydrocarbon by-products which must be removed by extractive distillation. Proposed free radical inhibitors are tertiary butyl catechol and 2,6-di-t-butyl-4-methyl phenol.
Stein, et al. in U. S. Patent 3,849,451 have improved upon the Kollar process of U. S. Patents 3,350,422 and 3,351,~35 by requiring a close control of the reaction temperature, between 90-200C and autogeneous pressures, among other parameters. Stein et al. also suggest the use of several reaction vessels with somewhat higher temperature in the last zones to insure more complete reaction. The primary benefits seem to be improved yields and reduced side reactions. Prescher et al. in U. S. Reissue Patent No.
Re.31,381 disclose a process for the preparation of propylene oxide from propylene and hydrogen peroxide wherein plural reactors such as stirred kettles, tubular reactors and loop reactors may be used~ They recommend, as an example, the use of a train of several stirred ket~les, such as a cascade of 3 to 6 kettle reactors or the use of 1 to 3 stirred kettles arranged in series followed by a tubular reactor~
Russell U. S. Patent No. 3,418,430 discloses a process for producing propylene oxide by reacting propylene with an organic hydroperoxide in solvent solution in the presence of a metallic epoxidation catalyst, such as a compound of molybdenum at a mole ratio of propylene to hydro-peroxide of 0.5:1 to 100:1 (preferably 2:1 to 10:1) at a temperature of -20 to 200C (preferably 50-120C) and a pressure of about atmospheric to 1000 psia, with a low olefin conversion per pass (e~g., 10-30~) whexein unreacted oxygen is removed from the unreacted propylene.

Sheng et al. U. S. Patent No. 3,434,975 discloses a method for making molybdenum compounds useful to catalyze the reaction of olefins with organic hydroperoxides wherein metallic molybdenum is reacted with an organic hydroperoxide, such as tertiary butyl hydroperoxide, a peracid or hydrogen peroxide in the presence of a saturated Cl-C4 alcohol.
The molybdenum-cataly~ed epoxidation of alpha olefins and alpha substituted olefins with relatively less ~table hydroperoxides may be accomplished according ~o U. S.
10 Patent 3,862,961 to Sheng, et al. by employing a cxitical amount of a stabilizing agent consisting of a C3 to Cg secondary or tertiary mcnohydric alcohol. The preferred alcohol seems to be tertiary butyl alcohol. Citric acid is used to minimize the iron-ca~alyzed decomposition of the organic hydroperoxide without adversely affecting the reac-tion between the hydroperoxide and the olefin in a similar oxirane producing proce$s taught by Herzog in U. S. Patent 3,928,393. ~he inventors in U. S. Patent 4,217,287 discov-ered that i barium oxide is present in the reaction mixture, the cataly~ic epoxidation of olefins with organic hydroper-oxides can be successfully carried out with good selectivity to the epoxide based on hydroperoxide converted when a relatively low olefin to hydroperoxide mole ratio is used.
The alpha-olefinically unsaturated compound must be added ~5 incrementally to the organic hydroperoxide to provide an excess of hydroperoxide khat is effective.
Selective epoxidation of olefins with cumene hydroperoxide (CHP) can be accomplished at high CHP to olefin ratios if barium oxide is present with the molybdenum catalyst as reported by Wu and Swift in "Selective Olefin ~27B~

Epoxidation at High Hydroperoxide to Olefin Ratios,~ ournal _ Catalysis, Vol. 43, 380-383 (1976).
Catalysts other than molybdenum have been tried.
Copper polyphthalocyanine which has been activated by contact with an aromatic heterocyclic amine is an effectiYe catalyst for t.he oxidation of certain aliphatic and alicyclic compounds (propylene, for instance) as discovered by Brownstein, e~ al.
de~cribed in U. S. Patent 4,028~423.
Various methods for preparing molybdenum ca~alysts useful in these olefin epoxidation methods are described in the following patents: U. S. 3~362,972 to Kollar; U. S.
3,480,563 to Bonetti, et al.; U. S. 3,578,690 to Becker;
U. S. 3,953,362 and U. S. 4,009,122 both to Lines, et al.
It has also been proposed to use the tertiary butyl alcohol that is formed when propylene is reacted with tertiary butyl hydrop~roxide as an intermediate in the syn-thesis of another organic compound. Thus, Schneider, in U. S. Patent No. 3,801,667, proposes a method for the prepa-ration of isoprene wherein, as ~he second step of a six step process, tertiarybutyl hydroperoxide is reacted with propyl-ene in accordance wi~h U. S. Patent No. 3,418,340 to provide tertiary butyl alcohol. Connor et al. in U. S. Patent No.
3,836,603 propose to use the tertiary butyl alcohol as an intermediate in a multi-step process for the manufacture of p-xylene.
Also pertinent to the subject discov~ry are those patents which address schemes for separating propylene oxide from the other by-products produce.d. These patents demon-strate a high concern for separating out the useful propylene oxide from the close boiling hexene oligomers. Tt would be ~27B~

a great progression in the art if a method could be devised where the oligomer by-products would be produced not at all or in such low proportions that a separate separation step would not be necessary as in these patents.
U. S. Patent 3,464,897 addresses the separation of propylene oxide from other hydrocarbons having boiling points close to propylene oxide by distilling the mixturP in the presence of an open chain or cyclic paraffin containing from 8 to 12 carbon atoms. Similarly, prspylene oxide can be separated from water using identical entrainers as dis-closed in U. S. Patent 3,607,669. Propylene oxide is puri-fied from its by-products by fractionation in the presence of a hydrocarbon having from 8 to 20 carbon atoms according to U. S. Patent 3,843,488. Additionally, U. S. Patent 15 3,909,366 teaches that propylene oxide may be purified with respect to contaminating paraffinic and olefinic hydocarbons by extractive distillation in the presence of an aromatic hydrocarbon havin~ from 6 to 12 carbon atoms.

SUMMARY OF T~ INVENTION
This invention is directed to a process wherein a hydroperoxide charge s~ock (t-butyl hydroperoxide or t-amyl hydroperoxide~ is reacted with a C3 to C20 olefin charge stock in liquid phase in a reaction zone in the presence of 25 a catalytically effective amount of a soluble molybdenum catalyst to form a product olefin epoxide corresponding to the olefin charge stock and a product alcohol corresponding to the hydroperoxide charge ~t-butyl alcohol or t-amyl alcohol), which process is improved in accordance with the present invention by maintaining a reaction medium composed ~7~

of more than 60 wt.~ of polar components (hydroperoxide charge stock, product alcohol and product epoxide) in the reaction zone by charging to the reaction zone at least about a 30 wt.% solution of the hydroperoxide charge stock in the corresponding product alcohol and chargi~g said ole-fin charge stock to said reaction zone in an amount relative to the amount of said charged so,ution of charged hydroper-oxide in product alcohol sufficient to provide a ratio of from about 0.5 to about 2 moles of charged olefin per mole of charged hydroperoxide.
The preferred olefin charge stock is propylene and the preferred hydroperoxide charge stock is t-butyl hydro-peroxide. The corresponding epoxide in this situation is propylene oxide and the corresponding product alcohol is t-butyl alcohol.

BACRGROUND OF THE INVENTION
Under ambient conditions t-butyl hydroperoxide and t-amyl hydroperoxide are comparatively stable materials.
However, as temperature increases, these hydroperoxides tend to become "destahilized" so that thermal and/or catalytic decomposition will be initiated leading to the formation of unwanted by-products such as ketones, lower molecular weight alcohols, tertiary alcohols, oxygen, e~c. This is a par-ticularly troublesome problem at temperatures of 50 to 180C (e.g., 100 to 130C) which are normally used when such a hydroperoxide is catalytically reacted with an olefin to form an olefin epoxide. This problem can be at least partially overcome by conducting the epoxidation reaction in the presence of an excess of the olefin reactant. However, ~ 2~7B~

the unreacted olefin must be separated from the epoxide reaction product for recycle and such separations are accom-plished with progressively more difficulty as the molecular weight of the olefin reactant increases. Problems can be encountered even with the lower molecular weight olefins and, in any event, the u~ility costs associated with the recovery and recycle of significant quantities of the olafin reactant add an appreciable burden to the cost of manufacture of the corresponding olefin epoxide and alcohol reaction products.

Further, use of excess propylene in order to in-crease reaction rate and therefore reduce the side reactions of TBHP or TAHP leads to the serious problem of propylene dimer formation. ~he formation of dimer is a second order reaction and hence is accelerated as the concen~ration of propylene increases. Also, the use of excess propylene affords a more non-polar medium which in turn tends to ren-der the molybdenum catalys~ less soluble during the reaction.

DETAILED DESCRIPTION OF THE PREFERRED EMBODXMENT

It ha~ been discovered in accordance with the present in~ention that in the production of olefin epoxides by reacting a C3~C20 olefin with t-butyl hydroperoxide or t-amyl hydroperoxide in liquid phase in the presence of a catalytically effective amount of a soluble molybdenum catalyst that an unexpectedly high selectivity to olefin epoxide, on the basis of hydroperoxide converted, can he obtained when the hydroperoxide is charged to the reaction zone in at least a 30 wt.% solution of the corresponding product alcohol and the olefin is charged to the reaction ~'7~

zone in an amount relative to the hydroperoxide charged to the reaction zone such that about 0.5 to 2 moles of olefin are charged per mole of hydroperoxide chaxged.

Reactants and Catalysts The method of this invention can be used to epoxi-dize C3-C20 olefinically unsaturated compound suchs as substituted and unsubstituted aliphatic and alicyclic ole-fins. The process i5 particularly useful in epoxidizing compounds having at least one double bond situated in the alpha position of a chain or internally. Representative compounds include propylene, normal butylene, isobutylene, pentenes, methyl pentenes, hexenes, octenes, dodecenes, cyclohexene, substituted cyclohexenes, butadiene, styrene, substituted styrenes, vinyl toluene, vinyl cyclohexene, phenyl cycloh~xenes and the like.
The invention finds its greatest utility in the epoxidation of primary or alpha olefins and propylene is epoxidized particularly advantageously by the inventive process.
It has been surprisingly discovered that the method of this invention does not work equally well for all hydro-peroxides. For example, with cumene hydroperoxide at low propylene excesses, the selectivity to propylene oxide, based on cumene hydroperoxide convexted is poor.
Tertiary butyl hydroperoxide (TB~P) and tertiary amyl hydroperoxide ~TAHP) are the hydroperoxides to be used in accordance with the present invention. Tertiary butyl hydroperoxide is preferred.

~ 3~ 68878-32 The TBHP should be charged in at least a 30 ~7t.~
solution in t-butyl alcobol, and preferably about a 40 to 75 wt.% solution.
Ca~alysts suitable for the epoxidation me~hod of this invention are molybdenum catalysts tha~ are soluble in the reaction medium.
Examples of suitable soluble catalysts include molybdenum compoundfi such as molybdenum octoate, molybdenum naphthenate, molybdenum acetyl ace~onate, molybdenum/alcohol complexes, molybdenum/glycol complexes, etc.
Other catalysts found to be useful are the molybdenum complexes of alkylene glycols with molybdenum compounds.
Briefly, ~hese complexes are made by reac~ing an ammonium-containing molybdenum compound with an alkylene glycol in the presence of water at an elevated temperature, such as about 80 to 130C. The ammonium-containing molybdenum compound is preferably ammonlum heptamolybdate tetrahydrate or ammonium dimolybdate hydrate. The alkylene glycols are preferably ethylene glycol and/or propylene glycol although others have been found to be useful.
Still other catalysts found to be useful ln the practice of the present invention are molybdenum complexes of monohydric alcohols. Briefly, an _L2 7 ~

alkanol such as 2-ethyl hexanol is reacted with a molybdenum oxide in the presence of ammonium hydroxide or by reactiny the alkanol with ammonium heptamolybdate in the presence of a controlled amount of water.

s Reaction Conditions The epoxidation reaction may be conduc~ed at a temperature in the range of 50-180C with a preferred range of between 90 and 140C. An especially preferred range is 100 to 130C with about 110C-120C being the most preferred single stage operating temperatureO
It has been discovered that the u e of only a small molar excess of olefin contributes to increased oxide concentrations, increased oxide selectivities and yields increased recoverable molybdenum. These benefits are due to the more polar reaction media (low propylene, high TB~P/TBA) which tends to stabilize TBHP and render the molybdenum catalyst more active and soluble throughout ~he entire reac-tion period. The lower ~emperatures of our invention further contribute to the catalyst's stability and prevents TB~P
decomposition via undesired pathways.
The catalyst concentrations in the method of this invention should be in the range of 50 to 1,000 ppm (0.01 to 0.10 wt.%) based on the total reactant charge. Catalyst ~5 concentration is calculated as molybdenum metal. A pre-ferred range is 200 to 600 ppm. Generally, about 250-500 ppm is the most preferred level. These catalyst levels are higher than those presently used in prior art methods, which tend to run from 50 to 200 ppm. Moreover, it has been dis-covered that the method of the present invention provides a ~.2'7~

process wherein the molybdenum catalyst is retained in solu-tion in the medium during the life of the reackionO
The epoxidation reaction of this in~ention is carried out in the presence of a polar solvent. The polar solvent should correspond to the hydroperoxide reactant (i.e., have the same carbon skeleton as the hydroperoxide).
Tertiary butyl hydroperoxide and T~A are copro-duced commercially by the oxidation of isobutane and if TBRP
is used as the hydroperoxide, TBA is the polar solvent. The TBA coproduced with the TBHP will normally ~upply all of the polar solvent required for the present invention.
It is preferred that the solution of TBHP in TBA
contain very little water, between zero and 1 wt.~. Pref-erably~ the water level should be less than 0.5 wt.%.
The reaction can be carried out to achieve a hydroperoxide conversion, typically 96 to 99~, while stillmaintaining high epoxide selectivities, typically also 96 to 99% basis the hydroperoxide reactedO For both of these values to be simultaneously so high is very unusual. This is important ~ecause the profitability of a commercial ole-fin epoxide plant, to a significant extent~ is increased as the yield of olefin epoxide increases.
The reaction time may vary considerably, from minutes to hours. Generally, the reaction times run from thirty minutes to three or four hours with 1.5-2.0 hours being about average. The preferred single stage reaction time/temperature is two hours at 110-120C. Preferably the reaction is conducted in two or more temperature stages.
The reaction procedure generally begins by charging the olefin to the reaction vessel. Next, the hydroperoxide, ~13-~2~

polar solvent and catalyst may be added and the contents heated to the desired reaction temperature. Alternatively, the olefin reactant may be heated to, at or near the prefer-red reaction ~emperature, and then the hydroperoxide, polar solvent and catalyst may be added~ Further heat may be provided by the exotherm of the reaction. The reaction is then allowed to proceed for the desired amount of time at the reaction temperature, generally 110-120C, or conducted for 1 hour at 50-120C followed by 1 hour at 120-150C. The mixture is cooled down and the oxide recovered. Generally, for the method of this invention, the oxide concentration runs from about 24-28% for propylene/TB~P mole ratio of 1.6-1.9:1 tTBHP wt.~ is 68-80%) and from about 31-32~ for propylene/TB~P mole ratio of 1.1:1-1.2:1 (TB~P content is 68-80 wt.%).
A series of reactors helps to achieve the objec-tives of high reaction medium polarity and low olefin concentration. The use of staged reactors makes it possible to stage the addition of olefin to thereby increase reactor medium polarity and in the case of propylene, to fur~her decrease the formation of propylene dimer. This concept can be further improved by using a continuously stirred tank ractor (CSTR) or a series of CSTR's because a CSTR inherently provides a lower concentration of reactants than a plug flow reactor (PFR).
A more effective approach is to use a CSTR or a series of CSTR's followed by one or more plug flow reactors because conversion can be more effectively forced to comple-tion in a plugged flow reactor.

It is possible and, indeed, desirable ~o operate each stage at a progressively higher temperature.
As an example, the CS~R may be operated at a temperature in the range of about 70 to 115C, 90 to 115C
being preferred, with 100-110C as the most preferred reac-tion temperature range. The PFR should be operated at a higher temperature, from over 115C to 150C, with 120-140C
as the most preferred range. The plug from reactor can be of any of a number of designs known to those skill~d in the art such as jacketed reactors, with heat transfer, adiabatic reactors and combinations thereof. The effluent from the CSTR may be termed an intermediate reaction mixture since the reaction is not complete. The residenc2 time of the reactants in each reactor is left to the operator although it is preferred that they be adjusted so that about 30 wt.%

to about 50 wt.% of the TBHP is converted in the CSTR.
AYerage residence times in the CSTR and the PFR will be adjusted in the manner known to those of ordinary skill in the art, based on the other reaction conditions such as catalyt concentrations, reaction temperatures, etc.

PREPARATION OF _ ROPYLE~r DXID--/ROM PRO ~L--NE
It has been discovered that under the reaction conditions or the present invention propylene oxide can be produced at high concentrations ~24-32%), high selectivi~ies (96-99%) on the basis of t-butyl hydroperoxide conver~ed and high yields (94-98%) of propylene oxide produced on the basis of t-butyl hydroperoxide charged. One particularly preferred set of operating conditions, especially for a 3~

con~inuous process include charging the propylene and hydro-peroxide reactants at a low molar ratio of propylene to hydroperoxide (e.g., about 0.5 to about 2 moles of propylene charge per mole of hydroperoxide charge). Another preferred procedure is the use of staged temperatures so tha~ the first 0.5 to 1.5 hours of the reaction are conduc~ed a~ a lower temperature t50-120~C) and the second stage also usually an 0.5 to about 1.5 hour reaction time, is conducted at a higher temperature (usually 120-150C).
A low molar ratio of propylene to TBHP is further aided by optional staged addition of prvpylene to a staged plurality of reactors. By this technique~ buildup of olefin at any one point in the staged series of reac~ors, relative to the hydroperoxide, is minimized.
Normally, the charge ra~io of propylene to hydro-peroxide is thought to be variable over the range of from about 2:1 to 20:11 expressed as a mole ratio. An initial mole ratio of olefin to hydroperoxide of less than 2:1 has been thollght to be undesirable because of a loss of selec-tivity. In this invention, the initial mole ratio of olefin to hydroperoxide of the feed should not exceed 2.0:1. The broad range, expressed in terms of the charge rates of the propylene and the TB~P to a continuously stirred tank reac-tor is from 0.5:1 to 2.0:1, and preferably from 0.9:1 to 1.8:1. Most preferably, the mole ratio of olefin to hydro-peroxide in the feed is 1.05:1 to 1.35:1.
When excess propylene is charged, the ratio of propylene to TBHP in the CSTR will be different from ~he initial charge ratio because both propylene and TBHP are consumed in the reaction that takes place. In ~his case, as ~2~7~33~

the TB~P conversion increases, so does the ratio of propyl-ene to TB~P. For example, if the initial molar feed ratio of the charged propylen and TBHP i5 1.15 moles of propylene per mole of TBHP, and if the rate of withdrawal of reaction medium from the CSTR is such that about a 50% conversion of ~he TBHP is main~ained in the CSTR, the average mole ratio of unreacted propylQne to unreacted TBHP will be about 1.3:1. If the rate of withdrawal of the reaction medium is such that about a 90% conversion of TBHP is main~ained in the CSTR, the average mcle ratio of unreacted propylene to unreacted TBHP will be about 2.5:1.
In this same situation, and if it be assumed tha~
the TBHP is charged as a 70 wt.% solution of TBHP in tertiary butyl alcohol (TBA), the charge to the CSTR will be composed of about 72.7 wt.~ of polar materials (the sum of the weights of TBHP and TBA charged, divided by the sum of the weights of propylene, TBHP and TBA charged.) During the course of the reaction, the propylene (a non-polar material) is converted to propylene oxide (a polar material) so ~hat at the 50% TB~P conversion level mentioned above, the reac-tion medium will be composed of about 84.6 wt.~ of polar materials (the s~m of the weights of unreactad TB~P, TBA
charged, TBA formed as a reaction product and propylene oxide divided by the said sum of these four materials and unreacted propylene). A~ the 90% TBHP conversion level mentioned above, the reaction mediwm will be composed of about 94 wt.~ of polar materials on thi 5 same basis.
The method and apparatus of this invention are illustrated but not limited by the following examples.

REACTION MEDIUM POLARITY
Example l In order to demonstrate the importance of the polari~y of the reaction medium in the practice of the pr~cess of the present invention, three series of batch runs were made using propylene, propane, ~ertiary bu~yl hydro-peroxide and ~er~iary bu~yl alcohol as feed ma~erials.
In all of the runs, the catalyst that was used was a molybdenum/ethylene glycol complex prepared ~s follows:

Catalyst Prepara ion To a one=liter round-bottomed Morton flask fitted with a mechanical stirrer, a nitrogen inlet, a thermometer, a Dean Stark trap, a condenser and a nitrogen bubbler, were added 100 g. of ammonium heptamolybdate tetrahydrat~ and 300 g of ethylene glycol~ The reaction mixture was heated to 85-110C for a~out 1 hour with nitrogen slowly passin~
through the flask. At the end of that time, the reaction was essentially complete and essentially all of the ammonium heptamolybdate was dissolved~ ~he reaction mixture was subjected to an aspirator vacuum at a temperature of abou~
85-95C for about 1.5 hours and then reheated to 90-100C
for an additional hour. On cooling, there was obtained 2 clear liquid catalyst composition conkaining 16.1% molyb-denum by Atomic Absoxption spectroscopy, 1.17~ nitrogen(Xj~ldahl) and 1.67% water (Karl Fisher analysis).

Epoxidation Runs The epoxidation runs summarized in Tables 1 and 2 where made in a 300 ml. stainless steel autoclave~ The 3~

propylene feed component was charged at ambient temperature and ~hen the t-butyl hydroperoxide (TBHP) feed component was charged premixed with 0.38 grams of the catalyst. This provided for a catalyst concentration of about 350 ppm of catalyst in the reaction medium. For the pure Propylene runs, the TBHP feed component consisted of abou~ a 72.36 wt.~ solution of TB~P in t-butyl alcohol which contained about 0.2 wt.~ of water. For the Propylene/Propane runs wherein propane was added to the propylene feed component and for the Propylene/TBA runs wherein additional t-butyl alcohol was added to the TBHP feed componentl the TBHP feed component consisted of a 73.0 wt.% solu~ion of TB~P in t-butyl alcohol that contained about 0.2 wt.% of water. The quantities of feed component were adjusted for each of the lS runs in order to provide the desired mole ratio of propylene to TBHP shown in Tables 1 and 2.
Thus, by way of example, in Run No. 1 of Table 1, the propylene feed component consisted of about 49.4 grams of propylene and the T~HP feed component consisted of about 20 93.36 g. of TB~IP, about 35.4 g. of t-butyl alcohol, about 0.2S gram of water and about 0.38 gram of catalyst.
In Run No. 2 of Table 1, the propylene feed compo-nent consisted of about 34.65 grams of propylene and about 35.45 grams of propane. The TB~P feed component consisted 25 of about 71.72 grams of TB~Pt about 26.33 g. of t-butyl alcohol, about 0.2 g. of water and about 0.38 g. of catalyst.
In Run No. 3 of Table 1, the propylene feed compo-nent consisted of about 36.4 g. of propylene and the TBHP
feed component consisted of about 74.36 g. of TBHP, about _l9W

~ ~7~33~$

64.27 g. of t-butyl alcohol, about 0.2 g. of water and about 0.38 g. of catalyst.
All of the runs reported in Table 1 were conducted at 120C for about 2.0 hrs. All of the runs reported in Table 2 were conducted at 110C for 1.0 hour and 130C for 1.0 hour.
The reactants ~mployed and the results obtained are reported in Tables 1 and 2.

. ~ ~ ~ 0 ~ D O CO ~~ ~ O ~n o ~ , w ~ ~ o r-i ~ .~
~ ... ... o~ a, Q t3 ~ n ~ N ~ N ~ I O _I 0 1~ 0 I
CO _I co CO O Cl~al o o l~ o a~ ~ o o~ 0 Lr) o ~ D ~ m ~ O ~r o ~ ~ I~
o L~ r L ) c~ L~~D CO CO r~ a~ oo co ~ o~ o~ CD

LO N a:~ N O ~) O N ~D _1 ~r ~1 ~ 0~ ~ t:l~ CO ~ O
ra f~ ~ ~ N O ~ LO_i ~ Ln ~ r ~ L~

LY o~ Ll~xcr u~ D O O~
L~ N O N ri O ~) r-i O N a~ 0~ 0 ~ Ln OD
~ ~ 1 N ~ (~ ~ ~ ~ O

~ ~ ~ o o a~ ~ ~ _1 ~ ~ ~ ~D ln ~ o ,l c~ o ~

~ ~i rl r~~ O ~ o O

~~ O

LO u r- c~ o ~ Nrll ~r Ll~ I CD O
L~

~ ~q ~ DO

~ ~ N ,~.~

o~ In o o ~ o c~
_1 0 _~ 0 0 ~1 0 0 ~ ~ ~ ~ ot`
2~ w ~ ~ o~ o a~ o dP Q~ O ~ r o ~ t ~ r~ 0 ~r 9~ u~ o ~ ~ N

~ ~ ~ I o ~ O O ~ O
~ CQ ;
.0 ~
~1 ~J o o O N r~ 1 0 N ~ ~
00 ~

a~
_~

~ ~ ~1 ~J N ~ N N

3~

Turning first to Table 1, it will be noted that the runs have been arranged in "tripletsn, based on the mole ratio of propylene to TB~P, with each set of "triplets~
having a progressively increasing mole ratio of propylene to ~BHP. With reference to Column 4, reporting the results obtained in terms of propylene oxide selectivity, based on TBHP, it will be noted that when the polarity of the charge was reduced in Run No. 1 through the addition of a mole of propane, so that the polar components of the charge ~TBHP
and TBA) consti~uted only about 5802 wt~% of the charge, there was a significan~ loss of selectivity, as compared with Run No~ 2 of the present invention wherQ the polar components of the charge constituted abou~ 74.6 wt.%. Run No. 3 wherein the polarity of the charge was increased through the use of an equivalent amoun~ of additional TBA, based on the amount of propane u~ed in Run No. 1, so the t~e polar components of the charge constituted about 80.6 wt.%
of the charge demonstrates that the poor selactivity of Run No. 1 was due to a reduction in the polarity of the charge rather than a ~dilu~ion~ of ~he reactants. ~un 3 had the same "dilution" realtive to Run 2, as did Run 1.
The same effec~ on propylene oxide selectivity basis TB~P reacted is noted in the second~ third and fourth set of data "triplets" (Runs 4-12).
~5 For the fif~h set of triplets ~Runs 13-15), it can be ca].culated that only 40.9 wt.% of the charge components were polar for Run No. 13 and that only 58.3 wt.% of the charge components were polar for Run No. 14, but essentially equivalent results were obtained. As shown by Run No. 15, increasing the percentage of polar components in the charge to 72.8 wt~ only marginally improved the selectivity of the TBHP to propylene oxide. At higher initial propylene to TBHP ratios (1.9-2.1 to 1) the effect of polar media sta-bilization of TBHP and catalyst is "washed-out" by the increased rates of reaction (less TB~P) product by increased propylene concentration. The penalty for increased propyl ene concentration is increased propylene dimer ~ade as evidenced by results in col D 8.

In the next set of data (Runs 16-19), the polar components constituted about 32.2 wt.% , 52.2 wt.%, and 47.1 wt.% in runs 16-18. Essentially equivalent results were obtained. There w~s no improvement in the selectivity of the TBHP to propylene oxide in Run No. 19 wh~n the polar components constituted about 67.1 w~.% of the charge. Note from Column 8, however, that there was a fur~her significant increase in the amount of propylene dimer that was formed at the higher mole ratios of propylene to TBHP of-Runs 16-21 which are outside the scope of the present invention. In runs 16-21 the selectivities of TBHP to propylene oxide were essentially equal because of the "wash" carried by sharply increased reaction rates (less TBHP decomposition becau~e of increased propylene to TBHP mole ratios).
Turning next to Table 2, it will be seen that the pattern is repeated, in Runs 22, 25 and 28 when the polar components constituted 58.4 wt~%, 54.7 wt.% and 52.2 wt.~, respectively, of the charge component~, the selectivity to the TBHP to propylene oxide (Column 4) was significantly less than the selectivity obtained in Runs 23, 26 and 27 of the present invention where the polar components constituted 74.5 wt.~, 71.6 wt.% and 69.6 wt.%, respectively, of the -~4 ~27B3~

charge. The selectivities of Runs 22, 25 and 28 were also significantly less than the selectivities obtained in Runs 24, 27 and 30 where the polar components constituted 79.3 wt.%, 78.2 wt.% and 76.0 wt.%, respectively of the charge.

Examples 2-4 show the importance of relatively high catalyst concentrations on the yields, selectivities, and conversions when the epoxidations are conducted at low reaction temperatures.
To a 300 ml 316 stainless steel autoclave (purged with nitrogen) was added 43.9g (1.0452 moles) of propylene at room temperature. Also at room temperature was added a premixed solution of 88.2g of TBHP (consisting of 60.75%
TBH~, 38.91% TBA and 0.34% water) and O.9g of molybdenum 2-ethyl-l-hexanol (5.96~ molybdenum) catalyst. The molybdenum 2-ethyl-1-hexanol catalyst was made by heating 29.0g of molybdenum trioxide with 299.5g 2-ethyl-1-hexanol and 20 ml concentrated NH40H and 250 ml toluene from room temperature to 140C over a 1-3/4 hour period removiny about 9 ml water and 165 ml toluene. The heating continued from 140C to 153C for another 4.25 hours at which poin~ about 16 ml water and 234 ml toluene had been recovered. The reaction mixture was fil~ered and the fil~rate appeared ~o contain water so it was dried ove~ molecular sieves. The dried material was refiltered and analyzed for molybdenum and found to contain 5.96~ molybdenum. This was the O.9g of catalyst that was premixed with the TBHP solution and charged to the autoclave at room temperature after the propylene was added. The mole ratio of propylene/TBHP in this run was 1.75:1 and TB~P/TBA was 1.2801 and the catalyst level was 0.0403 wt.% molybdenum basis total reactor charge.
The autoclave was heated with stirring to 110C over a 30 minute period and held at 110C for 90 minutes. The reac-tion mixture was cooled to room temperature and sampled under pressure. The total weight of the product recovered was 133.0g and the total weight of the liquid product (after propylene was stripped) was 93.0g.
The liquid produc~ was analyzed and found ~o con-tain 1.18~ TBHP.

Grams of TB~P remaining o 1.0974g moles of TBHP remaining - 0.0122 moles TBHP reacted = moles fed - moles remaining moles TB~P reac~ed = 0.5954 - 0.0122 = 0.5832 Conversion TB~P = moles reacted = 0.5832 = 97.94%
moles fed 0.5954 The total product analyzed under pressure was found to contain 24.729 wt.% propylene oxide and 0.152 wt.%
propylene glycol. It should further be noted that the total product contains only 13O922~ propylene unreacted.

Grams propylene oxide = 32.8896g moles PO = 0.5671 5 = = moles PO = 0.5671 = 97.23%
moles TBHP 0.5832 reacted Yield of PO ~= moles PO = 0O5671 = 95.24 moles TBHP 0.5954 fed The same liquid product was also analyzed by atomic absorption spectroscopy and found to contain 526 ppm molybdenum, or a 91.5% molybdenum recovery.

~2~d~3q:~

To a 300 ml 316 stainlPss steel au~oclave (purged with nitrogen) was added 45.3g (1.07857 moles~ of propylene.
Also at room temperature was added a premixed solution of 88.15g of TBHP (consisting of 60.75% TBHP, 38.91% TBA, and 0.34% water) and 0.45g of molybdenum 2-e~hyl-1-hexanol cata-lyst (5.96% molybdenum) whose preparation was described in Example 2. In this epoxidation the mole ratio of propylene to TBHP was 1.81:1 and the mole ratio of TB~P/~BA was 1.28:1 and the amount of moly catalys~ was 0.0200 wt.~ basis total reactor charge. The amount of catalyst used here is about half that used in Example 2. ~ere again the low propylene to TBHP ratio leads to sharply increased molybdenum recoveries.
The autoclave was heated with stirring to 110C
over a 30 minute period and held at 110C for 90 minutes.
The reaction mixture wa~ cooled to room temperature and sampled under pressure. The total weight of the product re-covered was 133.9g and the weight of the liquid product (after propylene was stripped) was 94.2g~ The liquid prod-uct was analyzed and contained 264 ppm molybdenum or a 93.0 molybdenum recovery.
The liquid product was analyzed and found to still contain 3.92% TBHP.
Grams of TBHP remaining - 94.2 x 3.92% = 3.69264g moles TBHP remaining - 0.0410 moles moles TBHP reacted = moles fed - moles remaining moles TBHP reacted = 0.59S0 - 0.0410 0.5540 Conversion TBHP = moles reacted = 0.5540 = 93.11%
moles fed 0.5950 -~7-3~

The total product analyzed under pressure ~7as found to contain 22.928 wt.~ propylene oxide and 16.14%
unreacted propylene.

grams propylene oxide = 133.9 x 22.928~ = 30.70g moles propylene oxide = 0.5293 moles Selectivity to PO = moles PO = 0u5293 = 9S.54%
moles TB~P 0.5540 reacted Yield PO = moles PO = 0.5293 = 88O96 moles T~P fed 0.5950 In essentially identical runs, except that the catalys~ concentration was reduced in Example 3, the yield of propylene oxide was some 6~ lower ~han in Example 2 with the higher catalyst concentration. Note, however, that the molybdenum recovery (soluble molybdenum) in the reactor effluent in Example 3 was very high (93.0%) and even higher than in ~xample 2 ~91.5%).

To a nitrogen purged 300 ml 316 stainless steel autoclave was added 48.3g (1.1500 moles) of propylene at room temperature. To the propylene was added a premixed solution of TBHP (124.2g) and molybdenum catalyst (1.2 g).

The TBHP part of the premixed TBHP/molybdenum catalyst solu-tion consisted of 124.2g having the following composition:
60.50~ TBHP, 39.30% TBA and 0.2% water. The mQlybdenum catalyst part of the premixed ~B~P/molybdenum catalyst ~olu-tion consisted of 1.2g of the molybdenum 2-ethyl-1-hexanol (6.50~ molybdenum content) catalyst.
The concentrated molybdenum catalyst was prepared by mixing 299.5g 2-ethyl-1-hexanol with 29.0g MoO3 and to this mixture was added 20 ml of concentrated ammonium hy-droxide. The catalyst preparation reaction mixture was ~27B3'~

heated to 180~ and held there for five hours removing some 21 ml of water. The reaction mixture was cooled and filter-ed. Atomic absorption analysis indicated the molybdenum content of the filtrate was 6.50% (97.7~ molybdenum incorpo-rated into soluble catalyst form).
In this reaction the propylene to TBHP mole ratio was only 1.38:1 and the mole ratio of TB~P/T~A was 1.27 1.
The amount of molybdenum catalyst used was 0.0449 wt~%

molybdenum basis total reactor charge. The autoclave and contents were heated to llO~C with stirring for 120 minutes (2.0 hours). The reaction mixture was cooled and pressured out into two sample bombs.

The ~otal product weight was 173.5g The total weight of liquid product was 142.2g The liquid product was examined and found to con-tain 1.70% TBH~.

grams TBHP remaining = 2.4174g moles TBHP remaining = 0.026g moles TBHP reacted = moles fed moles remaining moles TBHP reacted = 0.8349 - 0.0269 moles TBHP reacted - 0.8080 Conversion = 0.8080 - 96.78%
0.83~9 The liquid product was analyzed by atomic absorp-tion spectroscopy and found to contain 527 ppm molybdenum, which is essentially a 96.1% recovery of molybdenum.

9rAMPL-5 - ~

Examples 5 and 6 were conducted similarly to Ex-amples 2-4, except that different molybdenum 2-ethyl hexanol catalysts were utilized. The results, however, are essen-tially similar, very high recoveries of soluble molybdenum at the end of the reaction (96.0~ in both runs). These -2~-3~b~

results are su~marized in brief tahular form below. The runs were conducted at 110~C for 1.5 hours.

~E 3 Prcpylene/ ppm %
Wt~ m~ly ~ mole Po PO PO moly in ly EX. c~d ratio Wt.% Yield Sel. ~. Liquid RecoverY

4 0.0402 1.79:1 2~.71 g5~65 98.25 97.35 506 96.0 0.0393 1.82:1 25.28 97.57 99.86 97~71 498 96.0 When research into this area was first begun, it was discovered that results improved dramatically when dry T~HP (le~s than 0.4 wt.~ ~29) was used instead of the com-mercially available TBHP. Therefore, it is preferred that the TB~P/TBA solutions contain very little water, 0.5 wt.%
or less. In our initial propylene epoxidation experiments, high olefin/hydroperoxide mole ratios were chosen because of the repeated mention in patents and the literature that selectivities are lower when propylene/TBHP mole ratios are low. Over 130 epoxidation runs were conducted at propylene/
TBHP molec ratios of 6:1-10:1. At these ratios various catalysts and changes in conditions seemed to have no large effec~ on epoxidation result~O Further, the molybdenum recoveries at 6-10:1 ratios were in the 60-80% range.
However, when low initial propylene/TBHP mole ratios were used in an attempt to find a method which would help differentiate between the many molybdenum catalysts being synthesized, it was surprisingly discovered that the propylene oxide selectivities were excellent, provided that reaction temperatures, residence times, and molybdenum catalyst concentrations were adjusted properly.

It has been further discovered that enhancement of the results is achieved when the epoxidation reaction is conduc~ed at comparatively low reaction temperature using comparatively high concentrations of molybdenum catalyst.
In addition, it has been also surprisingly discovered that a high proportion of the molybdenum charged emerged as soluble molybdenum and this proportion increased uon reduction of the propylene/TBHP cnarge ratio. Further, it was found that low ratios of propylene to T~HP in the charge lead to low by-product propylene dimer make.
Table 4 gives data about the concentration of propylene dimer in reactor effluents. Propylene dimer was determined in reactor effluents using a GC mass spectrome~er.
Propylene dimer, as notedl is an objectionable by-product because it co-distills with propylene oxide and is best separated from propylene oxide by a costly extractive dis-tillation. The cost of the extractive distillation towers as well as the utilities cost to operate such a purification unit is very high. The examples of ~able 5 where the con-ventional propylene oxide process conditions are used revealsthat the propylene dimer levels seen in Table ~ are surpris-ingly low.
Table 5 presents examples where results are im-proved even further with high TBHP concentrations together with low propylene/TBHP mole ratios, reaction staging and lower catalyst levels. These examples represent the prefer-red reaction conditions and at 1.1-1.2:1 propylene/TB~P mole ratios. The examples of Table 6 show that the molybdenum catalyst may be recycled with good results.

Examples 21 through 25 of Table 7 show the results obtained when the inventive process was scaled up 3.3 fold to a lO00 ml reactor. The only procedural difference in these examples was that after the propylene was charged to the reactor, it was heated up to, at or near the reaction temperature before the TB~P/TBA/catalyst solution was added.
Exotherm was allowed to carry the reaction to the desired temperature. Even at these quantities, excellent results are maintaLned.
Ta~le 5A gives typical catalyst prepara~ions in-volving 2-ethyl-l-hexanol and ammonium heptamolybdate.
Table 8 presents examples which demonstrate that the reaction can proceed successfully with a two-part re-actor scheme. A CSTR is followed by a PFR at a slightly higher temperature. Low propylene/TBHP ratios are again demonstrated. Catalyst recoveries are sometimes reported as slightly greater than 100%. The excess should be taken as experimental errox, and ~he recovery taXen as essentially quantitative.
The inventive process provides high concentrations of propylene oxide (24-32%) and utilizes much less propylene due to the lower propylene~TBHP mole ratio and polar reac-tion media. In this process, 4 to 16% of the propylene is unreacted. Our selectivities (moles of propylene oxide formed per mole of TBHP consumed) do not drop as we lower the propylene/TBHP mole ratio. ~urprisingly, increased selectivities to propylene oxide basis TBHP are observed.
This is because the media is more polar.
It is surprising that selectivities to the alkyl-ene oxide are at least 96~, concentrations of the alkylene ~32-~7~3~6 oxide in the product stream can be at least 24~, yields to the alkylene oxide are at least 94% and hydroperoxide con~
versions are at least 96%, all simultaneously, using the method of this invention. Further, it is surprising that molybdenum recoveries at the lower ratios of propylene/TBHP
generally are >90%.

~ 1-I
'~

~ I ~ ~

N N N N

N N N

. ~ I ~ ~

~ ~ o o `~
o O
~ S ~ tr; ~ ~ ~

. ~ ~7 co ~ I
er u~ ~ In ~r I` co U c~
a~
~ Q~ ~ ~ ~ ~r ~ u~
u~ ~ r~ o Q) ~ ~ ~ ~ ~ O O
~ ~ -~ ~
`
U~
,~ o~ c~ o , co ~
_ ~ n) ~ ~ o o c~
~, U3 llt ~1 ~ N _I t~
. ~ ~
~ IQ) dP ~
1~ ~ ~1`
~3 ~-~ ~. ~ ~r ~ ~r ~ eP ,, ~ ~ ~,~ o ~
.~ ~ I a~-ol o o c~
a~
L~
_ o o o o o o ~ r~
' 8 ~ U ~o o u~ n o u~
~o CO O
o ~ o dP In ~ O O O O ~ O ~ ~
o o o O C ~n 10 a~

~7~33 o Q~
,.~

=o 3~

a ~ .s .
~o I 0 . ~ o o ~ ~
~ ~ O O 0 ~
~ P~ , ~

~27~3ED~;

o U~ o N ~r O

C~
R _I ~P I~ ~ r-a~

~o ~ ' o ~
P~ ~3 o, u~ er ~ ~ ~ ~ 0 _, ' r ~ r~
.~

~ O U~
~ 0 o a~
_I N N

o o o Q) ~0-~ U~

O ~ ~ ~ O
x ~ ~ ~ u~
~ ~ ~ c`, 0 o ~ 0 O~ I ~ O I
n~ ~ Q~
~ ~ oo a~ ,~ o o ~
_I ~ a)-,, a~o~ o o o ,~ ~
~ l ~
~D 8 ~ ~. ~
.
_ ~) ~ 0 5~ t) ~ ~D
~ 0 E~ ~ q ~ P~ ~ ~
~ I ~
~ ~ ` ~ a0) ~
~D ~ ~8~ co ~

U

~j ~E
;~ C~ , . . . . ~ ~
~ ~ ; ~ ( dP 0 ~3 ,c 13 9~ ~ u~

8 ~ o ~ o o o ~ ~OC~ ~

3 U~ y o ~ ~ aJ a~
C~ ~ O ~ ~ g dP 0 ~ ~ O O O O O ~
d~ ' o o o o o ~ ~ 1 0 a) ~ ~ ~ o ~ ~

. ~r o cr~
el 2 i~i ~ ~D ~ r o~ o~
~ ~ ~ o~
u~
o In CO
r C~

~ I ~ ~
~3 ~ ~ o o c~ In ~

æ

~ I ~ ~
~9 ~ 1` CO 1 ~ ~ ~ l ~ 8 ~ M oP oP r~
o u~ o ~ U ~ ~ ~

a~ *
o~ 0 r` r o ~ I o o ~L ~ N
o o o o o 3 ~ Q

,, f~ 1~ N ~
, 4 ~rZ7~

~ D 1` 0 ~ ~ O O~ ~D 0 ~ ~ c~ r OD c~

. ~D 1` N ~ D C~l In ~ r~

o O
~D ~ O r~ o 1` ~ 1`

3 a~ ~
u~
G~
~ ~ ~ O ~D er O

o~ ~ ~ ~ ~ u~ ~
P ~ O ~

~ ~ ;~ o ~ ~ ~1 ~ ~ ~J ~ O O O ~
K ~ a C) 1 ~ ~ ~ 1 0 ~ ~ ~ ~1 ~ ~ ~ o ~
o o o o o o ~ ~J c:~ o o ~ o I
o ~_ ~ydroperoxid~ Choice Representative runs were made to demonstrate that t-amyl hydroperoxide may also be used in the practice of the present invention.

EX~MPLE 32 Cata~ t Preparation To a one-liter round-bottomed Morton flask fitted with a mechanical stirrer, a nitrogen inlet, a thermometer, a Dean Stark trap, a condenser and a nitrogen bubbler, were added 100 g. of either ammonium heptamolybdate (Catalyst A) or ammonium dimolybda~e (Catalyst B) and 300 g of ethylene glycol. Th~ re2ction mixture was heated to 85-110C for about l hour with nitrogen slowly passing through the flask.
At the end of that time, the reaction was essentially com-plete and essentially all of the ammonium molybdate was dissolved. The reaction mixture was subjected to an aspira-tor vacuum at a temperature of about 85-95C for about 1.5 hours and then reheated to 90-100C for an additional hour.
On cooling, there was obtained a clear liquid catalyst composition containing (Catalyst A) 16.1% molybdenum by Atomic Absorption spectroscopy~ 1.17~ nitrogen (Kjeldahl) and 1.67% water ~Karl Fisher analysis). The acid number of the catalyst (mg KOH per gram of sample) was found to be 8Q.94 and 167.85 in duplicate analyses and (Catalyst B) 13.~4 molybdenum by ~tomic Absorption spectroscopy.

Epoxidation ~uns The epoxidation runs summarized in Table 9 where made in a 300 ml. stainless steel autoclave. The propylene 3~i feed component was charged at ambient temperature and then the t-amyl hydroperoxide (TA~P) feed component was charged premixed with 0~38 grams of the catalyst. This provided for a catalyst concentration of about 350 ppm of catalyst in the reaction medium. The TA~P feed component consisted of about a 70 wt.~ solution of T~HP in t-amyl alcohol (TAA) which contained about 0.2 wt.% of water. The reaction conditions employed and the results obtained are summarized in Table 9.

1~'7B3~

~1 ~ ~ ~ r~
~ ~ ~:r o~

o _, ~ ~
dP I~
_I ~ !!; dP
$ ~^ u~e 5 ~ ~ ~ ,, --~ 0-rl 0 ~ W _I ~ ~ tS
q ~ ~ ~ O r~
,~ ,, P~ _I ~ r~

~ o ~ ~ ~
~ ~ ~ 0 J _~ ~; ~ u~ i c~ L~ ~ 2 UO~
u~ ~ o o ~a ~
O O

oo ~vl o o ~ gJO ~ ~ ~ ~

~1o o o m From Table 9 it will be seen that in both instances there was excellent selectivity for the propylene oxide and an excellane conversion of the t-amyl hydropero~ide.
In an attempt to broaden the scope of the instant invention, VariQUS experiments were run using a hydroper-oxide other than TBHP or TAHP -- in ~his case cumene hydro-peroxide (CHP)~ I~ was surprisingly discovered tha~ while TBHP and C~P behave similarly at high propylene to hydro-peroxide mole ratios, at the low ratios of thi~ invention, below about 2:1, their behaviors divarged. The propylene oxide yields using CHP remained low when the propylene to cumene hydroperoxide mole ratios were low in contrast ~o the propylene oxide yields obtained using TB~P when high cata-lyst concentrations were used in both instances. When CHP
was used in the low ratio examples propylene oxide selec-tivities, basis CHP reacted, did increase when catalyst concentration increased, but not nearly to the extent that they did when TBHP was used in similar examples. Thus, it was also discovered that the choice of the hydroperoxide is also crucial in obtaining good resul~s, particularly wifh respect to the method of this invention. This discovery will be explored in detail with respect to the following examples. These examples were conducted according to the procedures outlined earlier using the parameters noted in the tables.
Tables 10 and 11 show the overriding inf luence of mole ratio of propylene to CHP. When CHP is used, the selectivity to propylene oxide, basis CHP reacted, decreases rapidly with decreasing charge ratios of propylene to C~P.
Examples 32 36 demonstrate that with d~creasing propylene/C~P

~44-33~

charge ratio, propylene oxide selectivity decreases while dimer content (on a ~pure PO" basis, i.e. propylene distilled out) generally increases, at low catalyst levels (69-78 ppm). Examples 37 and 38 demonstrate a similar trend for medium catalyst levels tabout 250 ppm), while Examples 39-45 reveal that these undesirable result6 hold true even for catalyst concentrations that are high (400 449 ppm).
With CHP, the propylene dimer make does not de-crease with lower propylene/C~P charge ratio as it does with lower propylene/TBHP charge ratios. Actually, the dimer make tends to increase with decreasing initial propylene/C~P
ratios. Further, note that Table 10 shows these trends for two different CHP concentrations of CHP in cumyl alcohol, 30 and 59%, and that the lower C~P amount (30%) actually gives better results to PO selectivity.
Table 12 recasts some previous examples in a form which demonstrates that while keeping the propylene/CHP mole charge ratio constant (a~out 3:1 for Examples 34, 37 and 39, and about 1.3:1 for Examples 35 and 41) and increasing the catalyst levels from the 60-80 ppm molybdenum catalyst range to ~he 250 to 450 ppm catalys~ range, the propylene oxide selectivity, basis CHP reacted, CHP conversion and propylene oxide yields all increase. Even the propylene dimer make decreases. However, also note that for Examples 35 and 41 at low propylene/CHP ratios, propylene oxide selectivities, basis CHP reacted, are still low. Thus, it appears that when CHP is used, high reactant ratios should be maintained, in contrast to the case where TBHP is used and excellent propylene oxide selectivities, basis TBHP reacted, are achievable with low propylene/TBHP mole ratios.

7B3~

Table 13 presents four previous examples plus two new ones t50 and 51) to demonstrate that at a given catalyst level (250 or 400-450 ppm) that propylene oxide selectivity, basis CHP rPacted, decreases and propylene dimer make in-creases when the C~P concentration increases from 30 to 59%.Propylene oxide yields also decrease.
~ able 14 presen~s the previous examples recast into yet another form to demonstrate that at propylene~CHP
charge ratios of 1.3:1 to 1.4~ is impo~sible ~o achieve high PO selec~ivi~ies or yields by varying either the CHP
concentration (30.0 to 43 to 59~) or the catalyst concentra-tion (60 to 250 to 450 ppm basis to~al charge.) In fact, at reactant mole ratios of 1.301 to 1.4:1, it appears that the optimum catalyst level may be the 200 to 350 ppm range.
Notice that the propylene dimer make increases with de-creasing PO selectivity and decreases with increasing PO
selec~ivity. This latter relationship is not seen with TBHP.
The CHP results seen herein are to some extent 20 confirmed by Kollar in U. S. Patent 3,351,635 (see Table 10) where it is seen that CHP conversion and epoxide selectivity decrease with decreasing mole ratio. The in~en~ion herein of using high catalyst concentrations was not discovered therein, even to the limited extent possible with CHP, as opposed to thè dramatic improvement possible with TBHP.
Finally, an Example 55 was conducted substantially the same as Example 39 except that it was performed in a single one-hour, 90C step as opposed to ~he staged reaction of Example 3g (one hour at 90C followed by one hour at 30 110C). Other minor differences were a charge ratio of 7.18:1 33~6 instead of 7.03:1 and a catalyst concentration of 79 ppm in-stead of 78. As with TBHP, the CHP conversion was much lower (66% as compared with Example 39's 97.7%).

cP ~ ~3 ~ ~ ~ ~ ~D O ~ o u~ ~
~ h ~ ~ o ~ er o co tl p~ ~ Ll O 0~ ~` co ~ ~ C~ ~ er Q~ ~r ~ ~
~ ..
~
t~-r1 C~ _I ~) U7 ~ N1~ U'~
~ Q a~ P C~ o ,, ,~ o E~ , 13 ~3 ~_1 ~ S~ 3 ~D r~ o r- o ~ ~
U~ ~ ~ ~ ~ 0~ ~ N ~) o ~ ~
_i ~ ~
~P C~ O O~ 0 /`~
I~ O cr~ a~ o c) N O ~ . ~ o o CD ~ C3 0 C r~ o~ ~
~~ ~ ~ O ~ I` r~ D 1` ~ Ul O
o~ ., ~ ~ ~ ~ O~ ~
~ C7 d~

~ ~ ~
~ . J~ ~
~ R~ ~ I` I~
U ~ ~ ~ .. , ~ ~ ~ ~ ~ ~ ~ ~ '' dP

i ~ O ~ ~ O~ ~I 11 1` ~ ~ ~
0 O O O ~ ~ ~

. , ~0 o _I b ~
.~ ~ o ~2~3~3i ~. ~ ~ ~ ~ ~
dP CO r~
il ~` ~
~ ~ dP ~ I` ~ r~ ~
~ cr. I` r ~ ~

o o o _, P

U~
~ ~ P
U~

æ u~ ~ ~n C5 Pi ,~ ~ ~ O ~
o ~1 .
' U~

CO a~
~ ~ 3 g ~ t.~ h ~ ~ . ~
~q QJ ~ IU

~ æ~ ~ O ~ ç

O ~

~ ~ q ~1 0 0 ~ ~
,~ ~ ~ a~ ~ I

E~~ o ~ I` O
~~ ~ r~ o o ~~ Q U~ ~
~ CJ C,~ ~ ~

~ ~æ~ ~
~ n~ d . . ~ . ~' Q
~ u~
~ ~1$
U~ ~1 O O
~ ~0 _I ~ ~

2~U~ ~
t~ ~ .
C~
er~
Li ~ o~ ~ o ~ t) dP ~ O~ O O a~ ~ o I_i ~ U~ ~ r i O 0 ,~

~'~ ~o CD
u ~
~ ~ ~ 0 ~i r ~ 8~ ~~
r~
~LI ~ U 1~) ~ t~ t~i~r ~Lt~
c~ ~ O, a) ~ 8 a~

~ ~) dP Ct~ N 1'-- --I
.~ u~

~i @ ~ ` N N

I ~ o o ~_1 In r- I,q 'U~ ~ ~ i O u7 0 h 3~ ~
~1 ~ ~ o ~ ~ C O
~ ~ 0 1` a~ o ~ ~

o~ o ~ ~

o ~ c~
~1 ~ r- r ~r N r~

~1 ~

~1 N
D
5~
n ~ ~''' I`
~ .
o~ ~~~

o ~ co o o O O N 1~

'Ul ~ UO~

~2~3~

Olefin_Choice Alth~ugh propylene has been used in the prior examples of the present invention as a matter of convenience and to provide for comparative data, other C3-C20 olefins may also be used in the practice of the present invention.
This is illustrated by the following specific examples.
When the higher olefins such as C4 to C20 are epoxidized, it is important to obtain an essentially qUantitiYe conversion of the olefin to the epoxide because the feed stock nd epoxide reac~ion product have similar physical properties and are separated only with great difficulty.

To a l-liter round bottomed Morton (fluted flask equipped with a mechanical stirrer, Dean Stark ~rap, ther-mometer, N2 inlet and bubbler, was added 35.31g of ammonium heptamolybdate tetrahydrate from Climax Molybdenum Co.
(molecular weight - 1235.86, g atoms moly - 0.2000, (NH4~6Mo7O24.4H2O) followed by 182.32g of 2 ethyl-l-hexanol 20 (99~ purity, alfa, molecular weight = 130.2, moles = 1.4) and 1404 ml H2O. Note the mole ratio ~f alcohol (2-ethyl~l-hexanol) to g atoms molybdenum = 7.0/l and the mole ratio of added H2O/g atoms molybdenum = 4.0/l. The reaction mixture was heated slowly to 178C and held at 178-180C for five hours during which time 29 ml H2O were xemoved with the Dean Stark trap. The cooled reaction mixture was filtered through glass filter paper to remove solids. The filtrate w~ight 176.3g.

~2~7~3~6 ~ molybdenum in filtrate by AA = 10.1%
% N in filtrate by Rjeldahl = 0.34%
g molybdenum fed = 19.187 g moly "out" in (soluble) filtrate = 17.81 % molybdenum lncorporated in catalyst = 92.80%

To a 250 ml round-bottomed flask ~itted with mag-ne~ic stirring bar, thermometer, condenser, N2 inlet and bubbler was added 42.0g of octene-l (molecular weight 112, 0.375 moles) followed by 35.5g of 72008% TB~P with 0.39g of 10 molybdenum catalyst 5810-60 (10.1% molybdenum) premixed with the ~BHP/TBA. The reaction mixture was heated slowly to 95C (exothermed to 99C) and then held there (93-96C~ for 2.0 hours. Af~er cooling, the reac~ion mix~ure was solids fr~e and weighed 74.1gn 15 wt.% TBHP = 1.70%
wt.% octene oxide = 46.182 wt.~ octene = 12.677%
g octene oxide = 34.221 moles epoxide = 0.26735 Selectivity = 0.26785 C8 epoxide 1.2703 = 98.91 Yield = 0.26735 20 C8 epoxide 0.2843 = 94.03 g TB~P remaining G 1 . 2597 moles TBHP remaining = 0.0140 moles TB~P fed = 0.2843 moles TB~P reacted = .2703 Conver~ion = 2703 TBHP .2843 - 95.08%

Several other examples are given in the table attached. The procedures and apparatus were exactly like that in Example 34.

~27~33~6 o o o o o o o o d) ~ ~( ~ t` ~ o o ~ o tn ~ ~8 ~ r~ D O ~ O~

, o r~
~ ~ a~ C5~ r u~ ~ ~ Lr ~r dP ~ ~ O~
I` m ~ a ~ o~ o r~
~ .

o ~ o ~o U- o In ~ O
~ ~ 1 u~ OP a, u~
~ ,~ ~$~ ~ ~ ~ ~ ~ o ~ ~ ~ ~ It~ o Ln i~i ~ ~ 3 0 ~ ~ ~

i~3 ~ a) ~ ~ o !~ ~ ~
~ - .
* In L~ ~ * Ln Q-U~ '~ i o o~

~ ~ l o o o o o o C:~ o o~ -~ Q-t) Il~ * o 11~ 0 U~

~ ~ O ~ o ~ I I I I o o o o z 1~1 *

*

3~

c, ~ a ~ a o ~ o c: a a _ O ~ ~ r~
0 ~ o~
o-~a ~ ~ ~ _ _ _. _ _ :~ co cn a~ co CO
u~ ~r o u7 d~ ~ ~ I` ~D a~ r~
I~ . ~ c~ ~ c~

t~ u~ 8 ~ ~ r~ o ~i ~ U 5; ,~
~ ~ 5~ a~

o ~ ,o~ ~ O u~ o ~ ~ o ~1 ~ ~ _ o I
. 3 ~ ~ ~ o _ ~ .. ~ Q~.~ ~
~ ~ ~ ~01 ~
~ b ~8~
~ ' ~ ~ O ~
~ ~ .
~Lr~ m ~ ~ ~1 o o ~ i o ,~1 oo o oo co 1~ ~.o O
~

o ~ o ~i I I I I I I ~

~ ~1 ~L2~7æ30~i Many modifications could b~ made by one ski 1 led inthe art in the invention without changing its spirit or scope which are defined only by the appended claims.

Claims (19)

1. In a method wherein a hydroperoxide charge stock selected from the group consisting of t-butyl hydro-peroxide and t-amyl hydroperoxide is reacted in a reaction zone in liquid phase with a C3 to C20 olefin charge stock in the presence of a catalytically effective amount of a solu-ble molybdenum catalyst to form a product olefin epoxide corresponding to the olefin charge stock and a product alcohol selected from the group consisting of t-butyl alco-hol and t-amyl alcohol corresponding to the hydroperoxide charge stock, the improvement which comprises:
maintaining a reaction medium composed of more than 60 wt.% of polar components in a reaction medium in said reaction zone composed of said olefin charge, said corresponding product olefin epoxide, said hydroperoxide charge, said corresponding product alcohol and said cata-lyst, by feeding to said reaction zone at least a 30 wt.%
solution of said hydroperoxide charge in solution in said corresponding product alcohol, and feeding said olefin charge stock to said reaction zone in an amount relative to the hydroperoxide charge in said solution such that the mole ratio of said olefin charge stock to said hydroperoxide charge stock is within the range of about 0.5 to about 2 moles of olefin charge per mole of hydroperoxide charge, said polar components of said reaction medium comprising said hydroperoxide charge, said corresponding alcohol and said corresponding product olefin epoxide.

2. A method in claim 1 wherein the peroxide is t-butyl hydroperoxide and the corresponding product alcohol is t-butyl alcohol.

3. A method as in claim 2 wherein the olefin charge is octene and the corresponding olefin epoxide is octene epoxide.

4. A method as in claim 2 wherein the olefin charge is dodecene and the corresponding olefin epoxide is dodecene epoxide.

5. A method as in claim 1 wherein the peroxide charge is composed of t-amyl peroxide and the corresponding product alcohol is t-amyl alcohol.

6. In a method wherein a hydroperoxide charge stock selected from the group consisting of t-butyl hydro-peroxide and t-amyl hydroperoxide is reacted in a reaction zone in liquid phase with a propylene charge stock in the presence of a catalytically effective amount of a soluble molybdenum catalyst to form propylene oxide and a product alcohol selected from the group consisting of t-butyl alco-hol and t-amyl alcohol corresponding to the hydroperoxide charge stock, the improvement which comprises:
maintaining a reaction medium composed of more than 60 wt.% of polar components in a reaction medium in said reaction zone composed of said propylene, said propylene oxide, said peroxide charge, said corresponding product alcohol and said catalyst, by feeding to said reac-tion zone at least a 30 wt.% solution of said hydroperoxide charge in said corresponding product alcohol, and feeding said propylene charge stock to said reaction zone in an amount relative to the hydroperoxide charge in said solution such that the mole ratio of said propylene charge stock to the said hydroperoxide charge stock is within the range of about 0.5 to about 2 moles of propylene per mole of peroxide charge, said polar components of said reaction medium comprising said hydroperoxide charge, said corresponding alcohol and said propylene oxide.

7. A method as in claim 6 wherein the peroxide charge is composed of t-amyl peroxide and the corresponding product alcohol is t-amyl alcohol.

8. In a method wherein a t-butyl hydroperoxide charge stock is reacted in a reaction zone in liquid phase with a propylene charge stock in the presence of a cataly-tically effective amount of a soluble molybdenum catalyst to form propylene oxide and t-butyl alcohol, the improvement which comprises:
maintaining a reaction medium composed of more than 60 wt.% of polar components in a reaction medium in said reaction zone composed of said propylene charge stock, said propylene oxide, said t-butyl peroxide, said t-butyl alcohol and said catalyst, by feeding to said reac-tion zone a solution of said t-butyl hyproperoxide in said t-butyl alcohol containing at least about 30 wt.% of said t-butyl peroxide, and feeding said propylene charge stock to said reaction zone in an amount relative to the charge of t-butyl hydroperoxide in said solution such that the mole ratio of said propylene to said t-butyl hydroperoxide charge stock in said charged solution is within the range of about 0.5 to about 2 moles of propylene per mole of t-butyl hydroperoxide, said polar components of said reaction medium comprising said t-butyl hydroperoxide, said t-tertiary butyl alcohol and said propylene oxide.

9. A method as in claim 8 wherein the water content in said solution of t-butyl hydroperoxide in t-butyl alcohol is less than 1 wt.%.

10. A method as in claim 9 wherein the water content is less than 0.5 wt.%.

11. A method as in claim 9 wherein the solubil-ized molybdenum catalyst concentration in said reaction medium is within the range of about 50 to about 1,000 ppm.

12. A method as in claim 11 wherein the catalyst concentration is within the range of about 200 to about 600 ppm.

13. A method as in claim 11 wherein the catalyst concentration is within the range of about 250 to 500 ppm.

14. In a continuous method wherein a t-butyl hydroperoxide charge stock is continuously reacted in a reaction zone in liquid phase with agitation with a propylene charge stock in the presence of a catalytically effective amount of a soluble molybdenum catalyst to form propylene oxide and t-butyl alcohol the improvement which comprises:
conducting said reaction in said reaction zone at a temperature within the range of about 100° to about 130°C in a reaction medium containing less than about 1 wt.% of water and about 200 to about 600 ppm of catalyst and composed of said propylene, said propylene oxide, said t-butyl peroxide, said t-butyl alcohol and said catalyst, maintaining a reaction medium in said reac-tion zone composed of more than 60 wt.% of polar components by continuously charging to said reaction zone a solution of said t-butyl hydroperoxide in said t-butyl alcohol con-taining at least about 30 wt.% of said t-butyl peroxide, continuously charging said propylene in an amount relative to the charge of t-butyl hydroperoxide in said solution such that the mole ratio of said propylene charge stock to said t-butyl hydroperoxide charge stock in said charged solution is within the range of about 0.5 to about 2 moles of propylene per mole of t-butyl hydroperoxide, and continuously removing a product stream from said reaction zone and recovering propylene oxide and t-butyl alcohol from said product stream, said polar components of said reaction medium comprising said t-butyl hydroperoxide, said t-tertiary butyl alcohol and said propylene oxide.

15. A continuous method as in claim 14 wherein said charge stocks are initially reacted with each other in a first continuous stirred tank reaction zone to give an intermediate reaction mixture, and wherein a stream of said intermediate reaction mixture is continuously withdrawn from said continuously stirred tank reaction zone and charged to a second plug flow reaction zone and wherein the said reac-tion is completed in said plug flow reaction zone.

16. A method as in claim 15 wherein the continu-ously stirred tank reactor is operated at a temperature within the range of about 70° to about 115°C and the second reactor is operated at a temperature within the range of about 115° to about 150°C.

17. A method as in claim 16 wherein the continu-ously stirred tank reactor is operated at a temperature of 90° to 115°C and the second reactor is operated at a tem-perature within the range of about 120° to 140°C.

18. A method as in claim 17 wherein the mole ratio of the propylene charge stock to the t-butyl hydroperoxide charge stock is within the range of about 0.9 to about 1.8 moles of propylene per mole of t-butyl hydroperoxide.

19. A method as in claim 18 wherein the mole ratio is within the range of about 1.05 to about 1.35 moles of propylene per mole of t-butyl hydroperoxide.