GB1602508A - Process for producing polyhydric alcohols - Google Patents

Description

(54) PROCESS FOR PRODUCING POLYHYDRIC ALCOHOLS (71) We, UNION CARBIDE CORPORATION, a corporation organized and existing under the laws of the State of New York, United States of America, whose registered office is, 270 Park Avenue, New York, State of New York 10017, United States of America. (Assignee of LEONARD KAPLAN), do hereby declare the invention, for which we pray that a patent may granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: This invention relates to the production of polyhydric alcohols and derivatives thereof, such as their ether and ester derivatives and oligomers of such alcohols.

This invention also produces monohydric alcohols such as methanol and their ether and ester derivatives.

Polyhydric alcohols are presently being produced synthetically by the oxidation of petroleum derived materials. Owing to the limited availability of petroleum sources, the cost of these petroleum derived materials has been steadily increasing. Many have raised the dire prediction of a significant oil shortage in the future. The consequence of this has been the recognition of the need for a new low cost source of chemicals which can be converted into such polyhydric alcohols.

This invention is directed to the process of making alkane polyols, for example alkane diols and triols, containing 2, 3 or 4 carbon atoms, and derivatives such as their esters. Key products of the process of this invention are ethylene glycol and its ester derivatives. Byproducts of this invention are the lesser valuable, but valuable neverthless, monohydric alkanols such as methanol, ethanol and propanols, and their ether and ester derivatives. The products of the process of this invention contain carbon, hydrogen and oxygen There are described in U.S. Patent 3,833,634, issued September 3, 1974, and U.S. Patent 3,957,857, issued May 18, 1976, processes for reacting hydrogen and oxides of carbon in the presence of rhodium carbonyl complex catalysts. U.S.

Patent 3,957,857 is concerned with a rhodium carbonyl complex which is a rhodium carbonyl cluster exhibiting a particular infrared spectrum. The conditions, broadly speaking, employed in those processes involve reacting a mixture of an oxide of carbon and hydrogen with a catalytic amount of rhodium in complex combination with carbon monoxide, at a temperature of between about 100"C to about 3750C and a pressure of between about 500 p.s.i.a. to about 50,000 p.s.i.a. As described in these patents, the process is carried out in a homogeneous liquid phase mixture in the presence of one or more compounds selected from among groups referred to in the patent, as organic oxygen ligands, organic nitrogen ligands and organic aza-oxa ligands. In addition to the aforementioned U.S. Patents, the following U.S. Patents and U.K. Patents amplify the development of the processes for making alkane polyols from mixtures of hydrogen and oxides of carbon: U.S.P. 3,878,292 Patented April 15, 1975 U.S.P. 3,878,290 Patented April 15, 1975 U.S.P. 3,878,214 Patented April 15, 1975 U.S.P. 3,886,364 Patented May 27, 1975 U.S.P. 3,940,432 Patented February 24, 1976 U.S.P. 3,929,969 Patented December 30, 1975 U.S.P. 3,952,039 Patented April 20, 1976 U.S.P. 3,948,965 Patented April 6, 1976 U.S.P. 3,944,588 Patented March 16, 1976 U.S.P. 3,974,259 Patented October 10, 1976 U.S.P. 3,989,799 Patented November 2, 1976 U.S.P. 4,013,700 Patented March 22, 1977 U.K.P. 1,522,491 Patented 11 July, 1975 U.S.P. 3,968,136 Patented July 6, 1976 U.K.P. 1,521,695 - U.S.P. 4,001,289 Patented January 4, 1977 U.K.P. 1,537,850 Patented December 31, 1975 U.K.P. 1,565,979 Patented September 29, 1976 U.K.P. 1,565,978 Patented September 29, 1976 U.K.P. 1,563,232 Patented September 29, 1976 U.S.P. 4,162,261 Patented July 24, 1979 U.S. Patent No. 3,852,039 issued April 10, 1976 to Walker et al describes the use of salts containing alkali metal cations to improve the yield of the desired alkane diols and triols of the invention. The process of the Walker et al patent involves providing a metal salt to the aforementioned homogeneous liquid phase reaction mixture to promote the production of alkane polyols, ethylene glycol being the primary product in terms of its commercial value. The salt promoter provided to the mixture is present in an amount to achieve the optimum rate of formation of said alkane polyol at the correlated catalyst concentration, temperature and pressure of such reaction mixture. A range of 0.5-1.5 atoms of cation per six atoms of rhodium is disclosed in the patent. When the amount of alkali metal cation in the reaction is greater or less than this amount, the productivity of reaction to polyhydric alcohol is significantly reduced. The present invention, however, provides for the selection of a salt promoter in terms of its basicity to minimize inhibition of alkane polyol production by the presence of an excess of salt.

The following postulate possible mechanisms which would result in the abovedescribed behaviour: a.) the inhibitor function of the salt is of higher kinetic order in salt than is the promoter function; b.) the promoter function of the salt has a stoichiometric limit after which only the inhibitor function of the salt remains.

The term "inhibitor function" means the function of the salt which results in a decrease in alkane polyol yield as salt concentration increases.

An above postulate can be illustrated by the following reaction scheme: The involvement of salt is described as follows (Rh symbolizes a rhodiumcontaining species):

Rh+salt (MX) < [Rh-M+ K Rh-+M+ t glycol INote: In the above reaction scheme the charge of the rhodium carbonyl complex is not shown; it contains a fixed or varying number of CO's and H's; the rate and equilibrium constants implicity contain any appropriate CO and H2 concentrations.] The salt acts as a promoter because its anion helps to produce the active catalyst and as an inhibitor because its cation has an adverse mass law effect on the equilibrium concentration of a direct precursor of the active catalyst. The model suggests the use of a salt and reaction conditions which produce a reactive anion and an impotent cation.

The model predicts that the rate will increase as a function of the stoichiometric concentration of salt promoter, and then decrease. As K increases, either becaue of an intrinsic property of M+ or use of a solvent of high dielectric constant (for example, sulfolane) or high constant of complexation with M+, (for example, crown ether, and to a lesser extent tetraglyme), the rate of decrease decreases. Any complexation of the cation by the solvent is incorporated implicity into the definition of K as a result of the customary definition of standard states.

More generally, however, such a microscopic solvent effect is just one example of the use of a complexing agent to influence the ion-pairing ability of M+. The simplest case, since it would not involve the introduction of an additional compound, would be complexation of M+ by X. . This is described by the following equilibra Rh- M+uRh-+M+ (1) X- M+=X-+MA (2) X-+ROH > XH+RO- (3) RO- M+RO-+M+ (4) Thus, the process of U.S. Patent No. 3,952, 039 recognizes that there is an optimum concentration for salt promoters to achieve maximum alkane polyol production and that amounts in excess of that optimum concentration are undesirable. The present invention contemplates increasing the concentration of the salt in excess of said optimum concentration for the purpose of enhancing catalyst stability in the reaction. Catalyst stability relates to the desirable feature of keeping the catalyst in solution. The invention also recognizes the fact that allowing for some excess of the salt over the optimum concentration will reduce the criticality of having to operate the process under strict control of salt concentration.

The process of this invention differs from the process described in U.S. Patent No. 3,952,039, in that there is provided in the aforementioned homogeneous liquid phase mixture a concentration of salt promoter in excess of the optimum such that the rate of formation of alkane polyol is not decreased from the maximum by more than 50, 4.

For the purposes of this invention, the ultimate salt promoter selected is one whose cation ion pairs least with the rhodium catalyst. One should employ such a promoter in the homogeneous liquid phase reaction mixture in an amount which is greater than the minimum for producing the optimum rate of formation of alkane polyols, particularly ethylene glycol, using that promoter.

The effects of concentration of the salt promoter on product formation in the homogeneous liquid phase mixture of the process of this invention has been found to be dependent upon the temperature, the rhodium concentration, the solvent employed and, to a lesser degree, the pressure.

The precise role of the rhodium carbonyl complexes, such as the rhodium carbonyl clusters, in the reaction of hydrogen with oxides of carbon to produce polyhydric alcohols is not fully appreciated at present. Under the reaction conditions of the present process the carbonyl complexes are believed to be anionic in their active forms.

Infrared spectra under reaction conditions of the present process have shown Rh(CO)4-, Rh,3(CO)24H3-2, Rh8(CO)15H-, Rh,3(CO)24H2 3, and [Rh12(CO)3436]2- anions, and other rhodium clusters to be present at various concentrations at different times of the reaction. These may represent the active rhodium carbon species responsible for polyhydric alcohol formation or may be merely symptomatic of some further intermediate transitory rhodium carbonyl structure which serves to convert the carbon monoxide and hydrogen to the polyhydric alcohol.

The salt promoters contemplated by the present invention include any organic or inorganic salt which does not adversely effect the production of polyhydric alcohols. Experimental work suggest that many salts are beneficial as either a copromoter and/or in aiding in maintaining rhodium in solution during the reaction. Illustrative of useful salt promoters are the ammonium salts and the salts of the metals of Group I and Group II of the Periodic Table (Handbook of Chemistry and Physics-SOth Edition) for instance the halide, hydroxide, alkoxide, phenoxide and carboxylate salts such as sodium fluoride, cesium fluoride, cesium pyridinolate, cesium formate, cesium acetate, cesium benzoate, cesium pmethylsulfonylbenzoate (CH3SO2C6H4COO)Cs, rubidium acetate, magnesium acetate, strontium acetate, ammonium formate, ammonium benzoate and the like.

Preferred are the cesium and ammonium salts.

In addition, the anion of the above salt may be any of the rhodium carbonyl anions. Suitable rhodium carbonyl anions include [Rh6(CO)ls]2-; [Rh6(CO),5Y]- wherein Y may be hydrogen or halogen, such as chlorine, bromine, or iodine [Rh6(CO)15(COOR"Iwherein R" is lower alkyl or aryl such as methyl, ethyl or phenyl; [Rh6(CO),412-; [R,(CO),6]3-; [Rht2(CO)3ol2-; Rh,3(CO)24H3-2; and Rhl3(CO)24H2-3; Rh,3(CO)24H-4.

The capabilities of seven cesium carboxylates (RCO2Cs) as inhibitors of alkane polyol formation in tetraglyme, 18-crown-6 and sulfolane are shown in Tables I and Ill and Figures 1 and 2. They decrease with increasing basicity of RCO2-, a result consistent with consideration of equilibria (5)-(8) Rh-Cs+=Rh-+Cs+ (5) RCO2-Cs+=RCO2-+Cs+ (6) RCO2-+R'OHoRCO2H+R'O- (7) R'O-Cs+=R'O-+Cs+ (8) within the framework of the previously discussed model: Since inhibitor capability [rate of fall-off of a plot of g glycols vs. mmoles salt (Figures 1 and 2), g0.65-g0.75 (Table 1)] depends on K5 K5+[Cs+] an increase in basicity of RCO2 at fixed stoichiometric [Cs+l leads to a decrease in [Cs+l and in its inhibitory effect. Note that this dependence on basicity of RCO2 is not merely a general "basicity effect" of added nucleophile (amine or anion) since variation of basicity of amine and anion affects oppositely the degree of inhibition by the amine and salt, respectively. The effect of amine basicity on inhibition of alkane polyol formation is disclosed in U.K. Patent No. 1,565,979.

TABLE 1 Inhibitory Capability of RCO2Cs vs. Basicity of RCO2-(a) g. of ethylene glycol from Kdissoon of RCO2H R 0.50 0.65 0.75 mmole RCO2Cs g065-g0,75 (H2O,250)x l05(b) H 3.06 4.26 2.46 1.8 18 Ph3C 2.00 3.70 2.08 1.6 11 Ph 2.25 4.00 2.44 1.6 6.2 Me 2.73 3.40 2.19 1.2 .17 Me2CH 2.39 3.81 3.04 0.8 1.4 Me3C 2.65 3.65 3.09 0.6 0.93 (a) 1/1 H2CO, 8000 psig, 4 hours, 3 mmoles Rh(CO)2 acac, 75 ml tetraglyme, 220 , 1.25 mmoles pyridine.

(b) For values in other solvents, see Table II below. Except for the unusually inaccurate results of Reutov, et al. with Me2CHCO2Cs, there is a monotonic relationship between acidity in H2O and in a dipolar aprotic solvent.

TABLE II K(RCO2H#RCO2-+H+)12

LON N f ~ C CE! ON ed X o z N o N o 14 as 1D2-E ev z o ON H 18x10-5 300x10-14(1) 100x10-14 600x10-13(6) 35x10-11 # 6.2 60(1,2) 50 .100(7,2c,6) 6000x10-12 22x10-23 12x10-7 7.4 CH3 1.7 5(1,2a,3) 5 3(8,7b) 1 5 5 1.1 (CH3)2CH 1.4 15(4) 3 24(4a,9) (1) S. M. Petrov and Yu. I. Umanskii, Zh, Fiz. Khim, 42, 3052 (1968).

(2) (a) I. M. Kolthoff, M. K. Chantooni, Jr., and H. Smagowski, Anal. Chem., 42, 1622 (1970); (b) J. Juillard, J. Chim. Phys. Physicochim, Biol., 67, 691 (1970); (c) T. Jasinski and K. Stefeniuk, Chem. Anal. (Warsaw), 10, 211 (1965) [Chem. Abstr., 64, 4244e (1966)].

(3) M. T z and R. Schaal, Bull. Soc. Chim. France, 1372 (1962).

(4) (a) K. P. Butin, I. P. Beletskaya, P. N. Belik, A. N. Ryabtsev, and O. A. Reutov, J. Organometal. Chem., 20, 11 (1969).

(b) Scaled so that K(CH3CO2H#CH3CO2-+H+)=5x10-14.

(5) A. P. Kreshkov, Ya. A. Gurvich, G. M. Galpern, and N. F. Kryuchkova, Zh. Anal. Khim 1166 (1972).

(6) S. M. Petrov, Yu. I. Umanskii, I. F. Mullin and A. N. Tetricknyi, Zh. Fiz. Khim., 47, 647 (1973).

(7) (a) I. M. Koltoff and M. K. Chantooni, Jr., JACS, 93, 3843 (1971); (b) J. Courtot-Coupez and M. LeD m zet, Bull. Soc. Chim. France, 1033 (1969); (c) K. Kalfus and M. Vecera, Coll. Czech. Chem. Comm., 37, 3607 (1972); (d) C. D. Ritchie and R. E. Unschold, JACS, 90, 2821 (1968).

(8) (a) I. M. Kolthoff, M. K. Chantooni, Jr., and S. Bhowmik, JACS, 90, 23 (1968); (b) M. LeD m zet, Bull. Soc. Chim. France, 4550 (1970).

(9) Scaled so that K(CH3CO2H#CH3CO2-+H+)=3x10-13.

(10) C. Madic and B. Tr millon, Bull. Soc. Chim. France, 1634 (1968).

(11) P. Reynand, Bull. Soc. Chim. France, 4597 (1967). 70 g sulfolane/100 ml solution.

(12) See, the original papers for units.

(13) R. G. Bates and Z. Pawlak, J. Soln. Chem., 5, 213 (1976). Mole fraction sulfonate=0.8.

Illustrative solvents which are generally suitable in making the homogeneous mixture include, for example, ethers such as tetrahydrofuran, tetrahydropyran, crown ethers (see, for example, "Structure and Bonding" vol. 16, 1973, Published by Springer-Verlag), diethyl ether, 1,2-dimethoxybenzene, 1,2-diethoxybenzene the mono- and dialkyl ethers of ethylene glycol, of propylene glycol, of butylene glycol, of diethylene glycol, of dipropyTene glycol, of triethylene glycol, of tetraethylene glycol, of dibutylene glycol, of oxyethylenepropylene glycol, etc; alkanols such as methanol, ethanol, propanol, isobutanol, 2-ethylhexanol, etc.; ketones such as acetone, methyl ethyl ketone, cyclohexanone, cyclopentanone, etc.; esters such as methyl acetate, ethyl acetate, propyl acetate, butyl acetate, methyl propionate, ethyl butyrate, methyl laurate, etc.; water, gamma-butyrolactone, deltavalerolactone, substituted and unsubstituted tetrahydrothiophene-l, I-dioxides (sulfolanes) and others. The mono and dialkyl ethers of tetraethylene glycol, gamma-butyrolactone, particularly sulfolane, 3,4-bis(2-methoxyethoxy) sulfolane, tetraglyme (dimethyl ether of tetraethylene glycol), and crown ethers, are the preferred solvents.

More particularly, the solvents include sulfolanes of the formula:

wherein each of R through R8 is at least one of hydrogen; hydroxyl; straight or branched chain alkyl, preferably having I to 12 carbon atoms, most preferably I to 6 carbon atoms in the alkyl cham, such as methyl, ethyl, isopropyl, butyl, octyl and dodecyl; a cycloaliphatic group including the monocyclic and bicyclic groups such as cyclopentyl, cyclohexyl and bicyclo [2.2.1] heptyl; or an aryl, alkyl-aryl, or aralkyl group such as phenyl, naphthyl, xylyl, tolyl, benzyl and beta-phenylethyl; an ether ofthe formula (O--R") wherein Rd may be aryl or lower alkyl having from I to 12 carbon atoms, preferably 1 to 4 carbon atoms in the alkyl chain; an alkylene or polyalkylene ether of the formula hOCnH2n)XOR wherein n has an average value of from I to about 4, x has an average value of from 1 to about 150, preferably 1 to about 20, most preferably 1 to about 4, and R"" may be hydrogen or alkyl having from 1 to 6 carbon atoms in the alkyl chain, such as poly(oxyethylene), poly(oxypropylene), poly(oxyethyleneoxypropylene), alkylene and polyalkylene glycols and lower alkyl ethers thereof; and a carboxylate group of the formula:

wherein y may have any value between 0 and 12, m and m" may be zero or one provided that when/either m or m" is one the other is zero, and R""" may be a lower alkyl group having from I to 12 carbon atoms, preferably from I to 4 carbon atoms, or aryl. Preferably the sulfone used in the practice of the present invention is tetrahydrothiophene-l,l-dioxide, better known as tetramethylene sulfone or sulfolane. In those instances where it may be desirable to use a substituted sulfolane those substituted in the 3 or 3,4 positions of the sulfolane ring are preferred.

The temperature which may be employed can vary over a wide range of elevated temperatures. In general, the process can be conducted at a temperature in the range of from about 100"C and upwards to approximately 3750C, and higher.

Temperatures outside this stated range are not excluded from the scope of the invention. At the lower end of the temperature range, and lower, the rate of reaction to desired product becomes markedly slow. At the upper temperature range, and beyond, signs of some catalyst instability are noted. Notwithstanding this factor, reaction continues and alkane polyols and/or their dervatives are produced. Additionally, one should take notice of the equilibrium reaction for forming ethylene glycol 2 CO+3H2HOCH2CH2OH At relatively high temperatures the equilibrium increasingly favors the left hand side of the equation. To drive the reaction to the formation of increased quantities of ethylene glycol, higher partial pressures of carbon monoxide and hydrogen are required. Processes based on correspondingly higher pressures, however, do not represent preferred embodiments of the invention in view of the high investment costs associated with erecting chemical plants which utilize high pressure utilities and the necessity of fabricating equipment capable of withstanding such enormous pressures. Suitable temperatures are between about 150"C to about 320"C, and desirably from about 210 C to about 300"C.

The novel process is effected for a period of time sufficient to produce the alkane polyols and/or derivatives thereof. In general, the residence time can vary from minutes to several hours, e.g., from a few minutes to approximately 24 hours, and longer. It is readily appreciated that the residence period will be influenced to. a significant extent by the reaction temperature, the concentration and choice of the catalyst, the total gas pressure and the partial presssures exerted by its components, the concentration and choice of diluent, and other factors. The synthesis of the desired product(s) by the reaction of hydrogen with an oxide of carbon is suitably conducted under operative conditions which give reasonable reaction rates and/or conversions.

The relative amounts of oxide of carbon and hydrogen which are initially present in the reaction mixture can be varied over a wide range. In general, the molar ratio of oxides of carbon to hydrogen is in the range of from about 20:1 to about 1:20, suitably from about 10:1 to about 1:10, and preferably from about 5:1 to about 1:5.

It is to be understood, however, that molar ratios outside the aforestated broad range may be employed. Substances or reaction mixtures which give rise to the formation of carbon monoxide and hydrogen under the reaction conditions may be employed instead of mixtures comprising carbon monoxide and hydrogen which are used in preferred embodiments in the practice of the invention. For instance, polyhydric alcohols are obtained by using mixtures containing carbon dioxide and hydrogen. Mixtures of carbon dioxide, carbon monoxide and hydrogen can also be employed. If desired, the reaction mixture can comprise steam and carbon monoxide.

The novel process can be executed in a batch, semi-continuous, or continuous fashion. The reaction can be conducted in a single reaction zone or a plurality of reaction zones, in series or in parallel, or it may be conducted intermittently or continuously in an elongated tubular zone or series of such zones. The material of construction should be such that it is inert during the reaction and the fabrication of the equipment should be able to withstand the reaction temperature and pressure. The reaction zone can be fitted with internal and/or-external heat exchanger(s) to thus control undue temperature fluctuations, or to prevent any possible "run-away" reaction temperatures due to the exothermic nature of the reaction. In preferred embodiments of the invention, agitation means to vary the degree of mixing of the reaction mixture can be suitably employed. Mixing induced by vibration, shaker, stirrer, rotatory, oscillation, ultrasonic, etc., are all illustrative of the types of agitation means which are contemplated. Such means are available and well-known to the art. The catalyst may be initially introduced into the reaction zone batchwise, or it may be continuously or intermittently introduced into such zone during the course of the synthesis reaction. Means to introduce and/or adjust the reactants, either intermittently or continuously, into the reaction zone during the course of the reaction can be conveniently utilized in the novel process and especially to maintain the desired molar ratios of and the partial pressures exerted by the reactants.

As intimated previously, the operative conditions can be adjusted to optimize the conversion of the desired product and/or the economics of the novel process. In a continuous process, for instance, when it is preferred to operate at relatively low conversions, it is generally desirable to recirculate unreacted synthesis gas with/without make-up carbon monoxide and hydrogen to the reaction. Recovery of the desired product can be achieved by methods well-known in the art such as by distillation, fractionation, extraction, and the like. A fraction comprising rhodium catalyst, generally contained in byproducts and/or normally liquid organic diluent, can be recycled to the reaction zone, if desired. All or a portion of such fraction can be removed for recovery of the rhodium values or regeneration to the active catalyst and intermittently added to the recycle stream or directly to the reaction zone.

The active forms of the rhodium carbonyl clusters may be prepared by various techniques. They can be preformed and then introduced into the reaction zone or they can be formed in situ.

The equipment arrangement and procedure which provides the capability for determining the existence of anionic rhodium carbonyl complexes or clusters having defined infrared spectrum characteristics, during the course of the manufacture of polyhydric alcohols from carbon monoxide and hydrogen, pursuant to this invention is disclosed and schematically depicted in U.S. Patent No. 3,957,857, issued May 18, 1976.

A particularly desirable infrared cell construction is described in U.S. Patent No., 3,886,364, issued May 27, 1975.

The "oxide of carbon" as covered by the claims and as used herein is intended to mean carbon monoxide and mixtures of carbon dioxide and carbon monoxide, either introduced as such or formed in the reaction. Preferably, the oxide of carbon is carbon monoxide.

The invention will now be further described by reference to the following Examples: Examples Materials used in the examples had the following characteristics: Tetraglyme (Ansul), cesium formate (Alfa), and cesium acetate (Alfa) were used without further purification. Sulfolane (Phillips) was purified as described in E. N. Arnett and C. F. Douty, J. Am. Chem. Soc., 86, 406 (1964), Cesium benzoate, [J. H. S. Green, W. Kynaston, and A. S. Lindsey, Spectrochim Acta, 17, 486(1961)1, (recryst. H2O. Anal. Found: C, 32.62; H, 1.90. Calcd. for C7H5O2Cs: C, 33.10; H, 1.98) and cesium pivalate [P. H. Reichenbacher, M. D. Morris, and P. S.

Skell, J. Am. Chem. Soc., 90, 3432 (1968)1, (washed with PhCI, recryst. H2O. AnaL Found: C, 24.75; H, 4.20. Calcd. for C5H9O2Cs: C, 25.66; H, 3.88) were prepared by use of literature procedures. Cesium p-methylsulfonyl benzoate (washed with ether, recryst. H2O. Anal. Found: C, 28.26; H, 2.05. Calcd. for C8H7O4SCs: C, 28.90; H, 2.13), cesium triphenylacetate (recryst. H2O. AnaL Found: C, 56.25; H, 3.74. Calcd. for C20H,502Cs: C, 57.16; H, 3.60), and cesium isobutyrate (washed with PhCI, recryst. H2O. Anal. Found: C, 21.01; H, 3.32. Calcd. for C4H7O2Cs: C, 21.84; H, 3.21) were prepared by reaction of CsOH and the corresponding acids.

[181 crown-6 solvent was obtained from Parish Chemical Company, Provo, Utah, and was heated under vacuum to remove possible volatile impurities and its purity was checked by vpc, nmr, melting point, and elemental analysis.

Procedure employed in examples: A 150 ml. capacity stainless steel reactor capable of withstanding presures up to 7,000 atmospheres was charged with a premix of 75 cubic centimeters (cc) of solvent, rhodium dicarbonylacetylacetonate, and promoter(s). The reactor was sealed and charged with a gaseous mixture, containing equal molar amounts of carbon monoxide and hydrogen, to the desired pressure. Heat was applied to the reactor and its contents; when the temperature of the mixture inside the reactor reached 1900C, as measured by a suitably placed thermocouple, the carbon monoxide and hydrogen (H2:CO=I:I mole ratio) pressure was adjusted to maintain the desired gas pressure. During the course of the reaction additonal carbon monoxide and hydrogen was added whenever the pressure inside the reactor dropped approximately 500 psi below the desired pressure. With these added repressurizations, the pressure inside the reactor was maintained within about 500 psi of the desired pressure over the entire reaction period.

After the reaction period, the vessel and its contents were cooled to room temperature, the excess gas vented and the reaction product mixture was removed.

Analysis of the reaction product mixture was made by gas chromatographic analysis using a Hewlett Packard FM (Trade Mark) model 810 Research Chromatograph.

Rhodium recovery was determined by atomic absorption anaylsis of the contents of the reactor after the venting of the unreacted gases at the end of the reaction. The rhodium recovery values recited below are the percent rhodium based on the total rhodium charged to the reactor that is soluble or suspended in the reaction mixture after the specified reaction time.

The same equipment and procedure was used in all the example in the the use of cesium carboxylates in excess of optimum concentration in [18] crown-6 solvent and its attendant increase in rhodium stability.

Figure 1 graphically depicts the inhibitory effect of the use of various carboxylate salts in excess of the optimum concentration. The dissociation constant of the conjugate acid for each salt is also provided.

Figure 2 illustrates the inhibitory effect of PhCO2Cs promoter in 2 solvents of different complexing ability. The inhibitory effect is shown to be weaker in the solvent of greater complexing ability, i.e., [18] crown-6.

Claims (14)

TABLE III Salts as Sole Promoter in [18] Crown-6 Salt, atoms cation Rates, M hr1 per 6 atoms Rh Temperature MeOH Glycol %Rh Recovery PhCO2Cs, 1.3 260 2.0 1.2 78 1.5 260 1.7 1.5 95 1.7 260 1.6 1.7 na 2.0 260 1.4 1.4 90 2.6 260 1.7 1.5 99 PhCO2K, 1.5 260 1.4 1.5 88

1.7 260 1.4 1.5 94 PhCO2Cs, 2.0 270 1.7 1.8 87

3.0 270 2.2 2.0 92 CH3CO2Cs, 1.7 270 3.3 1.7 84

2.6 270 3.1 2.2

3.4 270 2.4 1.8 100 (CH3)2CHCO2Cs, 2.0 270 2.8 1.6 82

3.0 270 2.5 1.9 90

4.0 270 2.7 1.7 105 PhCO2K, 1.7 270 2.3 2.3 84

2.0 270 2.8 2.3 89 PhCO2Cs, 2.0 280 3.5 2.5 79

3.0 280 3.0 2.5 87

4.0 280 2.8 2.2 91 CH3CO2Cs, 3.4 280 3.8 2.5 92

5.1 280 1.6 1.0 90 Reaction conditions: 75 ml solvent, 1.5 mmoles Rh(CO)zacac, 15,000 psi WHAT WE CLAIM IS: 1. A process of producing alkane polyols and/or derivatives thereof by the reaction of oxides of carbon and hydrogen in a homogeneous liquid phase mixture containing a solvent and a rhodium carbonyl complex catalyst in combination with a salt promoter; the catalyst concentration, the temperature and the pressure of between about 800 psia to about 50,000 psia being correlated so as to produce such alkane polyol; the promoter being provided in combination with the catalyst in an amount (depending on the promoter's basicity) to achieve not less than 50% of the optimum rate of formation of the alkane polyol at said correlated catalyst concentration, temperature and pressure of said mixture and using the same promoter and catalyst, the amount of the promoter being greater than the minimum amount which is sufficient to produce such optimum rate of formation.

2. A process as claimed in claim 1 wherein the homogeneous liquid phase mixture additionally contains an amine promoter.

3. A process as claimed in claim 1 or 2 wherein the solvent is tetraglyme.

4. A process as claimed in claim I or 2 wherein the solvent is sulfolane.

5. A process as claimed in claim I or 2 wherein the solvent is a crown ether.

6. A process as claimed in any one of the preceding claims wherein the oxide of carbon is carbon monoxide.

7. A process as claimed in any one of the preceding claims wherein the temperature is between 100"C and 375"C.

8. A process as claimed in claim 7, wherein the temperature is between 1500C and 320"C.

9. A process as claimed in claim 8, wherein the temperature is between 210 C and 300"C.

10. A process as claimed in any one of the preceding claims, wherein the molar ratio of oxides of carbon to hydrogen varies from 20:1 to 1:20.

11. A process as claimed in claim 10, wherein the molar ratio of oxides of carbon to hydrogen varies from 10:1 to 1:10.

12. A process as claimed in claim 11, wherein the molar ratio of oxides of carbon to hydrogen varies from 5:1 to 1:5.

13. A process for producing alkane polyols as claimed in claim I and substantially as hereinbefore described in any one of the foregoing Examples.

14. An alkane polyol whenever produced by a process as claimed in any one of the preceding claims.