METHOD OF PREPARING MONOESTERS

TECHNICAL FIELD

The present invention relates generally to mono- esters of dicarboxylic acids and monoesters of dihydric alco¬ hols and, more particularly, to monoesterification of symmet¬ rical, dicarboxylic acids and diols respectively with mono¬ hydric alcohols and monocarboxylic acids to effect excellent yields (>90%, usually >95%) .

Because monoesters of dicarboxylic acids or dihy¬ dric alcohols have both a reactive ester and acid or alcohol functionality, they are of value as intermediates in the syn¬ thesis of many types of useful chemicals. Such compounds are useful in the manufacture of fine organic chemicals, as star¬ ting materials for preparing ω-amino acids, in the synthesis of certain types of nylon, as precursors to macrocyclic e- tones and lactones valued in perfumery, and as intermediates in the synthesis of insect pheromones. As one example, mono- ethyl sebacate prepared according to the present invention may be used as a starting material in the following synthesis of traumatic acid — a material useful both as a detergent additive (U.S. Patent No. 3,523,636), and as an anti-viral agent (U.S. Patent No. 3,542,826):

0 0 _. 0

HO - C - (CH ₂ )gC - OCH ₂ CH ₃ S0C1-, Cl - C(CH ₂ ) gC0 ₂ CH ₂ CH ₃

H /poisoned Pd catalyst H - C - (CH_) _o C0„CH_CH y 2' 8 2 2

(CH ₃ CH ₂ 0) ₂ PCH ₂ COCH ₂ CH ₃ , base hydrolysis or alonic acid, pyridine & heat

0 H (CH. _o C00H

HO - C - C =

<\

H

Monoesters of a, ω-diacids prepared in accordance with the present invention can also be utilized as starting materials in the preparation of nylon in a process suggested by C.S. Rondestvedt, Jr. [J. Org. Chem. , 42, 3118 (1977)].

Similarly, the monoacetate derivative of 1,8-octanediol can be used to prepare the attractant of the Oriental fruit moth:

0

CH- JC, _l -0(CH-_,) _Q O0H— >CHJ,C _M -0(CH_,)_/CH CH ₃ C-0(CH ₂ ) _? ^CH 2 ^CH 2 ^CH 3

\ C=C

/ \

H H

BACKGROUND ART

Although monoesters of dicarboxylic acids and di- hydric alcohols are useful intermediates, a problem which has hindered prior art efforts to synthesize such compounds is the occurrence of esterification at both reactive sites. The result is the formation of a large quantity of diester as an undesired reaction product. These diesters are typically dif¬ ficult to separate from the reaction mixture.

One prior art approach to the preparation of such monoesters has been to start with the unwanted diester and, via selective hydrolysis, to convert one ester function to an alcohol or acid function. For example:

CH

With the notable exception of the diester of terephthalic acid [Rondestvedt, J. Org. Chem. , 42, 3118 (1977)3 _/ this ap¬ proach suffers from moderate yields (generally in the range of 40=70%) of the desired monoester, as well as some diffi¬ culty in separating the desired monoester from the reaction mixture, [i.e.. Organic Synthesis, Collective Vol. 4, p. 635 (1963)3.

Other approaches to the problem can be found in the literature, viz. , J. Am. Chem. Soc. , 70, 364, 3206 (1948); J. Org. Chem., 29, 1252 (1964)? Org. Syn. Coll. Vol. 2, pp. 276-77 and references therein. All of these procedures pre¬ sent poor yields of the monoester and difficulty in elimina¬ ting the diester from the reaction product.

Of substantial interest to the background of the invention is the fact that the prior art uniformly suggests that esterification reactions should be run under essentially anhydrous reaction conditions. Indeed, since the process of ester formation from an acid and an alcohol is reversible, the reaction is generally performed under conditions provi¬ ding for the water to be removed from the reaction mixture as soon as it is formed. This is necessary since water in the presence of a strong acid catalyst is known to effect the reverse process —- i.e., hydrolysis of an ester.

H+

ROH + R'C-OH < ^» R'C-OR + H„0

II II ²

0 o

DISCLOSURE OF INVENTION

According to the present invention monoesters of di- acids and monoesters of diols are prepared in exceptionally high yields accompanied by the substantial reduction of dies¬ ter products. The method of the invention, in contrast to the

methods of the prior art, requires the presence of a substan¬ tial amount of water. The method can be illustrated by the following examples:

H O HOOC-R-COOH + R'-CH ₂ OH - > R'CE^OOC-R-COOH + H ₂ 0

H+

II

H O

HO-R-OH + R'COOH R'COOR-OH + H O

H+

III IV

[wherein R is an alkyl group (-CH-) and R ¹ is an alkyl group or H.3

Although the exact amount of water employed may vary with the solubility of the diacid or diol in the water-alcohol or water- acid reagent mixture, the molar concentration of water should always be at least equal to the molar concentration of the monohydric alcohol or monocarboxylic acid (II or IV) . Indeed, it has been found that for a given diacid or diol, the more water used (consistent with solubility limits of the reac- tants) , the better the results. If less water is used, the desired monoesterification is virtually impossible to accom¬ plish in even moderate yield.

In the case of both the diacid and diol starting materials, the reaction proceeds at room temperature. A

_2 strong acid catalyst (Ka>10 ) such as sulfuric acid greatly aids the reaction. In its absence, esterification can still occur — albeit quite slowly. The process of selective mono¬ esterification depends upon the use of a technique of contin¬ uous extraction of the reaction mixture with a nonpolar sol¬ vent such as a liquid alkane, a cycloalkane, an aromatic hydrocarbon, a halide derivative of a hydrocarbon, or mixtures thereof. These nonpolar solvents are, of course, essentially insoluble in the aqueous reaction mixture. In a preferred

method the solvent is heated to boiling, condensed, and the condensate passed through the aqueous reaction mixture by using, for example, a continuous extraction apparatus employ¬ ing a reflux condenser.

Suitable solvents include hexane, heptane, cyclo- hexane, benzene, toluene, xylene, and carbon tetrachloride. Cyclohexane, in particular, works quite well. As a rule, starting material diols and diacids should be soluble in the aqueous reaction mixture yet virtually insoluble or of very low solubility in the nonpolar solvent used for the contin¬ uous extraction process. The resultant monoester product, however, is very soluble in the nonpolar solvent being passed through the reaction mixture. Thus, as soon as the less polar monoester product has formed, it is extracted out of the reac¬ tion mixture. The presence of a large amount of water (always >50% on a molar basis) permits this extraction to occur more readily and greatly retards the likelihood of diester forma¬ tion.

A diacid or a diol suitable for use in the practice of the present invention must be symmetrical and have some appreciable solubility in a water-alcohol or water-acid mix¬ ture. Thus, while diacids and diols containing from 2 to 12 carbon atoms are particularly suitable, the process should also be useful for larger molecules having some degree of sol¬ ubility in a water-alcohol or a water-acetic acid (for diols) mixture.

The symmetrical diols employed in processes accor¬ ding to the invention may be either primary or secondary acyc¬ lic or alicyclic compounds. Particularly useful compounds can be represented by formulae JL through 3_,

(1) HO-CH_(CH_) CH_-OH

2 2 n J_ wherein: n has a value of from 0 to 10 inclusive

OH OH I I

(2) CH -(CH ) CH (CH ) CH(CH-) CH_

_3 __> _?_. & y _- x j wherein: x has a value of from 0 to 4; y has a value of from 0 to 8? and 2x + y_ <8

wherein m has a value of from 1 to 2. The process fails for dihydric phenols such as resorcinol (

Symmetrical diacids useful in practice of the in tion include acyclic diacids of the general structural for 5_, wherein R represents a chemical bond or a straight chai alkylene having up to ten carbons, and the cyclic diacid, 1,4-cyclohexanedicarboxylic acid, of formula 6_. The proces is unsuccessful both with terephthalic acid, 7_, since it is insoluble in water-alcohol, and with phthalic acid, 8_, sinc it readily forms the corresponding anhydride, 9.

0 0

II II

Monohydric alcohols useful in the esterification of diacids according to the invention include both methanol and ethanol. It has been found that higher molecular weight, water-soluble alcohols, such as n-propanol and isopropanol, are unsatisfactory. Despite their solubility in water, such higher monohydric alcohols tend to be soluble in the nonpolar organic solvent used in the continuous extraction process. Moreover, when the nonpolar solvent "picks up" the higher al¬ cohol, this mixture very quickly extracts over the starting diacid making it virtually impossible to carry out the esteri¬ fication reaction in the aqueous layer. More significantly, the reaction vessel containing the refluxing nonpolar organic solvent will, when higher monohydric alcohols are employed, build up an appreciable concentration of the alcohol and an esterification reaction can occur in the organic solvent mix¬ ture, leading to the formation of the undesired diester.

Similar problems arise upon attempted use of propi- onic or higher water-soluble acids to monoesterif diols. These higher acids are sufficiently soluble in the nonpolar solvent to prevent their use in the continuous esterification process. Thus the only monocarboxylic acids useful for the monoesterification of diols are alkanoic acids having up to _. two carbon atoms. Of this group, acetic acid is the preferred monocarboxylic acid, although formic acid can be used ™ if the selected diol is soluble in an aqueous formic acid mixture.

The procedure employed to isolate the desired mono¬ ester from the starting reactants can vary depending on the nature of the starting diacid or diol. Gas chromatographic analysis ("GC") of the products derived from reaction invol¬ ving low molecular weight, non-crystalline diols such as ethyl- ene glycol did not indicate the presence of any starting diol. Thus no further isolation is necessary, save removal of the nonpolar solvent by distillation.

The C-2 through C-12 diacids, as well as the C-6 through C-12 diols, are all highly crystalline and virtually insoluble in the nonpolar solvent used for the continuous ex¬ traction process. As soon as the extraction solvent is allowed

to cool to room temperature, the diacid (or diol) that has been extracted over (if any at all) by the nonpolar organic solvent can be removed by simple filtration and may be re¬ cycled. As the number of carbons in the diacid or diol be¬ comes larger (i.e., 10 or more carbons), this process of re cycling the starting material becomes more desirable.

After removal of any starting diacid or _: diol by f tration, the nonpolar solvent can be removed by distillatio (under reduced pressure, if one so desires) and the purity of monoester so obtained is >95% in most systems examined. If one desires even higher purity, recrystallization of the monoesters from a suitable solvent may be easily performed since many of the monoesters are solids at room temperature Alternatively, in the case of diacid derived products, the monoester is still acidic and can be separated from any di¬ ester by extraction into an aqueous phase using a weak base such as sodium bicarbonate. Subsequent acidification of th aqueous layer containing the acid salt allows recovery of t desired monoester.

BEST MODE FOR CARRYING OUT THE INVENTION

The following examples illustrate in greater deta practice of the present invention.

Example I 1,10-decanediol monoacetate

A solution of 4.36 g (25 mmoles) of 1,10-decanedi in 100 ml (1.77 moles) of glacial acetic acid was mixed wit 130 ml (7.22 moles) of H ₂ 0 containing 0.25 ml (4.5 mmoles) cone. H ₂ S0 ₄ . This mixture, itself at room temperature, was then extracted continuously for two days using refluxing cy clohexane. The cyclohexane layer was then cooled to room t perature and 0.78 g (18% recovery) of crystalline 1,10-deca diol was recovered by simple filtration of this nonpolar or ganic phase. The cyclohexane was then removed from the fil trate under reduced pressure, affording 4.28 g (approximate 79% yield — 97% based on recovered starting material of pr duct. GC analysis showed the presence of less than 2% star

diol? the ratio of monoester to diester was 20 to 1. Hence the desired monoester was approximately 95% pure.

Example II

In a procedure similar to that of Example I, a solu¬ tion of 4.25 g (24.4 mmoles) of 1,10-decanediol and 0.25 ml of cone. H-SO. in 200 ml (3.53 moles) of glacial acetic acid and 30 ml (1.66 moles) of H ₂ 0 was extracted continuously with cyclohexane for 20 hours. The cyclohexane layer was then cooled to room temperature and the solvent was removed under reduced pressure using a rotary evaporator leaving behind 5.52 g of crude product. GC analysis indicated the presence of 3% diol, 41% monoester, and 56% of the diester — 1,10- deσanediol diacetate. Since this procedure used only ^< 33% H„0 (on a molar basis) , the need for water (to minimize for¬ mation of the undesirable diester) is established. In the previous Example I, the aqueous layer was ^80% H ₂ 0 and ^20% acetic acid on a molar basis.

Example III Monoethyl sebacate

A solution of 3.093 g (.15.3 mmoles) of sebacic acid in 100 ml (1.63 moles) of 95% ethanol was mixed with 120 ml of H-O (6.94 moles H ₂ 0, total) containing 1 ml (18 mmoles) of cone. H ₂ SO,. [This aqueous solution is approximately 81% H O and 19% ethyl alcohol on a molar basis.] This solution was extracted continuously for two days using refluxing cyclo¬ hexane. The cyclohexane layer was then cooled to room temp¬ erature and 0.74 g (24% recovery) of sebacic acid was recovered by simple filtration of this nonpolar organic phase. After filtering off the sebacic acid, the cyclohexane layer was washed 15 times with 1M aqueous NaHCO, solution (15 x 10 ml) in order to effect separation of monoethyl sebacate from any diethyl sebacate, which remained in the organic layer. Each of these sodium bicarbonate washes was added to an Erlenmeyer flask containing 100 ml of ice-cold 2M aqueous HC1 solution. The cyclohexane layer was then dried over anhydrous magnesium sulfate and filtered. Removal of the cyclohexane by evap¬ oration at reduced pressure afforded 103 g (.0.40 mmole.

2.6% yield) of diethyl sebacate. The monoester was recove from the aqueous hydrochloric acid mixture by thorough ext tion with ethyl ether. The combined extracts were washed with 10% sodium chloride solution, dried over anhydrous ma nesium sulfate, and filtered. Removal of the ether by eva oration at reduced pressure yielded 2.25 g (9.78 mmoles, 64 yield) .of monoethyl sebacate. The ratio of monoester to di ester in this procedure was therefore 64 to 2.6 or approxi mately 25 to 1. Since the purity of monoethyl sebacate wit out separating it from any diester is >96%, there appeared to be no substantial need for the sodium bicarbonate washes and the rest of the workup procedures.

Example IV

A procedure similar to that of Example III was ca ried out by mixing a solution of 3.09 g (15.3 mmoles) of se acic acid in 150 ml (2.44 moles) of 95% ethanol with 20 ml of H-O (1.53 moles H_0, total) containing 0.75 ml (13.5 mmoles) of cone. H SO.. This solution was then extracted c tinuously with refluxing cyclohexane for 27 hours. The rea tion products were isolated as described in Example III. Results: 617 mg (20% recovery) of sebacic acid which can b re-cycled in this monoesterification process; 753 mg (2.92 mmoles, 19% yield) of diethyl sebacate? and 1.72 g (7.48 mmoles, 49% yield) of monoethyl sebacate. The ratio of mon ester to diester in this procedure was therefore 49 to 19 o approximately 2-1/2 to 1. It is noteworthy that the aqueou solution in this procedure was approximately 38% water on a molar basis and the concentration of H-SO. was approximatel 0.08M. In Example III, the concentration of sulfuric acid was approximately Q.08M and the aqueous solution was 81% water on a molar basis.

Example V Monoethyl adipate

A solution of 4.00 g (27.4 mmoles). of adipic acid in 30 ml of absolute ^' ethanol (514 mmoles) was mixed with 18 ml (10 moles) of water containing 4 ml (72 mmoles) of cone.

H~SO.. This solution was then extracted continuously for five days using benzene. The benzene layer was then cooled to room temperature and trace amounts of crystalline adipic acid were removed by filtration prior to isolation of the reaction products. After filtration, the benzene layer was washed thoroughly with IM aqueous sodium bicarbonate solution, each wash being added to a flask containing sufficient aqueous hydrochloric acid to neutralize all of the NaHC0 ₃ , in order to liberate monoethyl adipate from its salt. Dilute aqueous Na-CO- solution, but not dilute aqueous NaOH, can be used in lieu of the sodium bicarbonate solution. If one uses aqueous sodium hydroxide to effect conversion of monoethyl adipate to a water-soluble salt, a substantial amount of hy¬ drolysis of the monoester occurs while it is dissolved in the the aqueous NaOH solution. The benzene layer (containing di¬ ethyl adipate) was then dried over anhydrous magnesium sul- fate and filtered. Removal of the benzene by evaporation at reduced pressure afforded 206 mg (1.02 mmoles, 3.7% yield) of diethyl adipate: bp (bath temperature) 55-65° at 0.2mm. The monoester was recovered from the aqueous hydrochloric acid mixture by thorough extraction with ethyl ether. The combined extracts were washed with 10% sodium chloride solution, dried over anhydrous magnesium sulfate, and filtered. Removal of the ether by evaporation at reduced pressure yielded 4.16 g (87.5% yield) of monoethyl adipate as a low melting solid,, mp. (after one recrystallization from ether-pentane) : 28-30°C.

Since the ratio of monoester to diester was approx¬ imately 96:4, there seems to be no substantial need for the sodium bicarbonate washes for the purification of the monoester via formation of a water-soluble carboxylate salt. Simple re¬ moval of the benzene should yield a product that is 96% pure. If greater purity is desired, the monoester can be recrystal- lized.

Example VI Monoethyl dodecanedioate

A solution of 4.616 g (20.1 mmoles). of dodecanedioic acid in 150 ml (2.44 moles) of 95% ethanol was mixed with 80.

OMPI '-~vmό y

ml (4.44 moles) of H ₂ 0 containing 1 ml ( _. 18 mmoles) of cone. H-SO.. The mixture was not homogeneous at this point; as much as 50% of the starting diacid precipitated out of solu tion when the water was added to the ethanol containing thi C-12 diacid. As the reaction proceeded, however, the diaci slowly re-dissolved in the aqueous layer and the reaction was able to be completed within the normal period of time.

This mixture was extracted continuously for 2-1/2 days using cyclohexane. The cyclohexane layer was then cooled to room temperature and 1.40 g (30% recovery) of dod canedioic ^" acid was recovered by simple filtration of this nonpolar organic phase. After filtering off this solid di¬ acid, the monoester was separated from any diester in the cyclohexane layer by thorough extraction with IM aqueous NaHC0_ solution. The exact procedure followed has previous been described in Example III. Results: The■yield of di¬ ethyl dodecanedioate was only 63 mg (0.22 mmole, 1% yield); the recovery of monoethyl dodecanedioate was 3.58 g (13.9 mmoles, 69% yield) . The mp of this monoester (after recrys tallization from ethanol-H ₂ 0) was 37-38°C. Since the ratio of monoester to diester was 69:1, there is no need for the lengthy workup procedure involving NaHC0_. After filtering off any recovered starting diacid, the cyclohexane can be re moved by evaporation under reduced pressure and the monoest should at that point be >98% pure. In a separate procedure, only 2 g of dodecanedioic acid was used and the rest of the reagents were kept the same. In this latter procedure, the aqueous layer remained homogeneous but the final results (product ratios, etc.) were the same as the first procedure. Hence it is not necessary that the diacid be totally dissol in the aqueous layer.

Example VII 1,8-octanediol monoacetate

A solution of 1.987 g (JL3.7 mmoles) of 1,8-octa- nediol in 25 ml (441 mmoles) of glacial acetic acid was mixe with 200 ml (11.11 moles) of H-0 containing 4 ml ( _. 72 mmoles)

OMP

of cone H ₂ S0.o This mixture was extracted continuously for 40 hours using hexane. The monoester is not very soluble in hexane and hence a mixture of hexane-cyclohexane (in which the monoester, but not the starting diol, is soluble) should be recommended if one is interested in developing this pro- cess for 1,8-octanediol. After cooling the hexane to room temperature, ether was added to make the organic layer homo¬ geneous. The ether-hexane layer was then dried over anhy¬ drous potassium carbonate and filtered. Removal of the sol¬ vent.under reduced pressure afforded 2.38 g (-92% yield) of product, GC analysis of which indicated the presence of 2% starting diol and 4% diester. The purity of the desired mono¬ ester was therefore 94%.

Example VIII When selected diacids and diols are reacted with aqueous solutions of a monohydric alcohol or a monocarbox¬ ylic acid in the presence of a strong acid catalyst, mono¬ esters can be isolated as illustrated in Table I.

TABLE I

Reagent Aqueous Reagent Catalyst Time Solvent • % .Recovery Ratio of Mixture Used for of Smarting Monoester Extraction Material to Diester

61.5 mmoles of 80 ml of 95% 1 ml 3 benzene 4% 90:1 oxalic acid ethanol and cone. days dihydrate 160 ml of H ₂ 0 H ₂ S0 ₄

29 mmoles of 100 ml of 1 ml 24 benzene <5% 92:1

1,4-cyclohexane- CH-OH and 200 cone. hours diearboxylic acid ml ^J of H O H ₂ S0 ₄

15 mmoles of 75 ml of 0.25 ml 1-1/2 5:1 (v/v) 35% 60:1 1,10-decanediol glacial ace¬ cone. days cyclohex¬ (which tic acid and H ₂ S0 ₄ ane: CC1. can be 155 ml of H ₂ 0 re-cycled)

7 mmoles of 150 ml of 0.25 ml 3.0 25% 66:5

1,12-dodecane- glacial ace¬ cone. hours cyclo¬ (re-cycl- diol tic acid and H ₂ S0 ₄ hexane able) 90 ml of H ₂ 0

Reagent Aqueous Reagent Catalyst Time Solvent % Recovery Ratio of

Mixture Used for of Starting Monoeste

Extraction Material to Diest

17 mmoles of 45 ml of 4.0 ml 4 benzene ^" 14%* 85:1

1,4-cyclohexane- glacial ace¬ cone. days (can be diol tic acid and H ₂ S0 ₄ precipita¬

180 ml of H ₂ 0 ted out)

17 mmoles of 60 ml of 4.0 ml 4 2:1 (v/v) 2% 90:8

1,4-cyclohexane- glacial ace¬ cone. days cyclohexane: diol tic acid and H ₂ S0 ₄ benzene

180 ml of H O

15 mmoles of 90 ml of 0.25 ml 1-1/2 1:1 (v/v) 22% 75:3

1,10-decane- glacial ace¬ cone. days hexane: (Re-cycl- diol t ic acid and H ₂ S0 ₄ cyclohexane able)

15 mmoles of 100 ml of None 3 cyclohexane 56% >40:1

1,10-decane- glacial ace¬ days diol tic acid and 130 ml of H ₂ 0

* Can be precipitated out of the benzene layer by the addition of hexane.