Dimethyl carbonate–water: an environmentally friendly solvent system for ruthenium tetraoxide oxidations

Abstract

Dimethyl carbonate (DMC)–water is an environmentally benign solvent system for ruthenium tetraoxide oxidations of various substrates including alkenes, alkynes, arenes, alcohols, ethers and aldehydes. Either hydrated ruthenium trichloride or hydrated ruthenium dioxide can be used as sources of ruthenium, while suitable cooxidants include sodium periodate, bleach and Oxone^®.

---

Green Context

The clean oxidation of organics is a pressing problem which is being addressed in many ways. This paper is about the development of a benign solvent system for Ru catalysed oxidations, which often utilise a mixed solvent system containing carbon tetrachloride, making it impossible to use industrially in several countries. It is shown that the use of aqueous dimethyl carbonate leads to the ready reoxidation of Ru species, without the need for complex solvent mixtures. While this is a significant step towards a green oxidation process, further improvements in the stoichiometric oxidant would make this an extremely attractive methodology for a number of oxidations.

DJM

Introduction

Ruthenium tetraoxide oxidation is a powerful method for oxidation of organic compounds. The reaction is fast and the products are usually isolated in a high yield. In the course of an oxidation, ruthenium tetraoxide is prepared in situ by addition of a suitable cooxidant to a solution or a suspension of either hydrated ruthenium trichloride or hydrated ruthenium dioxide in a mixture of solvents. As the reaction proceeds, ruthenium tetraoxide is reduced to lower valent species. In turn, those lower valent ruthenium species are reoxidized by a cooxidant to ruthenium tetraoxide. Thus, the cooxidant is used up, while the actual oxidizing agent is ruthenium tetraoxide. The main problem in ruthenium tetraoxide catalyzed oxidations is complexation of lower valent ruthenium species by the oxidation products, in particular carboxylic acids, to give insoluble complexes. This causes the reaction to slow down and eventually cease.^1,2 The problem has been addressed by Sharpless and coworkers, who added acetonitrile to the reaction mixture.¹ Acetonitrile is stable under the reaction conditions and is capable of complexing lower valent ruthenium species, thus keeping them in solution and allowing them to be reoxidized to ruthenium tetraoxide. However, in addition to acetonitrile, water has to be added to dissolve the cooxidant and yet another solvent is necessary to dissolve the formed ruthenium tetraoxide. That solvent is usually a chlorinated solvent, with carbon tetrachloride being the most commonly used. Since carbon tetrachloride has been identified as a solvent responsible for destruction of the stratospheric ozone, further use of carbon tetrachloride in the USA has been banned by the Environmental Protection Agency.³ Use of solvent systems consisting of water and acetonitrile was reported.⁴ However, it gave poor results in our hands. The solvent system that was successfully used in RuO₄-catalysed oxidative cyclisations (EtOAc–CH₃CN–H₂O)⁵ was also unsatisfactory.† In order to improve the reaction system we investigated dimethyl carbonate (DMC)–water as a solvent system. We speculated that DMC, as a highly oxygenated compound, would have an ability to complex ruthenium ion thus allowing the reoxidation of lower valent ruthenium species back to ruthenium tetraoxide. This solvent system has an advantage that it is environmentally friendly.⁶

Results and discussion

Either hydrated ruthenium trichloride or hydrated ruthenium dioxide can be used as a source of ruthenium. Suitable cooxidants are sodium periodate,⁷ sodium hypochlorite (household bleach)⁸ or Oxone^® (potassium hydrogen persulfate).⁹ Oxidations with Oxone^® as a cooxidant proceeded at a slightly lower rate compared to periodate as a cooxidant and a larger amount of precipitate formed in the course of a reaction. Nevertheless, Oxone^® is an inexpensive alternative to periodate cooxidants. Potassium bromate¹⁰ did not appear to be a suitable cooxidant. Reaction rates were slower and the yields were lower compared to periodate or hypochlorite cooxidants.

Examples of oxidation of organic substrates by ruthenium tetraoxide with sodium periodate as a stochiometric reagent in DMC–water solvent system are shown in Table 1. Most organic substrates are soluble in DMC. However, 1-octadecanol (Table 1, entry 2) was insoluble and was added to the reaction mixture as a powder. After about 1 h, it completely dissolved and the oxidation was completed after 2 h. It was reported that ruthenium tetraoxide oxidation of ethers proceeds cleanly and in good yields.^1,11 Ruthenium tetraoxide oxidation of tetrahydrofuran in DMC–water provided only a modest yield of the γ-butyrolactone as evidenced by GLC (Table 1, entry 7). The GLC yield peaked at 48% after 1.5 h. Apparently, the lactone hydrolyzed and was further oxidized to glutaric acid. It was recently reported that ethyl acetate–bleach may be a suitable solvent–cooxidant system for oxidation of ethers.¹² Aromatic compounds were oxidized by a catalytic amount of ruthenium tetraoxide with sodium periodate as a stochiometric oxidant in CCl₄–CH₃CN–H₂O solvent system in 60–91% yields.¹³ Oxidation of simple aromatic compounds in the DMC–water solvent system gave the corresponding carboxylic acids in good yields (Table 1, entries 8, 9). Oxidation of cyclopropylphenylmethane gave cyclopropyl phenyl ketone as the major product (Table 1, entry 10). The cyclopropyl group is known to activate the neighboring methylene group towards ruthenium tetraoxide oxidation.¹⁴

Table 1 Catalytic RuO₄ oxidation with NaIO₄ as a stochiometric oxidant in DMC–water solvent system

Entry	Starting material	Equiv. of NaIO4	Reaction time/h	Product (yield,^a %)
a Isolated yield.b RuO2·xH2O (4.5 mol%) was used as the ruthenium source.c GLC yield.
1	Benzyl alcohol	4.1	3	Benzoic acid (86)
2	1-Octadecanol	4.1	2	Stearic acid (92)
3	Cyclododecanol	2.05	2.5	Cyclododecanone (87)
4	4-Methoxybenzaldehyde	2.05	2.5	4-Methoxybenzoic acid (86)
5	1-Iodooctadecane	4.1	4	Stearic acid (79)^b
6	1-Hexyne	8.2	1	Pentanoic acid (88)
7	Tetrahydrofuran	4.1	1.5	γ-Butyrolactone (48)^c
8	Hexadecylbenzene	16.4	22	Heptadecanoic acid (91)
9	1,2,3,4-Tetrahydronaphthalene	16.4	18	Adipic acid (88)
10	Cyclopropylphenylmethane	8.2	20	Cyclopropyl phenyl ketone (56),^c cyclopropylacetic acid (9)^c

---

We used 2.05–16.4 equivalents of NaIO₄, for the reaction on 1–2 mmol of substrate (Table 1). The solvent was a mixture of 10 mL of DMC (5 mL of DMC was added initially to the reaction mixture and remaining 5 mL was used to dissolve the substrate prior to addition) and 5 mL of water. Vigorous stirring that ensured mixing of the two layers was essential. If the stirring was not vigorous enough, ruthenium compounds precipitated, as was indicated by black color of the solution, and the oxidation stopped. In such a system it was enough to increase the rate of stirring for oxidation to resume (yellow or red color of ruthenium tetraoxide returned within seconds). In contrast to ruthenium tetraoxide oxidations prepared in the presence of acetonitrile where a chlorinated solvent was necessary, addition of a chlorinated solvent such as dichloromethane or carbon tetrachloride was detrimental to the oxidation in DMC–water. Thus, addition of 0.184 g (1 mmol) of cyclododecanol dissolved in 5 mL of dichloromethane to the reaction mixture containing 5 mL of water, 5 mL of dimethylcarbonate, 5 mg of hydrated ruthenium tetraoxide and 0.877 g (4.1 mmol) sodium periodate resulted in precipitation of ruthenium species and termination of the reaction. Addition of either ethyl acetate or acetone, two other solvents commonly used in ruthenium tetraoxide oxidations, was also detrimental to the oxidation as evidenced by a reduced reaction rate. Addition of acetonitrile to the DMC–water solvent mixture was not detrimental to the ruthenium tetraoxide oxidation. However, it did not appear to provide any benefits either. The reaction rate and isolated yields of the oxidation products were similar in both cases.

Ruthenium tetraoxide oxidations with bleach (6% aqueous solution of NaClO) as a stochiometric oxidant gave good to excellent results in oxidation of alcohols, aldehydes, ethers and aromatic rings (Table 2). We obtained the best results by adding a solution of bleach to the reaction mixture in small portions over a period of time. Bleach was particularly suitable in degradation of aromatic rings. The oxidation was fast (1 h) compared to the RuCl₃–NaIO₄ oxidation system (18–22 h) and proceeded in a high yield. This method was successful in oxidative degradation of 2,4,6-trichlorophenol (Table 2, entry 9), a compound that has been identified as a persistent chlorinated pollutant. A drawback of the RuCl₃–NaClO oxidation system is that oxidation of aromatic substrates in which the benzene ring should be preserved proceeded in somewhat lower yields. Thus, oxidation of 4-methoxybenzaldehyde (Table 2, entry 4) proceeded in only 50% yield due to competing degradation of benzene ring. Even oxidation of 4-nitrobenzaldehyde (Table 2, entry 5) proceeded in only 72% yield. Attempts to improve the reaction by reducing reaction times or amounts of the cooxidant failed to improve the yield (Table 2, entry 6). Oxidation of tetrahydrofuran to γ-butyrolactone with RuCl₃–NaClO oxidation system proceded in 69% yield (Table 2, entry 7) and was superior to both RuCl₃–NaIO₄ (Table 1, entry 7) and RuCl₃–Oxone^® (Table 3, entry 5). It was recently reported that oxidation of tetrahydrofuran to γ-butyrolactone with RuCl₃–NaClO in ethyl acetate–water as solvent proceeded in only 44% yield.⁸ Further research on oxidation of ethers with RuCl₃–NaClO in DMC is in progress.

Table 2 Catalytic RuO₄ oxidation with NaClO as a stochiometric oxidant in DMC–water solvent system

Entry	Starting material	Equiv. of NaClO	Reaction time/h	Product (yield,^a %)
a Isolated yield.b GLC yield.
1	Benzyl alcohol	6	1	Benzoic acid (61)
2	1-Octadecanol	6	1	Stearic acid (81)
3	Cyclododecanol	3	1	Cyclododecanone (95)
4	4-Methoxybenzaldehyde	3	1	4-Methoxybenzoic acid (50)
5	4-Nitrobenzaldehyde	3	1	4-Nitrobenzoic acid (72)
6	4-Nitrobenzaldehyde	2.6	0.33	4-Nitrobenzoic acid (46)
7	Tetrahydrofuran	3.33	1	γ-Butyrolactone (69)^b
8	Hexadecylbenzene	20	1	Heptadecanoic acid (86)
9	2,4,6-Trichlorophenol	20	24	—

---

Table 3 Catalytic RuO₄ oxidation with Oxone^® as a stochiometric oxidant in DMC–water solvent system

Entry	Starting material	Equiv. of Oxone^®	Reaction time/h	Product (yield,^a %)
a Isolated yield.b GLC yield.
1	Benzyl alcohol	8	2	Benzoic acid (72)
2	Cyclododecanol	4	1	Cyclododecanone (94)
3	4-Methoxybenzaldehyde	7	25	4-Methoxybenzoic acid (69)
5	Tetrahydrofuran	6	4	γ-Butyrolactone (35)^b
6	Hexadecylbenzene	20	1	Heptadecanoic acid (88)

---

With Oxone^® as a cooxidant, most of the oxidation products were obtained in good yields (Table 3). However, often a somewhat larger excess of reagent was required, compared to RuCl₃–NaClO and RuCl₃–NaIO₄ systems. A lower amount of the reagent was needed to complete the oxidation when an aqueous solution of Oxone^® was added to the reaction mixture in small portions over a period of time.

The largest difference among the three stochiometric oxidants was encountered in oxidation of alkenes (Table 4). As expected, the RuCl₃–NaIO₄ system oxidized alkenes to the corresponding dicarboxylic acids in high yields (Table 4, entries 1 and 2). In order to avoid deactivation of the catalyst, a somewhat larger amount of DMC (20 mL instead of the usual 10 mL) was used. The RuCl₃–NaClO system in DMC failed to oxidize alkenes. The reaction was rapid and the alkenes were consumed quickly. However, only low yields of the corresponding diacids were isolated (Table 4, entries 3 and 5). Upon addition of bleach to the reaction mixture containing an alkene, a ruthenium catalyst and DMC, instead of a yellow–orange color of ruthenium tetraoxide, a black solution with greenish tint, reminiscent of the color of a deactivated ruthenium catalyst, was obtained. We speculated that, since oxidation of alkenes proceeds via the intermediate vicinal diols,¹⁵ the diols were water soluble and in the aqueous phase were not oxidized further by ruthenium tetraoxide, which was dissolved in the DMC phase. The fact that oxidation of alkenes proceeded in high yields when sodium periodate, which is capable of cleaving vicinal diols, was used as a stochiometric oxidant, and that yields were somewhat lower with Oxone^® (Table 4, entries 8 and 9), which does not oxidize diols in the aqueous phase, supports this hypothesis. To confirm the hypothesis, we run a RuCl₃–NaClO oxidation of an alkene (cyclooctene or cyclododecanone) for 1 h, separated the aqueous phase, extracted the aqueous phase with DMC in order to remove any ruthenium compounds that may have remained in it, and treated it with an aqueous solution of sodium periodate. Suberic and dodecanedioic acids were isolated in 43 and 15% yields, respectively (Table 4, entries 4 and 6). The results indicate that, at least in the case of lower molecular weight diols, the solubility of a diol in the aqueous phase presents a problem. Since the color of the solution indicated that the ruthenium catalyst was deactivated in the course of RuCl₃–NaClO oxidation of an alkene, we attempted to reactivate it. Addition of a larger amount of DMC to the reaction mixture did not provide an improvement. However, addition of 10 mL of acetonitrile to the reaction mixture resulted in a rapid color change from black-green to yellow. Thus RuCl₃–NaClO, in a solvent system consisting of DMC–acetonitrile (1∶1 by volume), oxidized cyclododecanone to decanedioic acid in a good yield (Table 4, entry 7). It is also possible to perform the oxidation in acetonitrile only (without DMC). However, in such a case a larger amount of acetonitrile was needed (50 mL).

Table 4 Catalytic RuO₄ oxidation of alkenes

Entry	Starting material	Stochiometric oxidant/eqiuv.	Reaction time/h	Product (yield,^a %)
a Isolated yield.b Solvent system consisted of 20 mL of DMC and 10 mL of water.c Acid was obtained by NaIO4 oxidation of the aqueous phase.d DMC (10 mL) and acetonitrile (10 ml) were used as the reaction solvent.
1	Cyclooctene	NaIO4 (8.2)	4	Suberic acid (91)^b
2	Cyclododecene	NaIO4 (8.2)	4	Dodecanedioic acid (87)^b
3	Cyclooctene	Bleach (10)	2	Suberic acid (2)
4	Cyclooctene	Bleach (10)	1	Suberic acid (43)^c
5	Cyclododecene	Bleach (10)	2	Dodecanedioic acid (12)
6	Cyclododecene	Bleach (10)	1	Dodecanedioic acid (15)^c
7	Cyclododecene	Bleach (10)	1	Dodecanedioic acid (73)^d
8	Cyclooctene	Oxone^® (10)	6	Suberic acid (44)
9	Cyclododecene	Oxone^® (10)	4.5	Dodecanedioic acid (63)

---

The described method represents a simpler, environmentally friendly, solvent system for a ruthenium tetraoxide oxidation of a wide range of substrates. Furthermore, together with bleach as a cooxidant, it may be have an application in ruthenium tetraoxide oxidative degradation of pollutants such as chlorophenols as well as polychlorinated biphenyls.

Experimental

The isolated products were identified by comparison with commercially available authentic samples. Compounds analyzed by GLC were identified by coinjection with authentic samples using a HP 5890 Gas Chromatograph equipped with a 0.53 mm × 30 m RTX-35 (Restek Corp.) column and FID.

Oxidation with sodium periodate as a stochiometric oxidant

In a typical experiment, 0.877 g (4.1 mmol, 2.05 equiv.) of sodium peroidate was dissolved in 5 mL of water. Hydrated ruthenium trichloride (5 mg, 1.2 mol%) was added followed by 5 mL dimethyl carbonate. A solution of cyclododecanol (369 mg, 2 mmol, 1 equiv.) in 5 mL dimethyl carbonate was added and the reaction mixture was stirred vigorously for 1 h. The layers were separated, and aqueous layer was extracted with ethyl acetate. Organic phases were combined, the excess of ruthenium tetraoxide was destroyed by addition of 2 mL of diethyl ether, and the solution was dried over anhydrous MgSO₄. The solution was filtered through a short column of Celite and evaporated to give 335.1 mg (92%) of cyclododecanone.

Oxidation with bleach as a stochiometric oxidant

In a typical experiment, to hydrated ruthenium trichloride (5 mg, 2.5 mol%) was added 10 mL dimethyl carbonate followed by hexadecylbenzene (355 μL, 1 mmol, 1 equiv.). Bleach was added in small portions over the course of the reaction. The color of the reaction mixture became black upon addition of bleach and, after a few seconds, turned yellow. An exothermic reaction ensued and the temperature of the solution rose to 45 °C. The reaction mixture was stirred for 1 h. The pH of the reaction mixture was adjusted to 2 by addition of 10 M hydrochloric acid. The layers were separated and the aqueous layer was extracted with diethyl ether. Organic phases were combined and the solution was dried over anhydrous MgSO₄. The solution was filtered through a short column of Celite and evaporated. Recrystallization (water–methanol) of the crude product gave 251.3 mg (86%) of heptadecanoic acid.

Oxidation with Oxone^® as a stochiometric oxidant

In a typical experiment, to hydrated ruthenium trichloride (5 mg, 2.5 mol%) was added 10 mL dimethyl carbonate followed by cyclododecene (190 μL, 1 mmol, 1 equiv.). An aqueous solution of Oxone^® (1 mmol or 0.615 g per 3 mL water) was added in small portions over the course of the reaction until a total of 10 mmol were added. The reaction mixture was stirred vigorously for 4.5 h. The layers were separated and aqueous layer was extracted with ethyl acetate. Organic phases were combined, the excess of ruthenium tetraoxide was destroyed by addition of 2 mL of diethyl ether, and the solution was dried over anhydrous MgSO₄. The solution was filtered through a short column of Celite and evaporated. Recrystallization (ethanol–water) of the crude product gave 208.5 mg (63%) of dodecanedioic acid.

Acknowledgement

We thank Nova Southeastern University for support in form of President’s Faculty Scholarship Award (Grant no. 338 319) to V. D.

Notes and references

1. J. H. Carlsen, T. Katsuki, S. V. Martin and B. K. Sharpless, J. Org. Chem., 1981, 46, 3936–3938.
2. A. E. M. Boelrijk and J. Reedijk, J. Mol. Cat., 1994, 89, 63–76.
3. EPA web site: http://www.epa.gov/ozone/title6/phaseout/accfact.html. Accessed on July 15, 2002.
4. F. A. Khan, B. Prabhudas, J. Dash and N. Sahu, J. Am. Chem. Soc., 2000, 122, 9558–9559.
5. V. Piccialli, Tetrahedron Lett., 2000, 41, 3731–3733; V. Piccialli and N. Cavallo, Tetrahedron Lett., 2001, 42, 4695–4699.
6. F. Rivetti, C. R. Acad. Sci. Paris, Sér. IIC, Chimie/Chemistry, 2000, 3, 497–503; Y. Ono, Appl. Catal. A: General, 1997, 155, 133–166.
7. F. G. Oberender and J. A. Dixon, J. Org. Chem., 1959, 24, 1226–1229.
8. S. Wolfe, S. K. Hasan and J. R. Campbell, Chem. Commun., 1970, 1420–1421.
9. A. Tenaglia, E. Terranova and B. Waegell, Tetrahedron Lett., 1990, 31, 1157–1158.
10. Y. Yamamoto, H. Suzuki and Y. Moro-oka, Tetrahedron Lett., 1985, 26, 2107–2108.
11. A. B. Smith and R. M. Scarborough, Synth. Commun., 1980, 10, 205–211.
12. L. Gonsalvi, I. W. C. E. Arends and R. A. Sheldon, Chem. Commun., 2002, 202–203.
13. M. T. Nunez and V. S. Martin, J. Org. Chem., 1990, 55, 1928–1932.
14. J. L. Coudret, S. Zollner, B. J. Ravoo, L. Malara, C. Hanisch, K. Dorre, A. de Meijere and B. Waegell, Tetrahedron Lett., 1996, 37, 2425–2428; T. Hasegawa, H. Niwa and K. Yamada, Chem. Lett., 1985, 1385–1386; G. Hugel, L. Lods, J. M. Meltor and G. Ourisson, Bull. Soc. Chim. Fr., 1965, 2894–2902.
15. T. K. M. Shing, V. W.-F. Tai and E. K. W. Tam, Angew. Chem., Int. Ed. Engl., 1994, 33, 2312–2313.

---

Footnote

† One of the goals of our research was to develop a general method for oxidation of iodoalkanes. The only satisfactory solvent systems were DMC–H₂O and CCl₄–CH₃CN–H₂O. Other solvent systems resulted in extended reaction times and poor yields. Detailed results of oxidation of iodoalkanes will be published elsewhere.

---