AN ENVIRONMENTALLY BENIGN PROCESS

FOR FRIEDEL-CRAFTS ACYLATION

ABSTRACT

We have developed a new method for Friedel-Crafts acylation of aromatic substrates with carboxylic acids, using a metal trifluoromethanesulfonate catalyst. The reaction is driven to completion (over several days) by the azeotropic removal of water. A wide variety of carboxylic acids may be used to acylate even weakly activated aromatic substrates in good yield, without the need to synthesize or use hazardous acid chlorides. Unlike aluminum chloride (the traditional "catalyst"), our catalyst is not toxic, water sensitive or corrosive, it is wholly reusable, and it may be used (in catalytic amounts) either as a solid or an aqueous solution. The work up is extremely simple, and yields a "crude" product which often requires no further purification. Only a small amount of aqueous sodium bicarbonate is produced, instead of the large quantity of strongly acidic waste generated when aluminum chloride is used.

INTRODUCTION

Friedel-Crafts acylation (Scheme 1) is an important industrial process, used for the preparation of a variety of pharmaceuticals, agrochemicals, and other chemical products.

Scheme 1. Friedel-Crafts acylation.

Its utility centers around the fact that it is one of only a very few ways of forming a new carbon-carbon bond onto an aromatic ring. Currently, the most widely-used method has three serious drawbacks:

  1. The carboxylic acid needs to be converted first (in a separate step) to a reactive and corrosive derivative, the carboxylic acid chloride. This requires the use of hazardous reagents such as thionyl chloride.
  2. The catalyst used (anhydrous AlCl3) is a water-sensitive, corrosive solid, which causes severe problems in waste streams, since aqueous neutralization of it results in a gelatinous mass. These waste streams have to be disposed of as acidic waste, adding significantly to the commercial cost and the environmental impact.
  3. The aluminum chloride catalyst forms a stable complex with the product, so that a full molar equivalent of catalyst (often more) is required AND the product can only be recovered by destruction of the catalyst (i.e., the catalyst cannot be reused). Both of these factors add significantly to the cost of operating the process.

Attempts have been made to overcome these problems, with only partial success. Use of catalytic quantities of a corrosive protic acid (HF, H2SO4 or CF3SO3H, all difficult to recycle) allows the carboxylic acid to be used. Another process from the carboxylic acid utilizes full molar amounts of (CF3CO)2O to form a reactive mixed anhydride in situ, but highly corrosive materials ((CF3CO)2O, P4O10) are still involved. Recyclable, non-toxic catalysts such as metal trifluoromethanesulfonate (triflate) salts have been used, but always with a hazardous acid chloride or acid anhydride which entails an extra step. Such procedures also require drying of the catalyst in vacuo at 200oC before re-use. Addition of certain metal salts makes it possible to recycle AlCl3, but this does not affect the hazardous nature of the catalyst or avoid use of the acid chloride as a second step.

The goal of this work was to design a process that addresses all three of the present problems, yet remains suitable for large scale manufacturing. The process described below allows aryl ketones to be prepared in one step from carboxylic acids using only a catalytic amount of a reusable catalyst which is both non-toxic and non-corrosive.

RESULTS AND DISCUSSION

In our procedure, an aromatic substrate is reacted with a carboxylic acid in the presence of a hydrated metal trifluoromethanesulfonate (triflate) salt, with azeotropic removal of water to drive the reaction. Toluene is both an effective substrate and a solvent for more activated substrates (e.g. anisole), though alternative solvents such as chlorobenzene may be used. The catalyst may be added as a solid or as an aqueous solution. After a simple workup, the crude product is essentially pure, except for small amounts of the ortho acylation product; if need be, the product may be further purified by recrystallization.

In a typical example, hydrated cerium (III) trifluoromethanesulfate (3.23 g) and para-toluic acid (1.36 g, 10 mmol) were refluxed together in toluene (125 mL), with azeotropic removal of water (Dean-Stark trap). After 100 hours reaction time, the mixture was cooled and extracted with 3 x 25 mL water (to recover the catalyst). The organic phase was then washed with 2 x 25 mL 8% NaHCO3 solution, dried over Na2SO4, then stripped to dryness in vacuo to give 4,4'-dimethylbenzophenone (1.65 g, 79 % yield) which was essentially pure (M.P. 84.5-86.5oC and by NMR). The aqueous extracts (excluding NaHCO3) may then be used directly as the catalyst for a fresh reaction, without further workup.

Variation of catalyst

A wide variety of metal triflates were tested, with triflates of the lighter lanthanides generally being the catalysts of choice (table 1). In work by other authors, scandium and ytterbium(III) salts are often the more effective recyclable catalysts, but these were always used under anhydrous conditions (unlike in our work, where water is necessarily present). In(III) and Bi(III) triflates could not easily be recycled because of hydrolysis.

Addition of triflic acid to the mixture (as an activator6a) had no noticeable effect. Cerium (III) p-toluenesulfonate was completely ineffective as a catalyst.

Table 1. Variation of catalyst in acylation of toluene.

Metal triflate

Sc

Y

In

La

Ce

Pr

Dy

Er

Yb

Bi

Th

Benzoic acid rxn. % yield

(168 h reflux)

-

20

-

-

26=

77

94

12

17

-

11

p-Toluic acid rxn. % yield

(24 h reflux)

7

17

60*

16

28

31

32

-

20

47*

45*

* Gave impure product = Believed to be low due to physical losses

All catalysts were used as the metal(III) triflate, except for Th(OTf)4.

All reactions used toluene as both solvent and substrate.

Bi and In triflates could not be recycled due to hydrolysis. Th salts are toxic.

Although 0.6 molar equivalents of catalyst (if assumed dry) were used in the above experiments, smaller amounts may be used. For example, a yield of 73% was obtained from toluene and benzoic acid (after 144 h), using only 0.06 molar equivalents Ce(OTf)3.

Scope and limitations of the reaction

Even weakly activated aromatics such as toluene are acylated in good yield; more highly activated compounds such as anisole react smoothly in refluxing toluene. However, non-activated substrates such as benzene gave little or no reaction. A variety of carboxylic acids are effective acylating agents, including many aliphatic acids (which fail in some alternative procedures), as shown in table 3 below. One notable exception occurs with a strong electron withdrawing group conjugated to the carboxyl group, e.g., in p-nitrobenzoic acid, where acylation only occurs in low yield.

Table 3. Variation of reaction substrate.

Aromatic substrate

Carboxylic acid

Catalyst

Rxn.

Time (h)

% Yield crude

ketone

Benzene

p-anisic acid

Ce(OTf)3

168

1*

Anisole

p-toluic acid

Ce(OTf)3

48

88

Anisole

p-anisic acid

Dy(OTf)3

168

92

Anisole

2-methylbutanoic acid

Dy(OTf)3

24

44

Toluene

p-toluic acid

Ce(OTf)3

100

79

Toluene

p-anisic acid

Ce(OTf)3

168

63

Toluene

p-nitrobenzoic acid

Dy(OTf)3

24

2*

Toluene

m-nitrobenzoic acid

Dy(OTf)3

168

39

Toluene

Cyclohexanecarboxylic acid

Ce(OTf)3

168

61

SUMMARY

The new Friedel-Crafts acylation method is applicable to both weakly and strongly activated aromatic substrates, and a variety of aromatic and carboxylic acids. Good yields of aryl ketones may be obtained directly from the carboxylic acid in one step, using a process which totally avoids the use of hazardous reagents.

ACKNOWLEDGEMENTS

College of the Holy Cross, SUNY Potsdam, Brandeis University and Johnson State College, for use of facilities and for funding. I would like also to thank Prof.. James B. Hendrickson and Jian Wang (Brandeis University) for their support and encouragement.

REFERENCES

Back to the Martin Walker's Research Page
Back to Martin Walker's Home Page