Although the synthesis of y-butyrolactone from THF is not desirable for economic reasons, the process in general is of high value for pharmaceutical industry and fine chemicals when applied to substituted THFs and other cyclic ethers, depending on the need of the client.

General problems for this process are usually the use of expensive and toxic oxidants, low TONs, low selectivity and working at elevated temperatures (energy).

Since laws concerning emissions get ever more stringent, the development of new, highly efficient catalyst systems that promote the synthesis and modification of various increasingly sophisticated organic compounds is of paramount importance.

The presented process satisfies the urge for'green chemistry'by using cheap air oxygen as the terminal oxidant (avoiding toxic and expensive organometallic oxidants or peroxides) in a catatalytic process with unlimited catalyst lifetime and plain water as the side product. Highly functionalised products will be available with the appropriate starting materials.

The selective oxidation of ethers to their corresponding esters or lactones in presence of transition metal compounds is an important reaction and is well documented.

Among the oxidants used in stoichiometric amounts are metal oxides such as CrO3, ZnCr207, KMn04, Zn (Mn04) 2, Mn207, and Ru04 [4]. Together with terminal oxidants such as Na104, LiCIO or organic peroxides, catalytic amounts of Ru04 [5], RuCI3 [6], RuC12 [1, 3-bis (diphenylphosphino) propane] 2 [7] or of the type [Rul" LX) (LY) Cl] [8], promote this transformation. For more than a decade now, much research has been directed to employing molecular oxygen as a cheap, safe and clean, i. e. an environmentally more attractive, 'green', alternative to the above mentioned oxidants [9]. For example, CoCl2 [10], Co (acac) [11], Pt [12], and dinuclear Cu [13] systems have been established in the oxygenation of THF with molecular oxygen, usually working at elevated temperatures. In presence of carbon monoxide [14] or carbon dioxide [15], this process is promoted by PdCfs/CuOs [14], Fez) 2 [15a], or Rh complexes [15a, b]. Recently, a whole array of mono nuclear, late transition metal complexes have been reported to catalyse the selective oxidation of THF, yielding y-butyrolactone as well as small quantities of 2- hydroxytetrahydrofuran and 4-hydroxybutanal [3].

The invention works as follows : In a typical catalytic run, 0.1 mmol of the catalyst precursor 1,2 or 3 is dissolved in 30 ml of dry THF and stirred under an oxygen atmosphere. A plain air atmosphere leads to similar results. The reaction is monitored by IR spectroscopy or GCMS, and the amounts of y- butyrolactone formed are derived from reference calibration. Fractional distillation of the reaction mixtures allows isolation the pure product y- butyrolactone.

Dinuclear 1 (new), 2 (known compound, [1] ) and 3 (known compound, [16]) were used as catalyst precursors for the first time in this process. The highest selectivity for butyrolactone was obtained with 3. The catalysts allow (i) to replace stoichiometric, toxic co-oxidants by cheap,'green'molecular oxygen (air), (ii) a selectivity of up to 70%, (iii) isolation of the lactone by simple distillation, (iv) working at room temperature, and (v) overall formulation of a green chemistry which is applicable to cyclic ethers: 0 0 0 cat., ox. >+ others \/-H20 \ A possible disadvantage is the lack of knowledge of the actual active catalytic species and the catalytic cycle. Investigations regarding this are underway.

The invention is described and understood with the following examples.

Example 1 Synthesis of Ind (CO) 3Mo-Ru (CO) 2Cp (1) 1.5 ml (2.4 mmol) of a 1.6 M solution of n-BuLi in hexanes are added dropwise to 300, u1 (2.6 mmol) of freshly distilled indene in 20 ml THF at-78°C. After stirring for 2 h at RT, 528 mg (2.0 mmol) Mo (CO) 6 are added and the mixture is refluxed for 18 h until the formation of NaMo (CO) 31nd is complete (IR). A solution of 604 mg (2.0 mmol) CpRu (CO) 2Br in 20 ml THF is added dropwise at RT, the mixture refluxed until the starting materials are consumed (25 h, IR) and subsequently concentrated and separated into its components by column chromatography. Fraction 1 (eluent : hexane/THF 25/1): 495 mg (0.96 mmol, 48 %) red-orange cubes of Ind (CO) 3Mo-Ru (CO) 2Cp (1) (from hexane/THF at- 25° C). IR (hexane) [cm~] : 2023 (vw), 2008 (vw), 1963 (vs), 1955 (vs), 1922 (vw), 1902 (w), 1884 (m), IR (THF) [cm-1] : 2015 (vw), 2006 (vw), 1953 (vs), 1988 (sh), 1873 (m) [1].'H-NMR (C6D6) [ppm]: 7.29 (m, 2 H, Ind, H4, 7), 6.75 (m, 2 H, Ind, H5, 6), 5.44 (d, 2 H, 3J = 6. 0 Hz, Ind, H1, 3), 5.05 (t, 1 H, 3J = 6. 0 Hz, Ind, H2), 4.48 s, 5 H, RuCp). 13C-NMR [ppm]: 219.4 (s, CO), 124.1 (d,'JcH = 160 Hz, Ind, C5, 6), 123.4 (d, 1JCH =165 Hz, Ind, C4, 7), 109. 4 (s, Ind, C7a, 7b), 91.8 (d, 1JCH =175 Hz, Ind, C2), 87.5 (d, 1JCH = 179 Hz, RuCp), 89.7 (d, 1JCH= 177 Hz, Ind, C1, 3). Anal. Calcd. for C19H1205MoRu (FW = 517. 3108): C, 44. 11 ; H, 2.34. Found C, 44.81 ; H, 2.43.

Catalytic runs In a typical catalytic run, 0.1 mmol of the catalyst precursor was dissolved in 30 ml of dry THF and stirred under atmospheres of Os, Os/CO (1 : 1) and 02/C02 (1 : 1) respectively, which were supplied by balloon. The control runs under Ar, CO and C02 were carried out in sealed Schlenk flasks under a pressure of 0. 5 bar. The reactions were monitored by IR spectroscopy and the amounts of y-butyrolactone formed were derived from reference calibration (area for the v-CO band at 1780 cm~1). A rough quantitative approximation of the sum of all other products was obtained by reference to propylformate calibration (area for the v-CO band at 1726 cm-', relevant for all formed CO compounds other than y-butyrolactone). Fractional distillation of the reaction mixture of a typical 3 + Os run allowed us to isolate the pure products y- butyrolactone, succinic acid and propylformate as well as to quantify the sum of all other products. The formate acetal 6 [10] and others were detected in the mother solution and mixed fractions (NMR) but were not isolated. The organic compounds were identified by comparison of spectral data with literature results and/or authentic samples.

The Results As expected, upon exposure to oxygen, the orange solutions of 1 and 2 as well as the yellow solution of 3, turned black with time and a black precipitate was formed, reflecting the oxidative decomposition of the starting materials.

Organic product formation was observed from the start of the reaction. It did not alter while and after 1-3 were oxidised (IR control). The red solution of 4 turned yellow, indicating the formation of oxo-complexes [19].

Table A. Oxidation of THF under various conditions Catalyst atmospher TON (7 days) TON (7) precursor e 0, o \ (15 (7) ~ sCHO HOOC CCON other than 7, (15 days) Ind (CO) 3Mo-Ru (CO) 2Cp (1) 02 55 100 Cp (CO) 3Mo-Ru (CO) 2CP (2) 02 49 100 80 [Ru (CO) 2Cp] z (3) 2 100 77 155 [Ru (CO) 2CP] 2 (3) 02 290 48 12 120 290 [Ru (CO) 2CP] 2 (3) Oz/COzl/l 80 70 80 [Ru (CO) 2CP] 2 (3) 02/CO 1/1 75 51 75 [Ru (CO) 2Cp] 2 (3) C02--0 [Ru (CO) zCp] 2 (3) CO--0 Cp (CO) 3Mo-Mo (CO) 3Cp (4) 02 5 6 5 RuCI3 (5) °2 30 50 61 Since in control runs of THF/02 as well as THF solutions of 1,2, or 3, under Ar, CO or C02 atmospheres respectively, the formation of organic products could not be detected, it is evident that the combination of THF, a metal complex and oxygen is crucial to perform the oxidation reaction. A comparison of the TONs of catalytic runs with 1-5, reveals that the dinuclear ruthenium complex 3 is the most active species of the series, with a TON of about 180 after 7 days (table 1).

In contrast, only a few turnovers were detected when the molybdenum compound 4 was used. The MoRu complexes 1 and 2 yield TONs about 150 and the TON of RuC13 (5) is about half of that. In either case, the catalyst could be recovered and used repeatedly. In contrast to reports for Pd/Cu [14] and Fe, Rh [15] systems, 02/CO and 02/C02 atmospheres did not enhance, but decrease the catalytic activity of 3, likely as a result of the lower partial oxygen pressure in the system.

These trends are clearly reflected by the rate of formation of y-butyrolactone (figure 1). Under oxygen atmosphere, 3 was the most active promoter from the start of the reaction, while all other catalyst/atmosphere combinations showed lower activity which seems to be dependent mainly on the concentration of active ruthenium species. However, a qualitative comparison of the selectivities towards y-butyrolactone shows that 3 yields a significantly higher selectivity (up to 70%) than the other complexes of the series. This shows that the nature of the catalyst precursor is a dominant factor and suggests that different active species are formed upon their oxidation. The exceptionally low activity of RuCl3 can partly be explained with an induction period (figure 1), as described for a K2PtCI4/D20 system [12]. For the compounds presented here, the induction, i. e. the decomposition, of 1-3 proceeds faster than the formation of the active catalyst from RuCI3. In addition, the solubility of the catalyst influences its performance.

References [1] T. Straub, M. Haukka, T. A. Pakkanen, J. Organomet. Chem. 612 (2000) 106.

[2] P. Pinto, M.-J. Calhorda, T. Straub, V. Miikkulainen, J. Raty, M. Suvanto, T. A. Pakkanen, J. Mol. CatalA., 2001, 170, 209-218.

[3] M. Shi, J. Chem. Research (S) 1998,592.

[4] (a) J. T. Harrison, S. Harrison, Chem. Commun. 1966,752 ; (b) H.

Firouzabadi, A. R. Sardaria, H. Moosavipour, G. M. Afshari, Synthesis 1986, 285; (c) H. T. Schimdt, H. J. Schaefer, Angew. Chem., Int. Ed. Engl. 18 (1979) 69; (d) S. Wolfe, C. F. Ingold, J. Am. Chem. Soc. 105 (1983), 7755; (e) M. Trommel, M. Russ, Angew. Chem. 99 (1987) 1037; (f) L. M. Berkowitz, P. N. Rylander, J. Am. Chem. Soc. 80, (1958), 6682.

[5] A. B. Smith III, R. M. Scarborough, Jr. , Synth. Commun. 10 (1980) 205.

[6] (a) P. H. J. Carlsen, T. Katsuki, V. S. Martin, K. B. Sharpless, J. Org. Chem.

46 (1981) 3936. (b) G. Balavoine, C. Eskenazi, F. Meunier, J. Mol. Catal. 30 (1985) 125.

[7] (a) M. Bressan, A. Morvillo, J. Chem. Soc. , Chem. Commun. 1989,421. (b) M. Bressan, A. Morvillo, G. Romanello, Inorg. Chem. 29 (1990) 2976.

[8] (a) D. Chatterjee, A. Mitra, S. Mukherjee, Polyhedron 18 (1999) 2659; (b) D.

Chatterjee, A. Mitra, B. C. Roy, React. Kinet. Catal. Lett. 70 (2000) 147; (c) D. Chatterjee, A. Mitra, B. C. Roy, J. Mol. Catal. A. 161 (2000) 17.

[9] T. Mukaiyama, T. Yamada, Bull. Chem. Soc. Jpn. 68 (1995), 17, and references therein.

[10] P. Li, H. Alper, J. Mol. Catal. 72 (1992) 143.

[11] (a) E. Hata, T. Takai, T. Mukaiyama, Chem. Lett. 1993,1513 ; (b) M. T.

Reetz, K. Töllner, Tetrahedron Lett. 36 (1995) 9461.

[12] (a) A. Sen, M. Lin, L. -C. Kao, A. C. Hutson, J. Am. Chem. Soc. 114 (1992) 6385; (b) M. Sommovigo, H. Alper, J. Mol. Catal. 88 (1994) 151; (c) K.

Heyns, H. Buchholz, Chem. Ber. , 109 (1976) 3707; (d) T. -J. Wang, Z.-H.<BR> <P>Ma, M. -Y. Huang, Y. -Y. Jiang, Polym. Adv. Technol., 7 (1996) 88.

[13] S. Minakata, E. Imai, Y. Ohshima, K. Inaki, l. Ryu, M. Komatsu, Y. Ohshiro, Chem. Lett. 1996,19.

[14] A. R. EI'man, E. V. Slivinskii, S. M. Loktev, lzv. Akad. Nauk SSSR, Ser. <BR> <BR> <P>Khim. , 9 (1988) 2188.<BR> <P>[15] (a) M. Aresta, C. Fragale, E. Quaranta, I. Tommasi, J. Chem. Soc. , Chem.<BR> <P>Commun. , 1992,315 ; (b) A. K. Fazlur-Rahman, J. -C. Tsai, K. M. Nicholas,<BR> J. Chem. Soc. , Chem. Commun. , 1992,1334.

[16] D. H. Gibson, W.-L. Hsu, A. L. Steinmetz, B. V. Johnson, J. Organomet.

Chem. , 208 (1981) 89.

[17] H. Nagashima, K. Mukai, Y. Shiota, K. Yamaguchi, K. Ara, T. Fukahori, H.

Suzuki, M. Akita, Y. Moro-oka, K. Itoh, Organometallics 9 (1990) 799.

[18] T. Straub, A. M. P. Koskinen, OMCOS 11, Taipei, Taiwan, 2001,268.

[19] (a) M. Cousins, M. L. H. Green, J. Chem. Soc. 1964,1567 ; (b) M.

Cousins, M. L. H. Green, J. Chem. Soc. 1969,16 ; (c) K. Isobe, S.

Kimura, Y. Nakamura, J. Organomet. Chem. 331 (1987) 221.