Photocatalytic properties of iron porphyrins revisited in aqueous micellar environment: oxygenation of alkenes and reductive degradation of carbon tetrachloride

Abstract

N,N-Dimethyltetradecylamine N-oxide (DTAO) is an appropriate surfactant to form micelles able to host the iron(III) meso-tetrakis(2,6-dichlorophenyl)porphyrin [Fe(III)(TDCPP)]. The so obtained microheterogeneous catalyst can induce biomimetic redox processes on organic substrates in aqueous medium, using sunlight and oxygen as clean reagents. The primary photochemical process consists in the photoinduced reduction of Fe(III) to Fe(II) with the contemporaneous oxidation of the axial ligand to radical species. The micelle environment may control some main parameters affecting the reactivity of these intermediates and, therefore, the chemoselectivity of the hydrocarbon oxidation processes. In contrast to what is observed in homogeneous organic solution, both cyclohexene and cyclooctene can be oxidised to the corresponding epoxides, with a selectivity higher than 90% in the case of cyclooctene. On the other hand, the main oxidation product of cyclohexene is cyclohex-2-en-1-one as expected in a hydrophobic micellar environment. The Fe(III)(TDCPP)/DTAO photocatalyst is very promising also in view of obtaining catalytic systems capable of converting small amounts of toxic halogenated alkanes present in water into less dangerous products. In particular, CCl₄ can be reduced by ethanol or cyclohexanol with high quantum yields (>10⁻¹), with good conversion (ca. 75%) and turnover values (>1500).

---

Green Context

There is an increasing trend towards the use of water as a medium for carrying out organic transformations. There are obvious advantages to the environment through the avoidance of volatile organic compounds and process costs can be reduced since highly effective solvent recovery steps are unnecessary. However, the incompatibility of water with many organic compounds hinders this approach and additives may often be necessary to enable effective reactions. Here the use of a surfactant to form micelles with a highly useful iron porphyrin catalyst is described. By doing this the authors demonstrate a very interesting new oxidation system that can for example, give high yields of epoxides from the direct reaction of alkenes with oxygen using surfactant–iron porphyrin catalysts and irradiation only.

JHC

Introduction

The use of water as an alternative reaction medium, instead of toxic and expensive organic compounds, is a very attractive challenge to reduce the environmental impact of chemical processes. In this framework, a possible approach is the use of appropriate, water soluble, surfactants able to form micelles with hydrophobic cores. Herein, this strategy is followed in order to induce biomimetic redox processes based on the use of iron porphyrins catalysts and of clean reagents such as molecular oxygen and sunlight.

In the last two decades, the catalytic activity of iron porphyrin complexes has attracted the attention of many researchers in view to build up model systems of cytochrome P-450 oxygenases, accounting for both monooxygenation of organic substrates^1–6 and reduction of halogenated compounds.^7–12 These models are investigated for two main reasons: (i) better understanding of the essential steps of the enzyme mechanism in living organisms; (ii) the growing demand of both fine and industrial chemistry for new active catalysts, working with high efficiency and selectivity under mild temperature and pressure conditions.

There is a similarity between the photochemical technique and other more conventional methods of activation of oxygenase model systems based on the use of chemical reagents. In particular, it has been demonstrated that the biomimetic catalytic activity of Fe(III) porphyrins may originate by intramolecular photoreactions at suitable wavelengths (λ = 300–400 nm) which lead to the reduction of the metal center as well as to the formation of reactive radical species. These primary photoproducts can induce the subsequent reductive activation of O₂ to give hydrocarbon oxidation^13–19 as well as the reductive degradation of carbon tetrachloride.^20,21

In this work we use N,N-dimethyltetradecylamine N-oxide as surfactant [(C₁₄H₂₉)Me₂NO, DTAO] to create water soluble micelles where the synthetic iron(III) meso-tetrakis(2,6-dichlorophenyl)porphyrin chloride [Fe(III)(TDCPP)Cl] and different organic substrates may reside. An interesting peculiarity of this iron porphyrin is that the chlorine atoms in its meso-aryl groups provide a steric protection of the porphyrin ring against its radical induced oxidative degradation.²²

Results and discussion

Photoreduction of the DTAO/Fe(III)(TDCPP) system under anaerobic conditions

Aqueous mixtures of DTAO (5 × 10⁻² M) and Fe(III)(TDCPP) (2 × 10⁻⁵ M) containing pyridine (4 × 10⁻² M) were irradiated (λ > 350 nm) under anaerobic conditions. The UV–VIS spectral variations displayed in Fig. 1 showed that the Soret band was progressively red shifted while the growth of a band at 540 nm occurred. Addition of the reducing agent, sodium dithionite, reinforced the spectral variations seen under irradiation. These results are analogous to those previously obtained irradiating iron porphyrin complexes in homogeneous solution containing pyridine²³ and are a clear indication that Fe(III)(TDCPP) inside the micelle is photochemically reduced to Fe(II)[TDCPP(py)₂] (bis-pyridine hemochrome). There is a major difference between the photochemical behaviour of Fe(III)(TDCPP) inside DTAO micelles and in homogeneous organic solution, where we found that the iron(II) porphyrin could be accumulated also in the absence of pyridine.²⁰

Fig. 1 UV–VIS spectral variations of a de-aerated aqueous mixture of Fe(III)(TDCPP) (2 × 10⁻⁵ M)/DTAO (5 × 10⁻² M) containing pyridine (4 × 10⁻² M) irradiated with λ > 350 nm. Dotted lines: during irradiation; solid line: after addition of sodium dithionite.

A number of workers has demonstrated the ability of non-emitting iron porphyrins to undergo intramolecular redox photochemistry according to the first equilibrium of Scheme 1.^{14,20,21,23,24} Irradiation in axial-ligand-to-metal charge transfer bands in the near-UV can bring about the reduction of the metal centre with the concomitant oxidation of the axial ligand to a radical species. In the system investigated here the axial ligand Cl⁻ may be oxidised to the Cl^• radical. The efficiency of this process depends on: (i) the efficiency of the charge separation that occurs in competition with radiationless deactivation of the excited state; (ii) the possibility for the radical to diffuse away from the first coordination sphere of the metal; (iii) the presence of species able to trap either Fe(II) porphyrin or the oxidised ligand. The results reported indicate that the diffusion of the primary photoproducts is strongly inhibited when the iron porphyrin is confined inside the micelle. Therefore, we observe reduction of Fe(III)(TDCPP) only in the presence of a large excess of pyridine, which effectively reacts with the iron(II) porphyrin to give the relatively stable bis-pyridine hemochrome Fe(II)[TDCPP(py)₂].

Scheme 1 Catalytic cycle for the monooxygenation of cyclohexene by photoexcited Fe(III)(TDCPP) in DTAO micelles.

Photooxygenation of cycloalkenes

Water solutions of DTAO (5 × 10⁻² M) and Fe(III)(TDCPP) (2 × 10⁻⁵ M) were able to dissolve considerable amounts of cyclohexene or cyclooctene (5 × 10⁻² M). Irradiation of the so-obtained mixtures (λ> 350 nm) in the presence of oxygen led to the oxidation of the two cycloalkenes to give the products reported in Table 1, which also shows the ratio between the total substrate oxidised and the Fe(III)(TDCPP) destroyed during the photochemical experiment (total turnover).

Table 1 Photooxygenation of alkenes^a

Substrate	Product	Conc.^b/μM	Total turnover^c
a Samples of 3 ml of aqueous solutions containing Fe(III)(TDCPP) (2 × 10^−5 M), DTAO (5 × 10^−2 M) and the alkene (5 × 10^−2 M) were irradiated for 4 h under 1 atm 100% O2 at 22 ± 1 °C (λ > 350 nm).b Relative error on reported values is 20%.c Mol of total oxidised substrate/mol of consumed Fe(III)(TDCPP).
Cyclohexene	Cyclohex-2-en-1-one	77
Cyclohex-2-en-1-ol	23	34
Cyclohexene oxide	25
Cyclooctene	Cyclooctene oxide	40	40

---

The nature and distribution of the products obtained from the oxidation of cyclohexene may be explained on the basis of previous work carried out in homogeneous organic solutions.^13,16,18,19 As shown in Scheme 1, the photogenerated Cl^• radicals may easily abstract the allylic hydrogen atom from cyclohexene to give the resonance-stabilised radical 3. The very fast reaction of O₂ with Fe(II)(TDCPP) in the presence of cyclohexenyl radicals should lead to the formation of the Fe(III)porphyrin–peroxide complex 5, which, in non-polar environments, is known to undergo an intramolecular decomposition, giving the starting Fe(III) porphyrin–hydroxy complex 6 and the corresponding ketone.²⁵ Since cyclohex-2-en-1-one represents about 60% of the overall oxidised cyclohexene, we infer that the photocatalytic process occurs to a significant level in the non-polar region of the micelles. In the proximity of the polar interface, 5 should undergo a heterolytic cleavage of the O–O bond, with formation of cyclohex-2-en-1-ol as well as of the high-valence iron–oxo–porphyrin complex 7, which is able to insert directly its oxygen atom in double bonds to yield cyclohexene oxide, regenerating the initial complex 1.^1,26–28 This statement is confirmed by the observation that cyclohex-2-en-1-ol and cyclohexene oxide are obtained in a 1∶1 molar ratio. It is of note that some of the reactive iron complex intermediates reported in Scheme 1 resemble those formed during the catalytic cycle of cytochrome P-450, especially the high valence iron–oxo species 7.^1–4 It is possible that at the end of the cycle, the starting axial ligand Cl⁻ is replaced by OH⁻. This axial ligand, however, is expected to undergo photooxidation to give OH^• radicals able to induce the formation of 3 as well as Cl^•.

The above results confirm our previous finding indicating that the control of the reaction environment through the employment of heterogeneous or organized systems can direct the selectivity of hydrocarbon photooxidation process.^13,19,29 In particular, proper conditions have been found here that are connected to the unusual reaction environment inside the DTAO micelles. Therefore, at variance with the above behaviour, photooxidation of cyclohexene by Fe(III)(TDCPP) dissolved in ethanol–cyclohexene (3∶2, v/v) mixed solvent led to the formation of a mixture of various oxygenation products which, however, did not include the epoxide.^13,30

When the DTAO/Fe(III)(TDCPP) system was photoexcited in the presence of cyclooctene, the only oxidation product was the corresponding epoxide (Table 1). These findings are in line with the fact that, in contrast to cyclohexene, cyclooctene is not able to form resonance-stabilised radicals analogous to 3 because ring tensions prevent a coplanar disposition of C[double bond, length half m-dash]C and a neighbouring radical carbon.³¹ Therefore, extraction of the allylic hydrogen atom by Cl^• is not favoured and allylic oxidation products are not formed. In accord with previous work concerning the oxidation of cyclooctene by peroxy radicals,³¹ we propose that this alkene gave large proportions of epoxide in the experiment described here via addition of the iron peroxo intermediate 4 to its double bond.

Photoreduction of carbon tetrachloride

The ability of photoexcited Fe(III)(TDCPP) to induce the reduction of CCl₄ has been previously demonstrated.²⁰ This process occurs when the porphyrin complex is irradiated in the near-UV in the mixed solvent ethanol–CCl₄ (2∶3 v/v) according to eqn. (1)

CCl₄ + CH₃CH₂OH → CHCl₃ + HCl + CH₃CHO (1)

The proposed mechanism for eqn. (1) is summarised in Scheme 2. The primary photochemical act leads to the reduction of Fe(III) to Fe(II) and to the oxidation of the axial ligand L to L^•, which, in turn, oxidises CH₃CH₂OH to CH₃CH^•OH and H⁺. The iron(II) porphyrin, mimicking the catalytic cycle of cytochrome P450 under anaerobic conditions, is able to reduce the coordinated CCl₄ yielding Cl⁻ and ^•CCl₃.^7–12 The two radicals CH₃CH^•OH and ^•CCl₃ can initiate a chain process leading to the formation of CHCl₃, CH₃CHO and HCl with a quantum yield of ca. 3 × 10⁻².²⁰ As described in the following, the employment of DTAO–water micellar solution significantly increases the photochemical efficiency of Fe(III)(TDCPP) and allows the initial concentration of CCl₄ to be reduced and to use the water incompatible C₆H₁₁OH as reducing species.

Scheme 2 Catalytic cycle for the reductive degradation of CCl₄ by photoexcited Fe(III)(TDCPP) in DTAO micelles.

The aqueous system DTAO/Fe(III)(TDCPP) (5 × 10⁻² M/ 2 × 10⁻⁵ M) in the presence of ethanol (10% v/v) and CCl₄ (2 × 10⁻² M) was prepared as described in the Experimental section and irradiated with wavelengths >350 nm for 4 h. Table 2 shows that ethanol was oxidised to acetaldehyde while CCl₄ underwent a reduction process leading to the formation of Cl⁻ ion. Some experiments were also carried out with monochromatic light at 365 nm in order to obtain the quantum yield value for the photoreduction of CCl₄, given as the ratio of the number of mol of Cl⁻ ions to the number of mol of absorbed photons. We calculated a value of (3 ± 0.2) × 10⁻¹, which, interestingly, was one order of magnitude higher than that previously obtained in an analogous experiment carried out by dissolving the iron porphyrin in the mixed solvent ethanol–CCl₄ (2∶3 v/v).²⁰ Finally, the good stability of Fe(III)(TDCPP) inside DTAO micelles is demonstrated by a turnover value, calculated as mol of reduced CCl₄ per mol of consumed porphyrin, of ca. 2500.

Table 2 Photoreduction of carbon tetrachloride^a

Electron donor	Detected product	Conc./mM	Turnover^b
a Samples of 3 ml aqueous solutions containing Fe(III)(TDCPP) (2 × 10^−5 M), DTAO (5 × 10^−2 M), CCl4 (2 × 10^−2 M) and the electron donor were irradiated for 4 h at 22 ± 1 °C (λ > 350 nm).b Mol of product/mol of consumed Fe(III)(TDCPP).c Relative error on reported values is 5%.d Relative error on reported values is 20%.
Ethanol	Chloride anion	16^c	2500
(co-solvent, 10% v/v)	Acetaldehyde	11^d	1700
Cyclohexanol	Chloride anion	5.6^c	1600
(7 × 10^−2 M)	Cyclohexanone	2.4^d	700

---

Table 2 also reports the results obtained when cyclohexanol was used in the place of ethanol for the reduction of CCl₄. This alcohol, almost insoluble in water, must largely reside at the micellar interface. Therefore the microheterogeneous environment is able to maintain high concentrations of cyclohexanol in proximity to the metal centre also when it is dissolved in small amounts, comparable to those of carbon tetrachloride (7 × 10⁻² M and 2 × 10⁻² M, respectively).

Irradiation of the Fe(TDCPP)DTAO system in the presence of cyclohexanol and CCl₄ induced a catalytic cycle analogous to that already discussed for ethanol (Scheme 2), leading to the formation of cyclohexanone. The photoreduction of CCl₄ to Cl⁻ was characterized by a quantum yield of (1 ± 0.1) × 10⁻¹ at 365 nm and a turnover value of 1600. Table 2 shows that the amount of ketone produced was significantly less than that of chloride anion, indicating that CCl₄ may be reduced with the involvement of species other than cyclohexanol. Possible candidates are cyclohexanone and the surfactant aliphatic chains, which may be involved in the radical chain process initiated by ^•CCl₃ and L^•.

In order to investigate deeper into the application perspectives of the described photocatalytic system in water detoxification, we carried out some experiments decreasing the concentration of both DTAO and Fe(TDCPP) by on order of magnitude. Interestingly, irradiation (λ = 365 nm) of aqueous solutions of DTAO (5 × 10⁻³ M) and Fe(TDCPP) (2.5 × 10⁻⁶ M) containing cyclohexanol (1.5 × 10⁻² M) and CCl₄ (5 × 10⁻³ M) still led to the reduction of CCl₄ to Cl⁻ with a quantum yield of (2.5 ± 0.2) × 10⁻², which is the same as that previously obtained in a homogeneous solution of EtOH and CCl₄.

Conclusions and perspectives

We have demonstrated that DTAO micelles are able to host the catalyst Fe(III)(TDCPP) and several organic substrates. In this way, it is possible to realise biomimetic redox processes in aqueous media utilising sunlight and oxygen as clean reagents. Inside the micelles, irradiation induces the homolytic cleavage of the Fe(III)–axial ligand bond as well as the subsequent oxygenation of cyclohexene and cyclooctene. In contrast to what is observed in homogeneous organic solution, both cyclohexene and cyclooctene can be oxidised to the corresponding epoxides, with a selectivity >90% in the case of cyclooctene. On the other hand, the main oxidation product of cyclohexene is cyclohex-2-en-1-one, as expected in the hydrophobic micellar environment inside the micelle core. The Fe(III)(TDCPP)/DTAO photocatalytic system appears to be very promising also in view of obtaining catalytic systems capable of converting small amounts of toxic halogenated alkanes present in water into less dangerous products. In particular, CCl₄ can be reduced either by ethanol or cyclohexanol with high quantum yields (>10⁻¹), with good conversion (ca. 75% with ethanol) and turnover values (>1500).

Experimental

Materials

Iron(III) meso-tetrakis(2,6-dichlorophenyl)porphyrin chloride [Fe(III)(TDCPP)Cl] was kindly supplied by Drs D. Mansuy and P. Battioni of the Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, Ura 400 CNRS Université Paris V. All the solvents were spectroscopic grade products (Fluka). Cyclohexene, cyclooctene, cyclohexanol and carbon tetrachloride were reagent grade and were distilled before use.

N,N-Dimethyltetradecylamine N-oxide was prepared modifying a literature procedure;³² 0.1 mol of N,N-dimethyltetradecylamine (Fluka, purified by vacuum distillation) and 0.15 mol of hydrogen peroxide (35% w/w solution in water by Carlo Erba) were dissolved in 60 ml of ethanol and stirred (room temperature) in a flask for ca. 24 h. Addition of 10 mg of MnO₂ destroyed the excess hydrogen peroxide. The solution was filtered and evaporated under vacuum. The obtained crude solid was dissolved in benzene and dehydrated using a Dean–Stark apparatus; the benzene was evaporated under vacuum. The obtained solid was dissolved in methylene chloride, filtered off and evaporated under vacuum. The crude solid was re-crystallised twice from acetone and from acetone–diethyl ether, filtered and dried in vacuum over P₄O₁₀. The critical micellar concentration of DTAO, determined by surface tension, is 3.2 × 10⁻⁴ M.

Apparatus

Irradiation with wavelengths >350 nm was carried out using a medium-pressure mercury lamp (GR.E 400, Helios Italquartz) and a glass cut-off filter. Irradiation at 365 nm was performed with a 250 W xenon source equipped with a f/3.4 grating monochromator (Applied Photophysics). All photochemical experiments were carried out in a thermostable cell holder at 22 ± 1 °C. UV–VIS spectra were measured with a Kontron Uvikon 940 spectrophotometer and with a Lambda 6 spectrophotometer from Perkin-Elmer equipped with an integrating sphere.

Gas chromatographic analysis were carried out with a DANI 8521-a equipped with a column, packed with Carbowax 20 M 5% on Chromosorb W-AW, and a flame ionization detector.

Potentiometric titrations were performed using an electronic voltmeter AMEL 337 connected to a reference calomel electrode and a working Ag/AgCl electrode.

Procedures

Preparation of the DTAO/Fe(III)(TDCPP) system. Aqueous solutions of DTAO (5 × 10⁻² M) and Fe(TDCPP) (2 × 10⁻⁵ M) were prepared as follows. Weighted amounts of surfactant and porphyrin were dissolved in CH₂Cl₂. The solvent was allowed to evaporate under stirring and water was added to re-dissolve the two components giving a transparent, liquid, microheterogeneous mixture. The substrates were added directly to the aqueous system in the desired amount. Irradiations were carried out in a 3 ml quartz cell, 1 cm path length (22 ± 1 °C).

Irradiation of de-aerated samples. The aqueous mixture of DTAO and Fe(III)(TDCPP) containing pyridine (4 × 10⁻² M) was irradiated (λ > 350 nm) after de-aeration by four vacuum-line freeze–thaw–pump cycles.

Photooxidation of alkenes. The aqueous system DTAO/Fe(III)(TDCPP) containing cyclohexene or cyclooctene (5 × 10⁻² M) was irradiated (λ > 350 nm) for 4 h under 1 atm of O₂. The degradation of Fe(III)(TDCPP) was followed through the decrease of its Soret band. The photooxidation products were detected by gas chromatographic analysis. Blank experiments in the absence of light or iron porphyrin gave no detectable products.

Photoreduction of CCl₄. Three aqueous systems were irradiated (λ > 350 nm) for 4 h: (i) DTAO/Fe(III)(TDCPP)/CCl₄/ethanol (5 × 10⁻² M/ 2 × 10⁻⁵ M/ 2 × 10⁻² M/ 10% v/v), (ii) DTAO/Fe(III)(TDCPP)/CCl₄/cyclohexanol (5 × 10⁻² M/ 2 × 10⁻⁵ M/ 2 × 10⁻² M/ 7 × 10⁻² M), (iii) DTAO/Fe(III)(TDCPP)/CCl₄/cyclohexanol (5 × 10⁻³ M/ 2 × 10⁻⁶ M/ 5 × 10⁻³ M/ 1.5 × 10⁻² M). Irradiation of the first two systems were carried out in a 3 ml quartz cell, 1 cm path length. By contrast, we were compelled to use a 12 ml quartz cell, 4 cm path length in order to absorb completely the incident light in the experiment carried out in the presence of lower amounts of both DTAO and Fe(III)(TDCPP). After irradiation, the products derived from alcohol oxidation were determined through gas chromatographic analysis. To determine Cl⁻, 0.5 ml of the irradiated mixtures (i) and (ii) (diluted to 10 ml) and 10 ml of (iii) were titrated with AgNO₃. No reactivity was observed in the absence of light or Fe(III)(TDCPP).

The above described systems were also irradiated with monochromatic light at 365 nm (±2 nm) in order to obtain the quantum yield values for the photoreduction of CCl₄, given as the ratio of the number of mol of Cl⁻ ions to the number of mol of absorbed photons.

Acknowledgements

This research was supported by MURST (Programmi di ricerca scientifica di rilevante interesse nazionale) and by CNR.

References

1. M. Sono, M. P. Roach, E. D. Coulter and J. H. Dawson, Chem. Rev., 1996, 96, 2841.
2. D. Dolphin, T. G. Traylor and L. Y. Xie, Acc. Chem. Res., 1997, 30, 251.
3. P. R. Ortiz de Montellano, in Cytochrome P-450, Structure, Mechanism and Biochemistry, Plenum Press, New York, 2nd edn., 1995..
4. B. Meunier, Chem. Rev., 1992, 92, 1411.
5. D. Mansuy, Pure Appl. Chem., 1990, 62, 741.
6. P. E. Ellis and J. E. Lyons, Coord. Chem. Rev., 1990, 105, 181.
7. D. Brault, C. Bizet, P. Morliere, M. Rougee, E. J. Land, R. Santus and A. J. Swallow, J. Am. Chem. Soc., 1980, 102, 1015.
8. D. Brault and P. Neta, J. Phys. Chem., 1982, 86, 3405.
9. D. Brault, Environ. Health Perspect., 1985, 64, 53.
10. D. Mansuy, M. Lange, J. C. Chottard, P. Guerin, P. Morliere, D. Brault and M. Rougee, J. Chem. Soc. Chem. Commun., 1977, 648.
11. D. Mansuy and M. Fontecave, Biochem. Biophys. Res. Comm., 1982, 104, 1651.
12. D. Mansuy and M. Fontecave, Biochem. Pharm., 1983, 32, 1871.
13. A. Maldotti, L. Andreotti, A. Molinari and V. Carassiti, J. Biol. Inorg. Chem., 1999, 4, 154.
14. D. N. Hendrickson, M. G. Kinnaird and K. S. Suslick, J. Am. Chem. Soc., 1987, 109, 1243.
15. M. W. Peterson, D. S. Rivers and R. M. Richman, J. Am. Chem. Soc., 1985, 107, 2907.
16. A. Maldotti, C. Bartocci, R. Amadelli, E. Polo, P. Battioni and D. Mansuy, J. Chem. Soc. Chem. Commun., 1991, 1487.
17. L. Weber, R. Hommel, J. Behling, G. Haufe and H. Hennig, J. Am. Chem. Soc., 1994, 116, 2400.
18. A. Maldotti, C. Bartocci, G. Varani, A. Molinari, P. Battioni and D. Mansuy, Inorg. Chem., 1996, 35, 1126.
19. A. Maldotti, A. Molinari, L. Andreotti, M. Fogagnolo and R. Amadelli, Chem. Commun., 1998, 4, 507.
20. C. Bartocci, A. Maldotti, G. Varani, P. Battioni, V. Carassiti and D. Mansuy, Inorg. Chem., 1991, 30, 1255.
21. A. Maldotti, C. Bartocci, R. Amadelli and V. Carassiti, J. Chem. Soc., Dalton Trans., 1989, 1197.
22. T. G. Traylor and S. Tshuchiya, Inorg. Chem., 1987, 26, 1338.
23. C. Bartocci, F. Scandola, A. Ferri and V. Carassiti, J. Am. Chem. Soc., 1980, 102, 7067.
24. J. Sima and J. Makanova, Coord. Chem. Rev., 1997, 160, 161.
25. R. D. Arasingham, A. L. Balch, C. R. Cornman and L. Latos-Grazynski, J. Am. Chem. Soc., 1989, 111, 4357.
26. T. G. Traylor and F. Xu, J. Am. Chem. Soc., 1990, 112, 178.
27. J. T. Groves and Y. Watanabe, J. Am. Chem. Soc., 1988, 110, 8443.
28. K. Yamaguchi, Y. Watanabe and I. Morishima, J. Am. Chem. Soc., 1993, 115, 4058.
29. A. Maldotti, R. Amadelli, A. Molinari and V. Carassiti, in Green Chemistry: Challenging Perspectives, ed. P. Tundo and P. Anastas, Oxford University Press, New York, 2000, p. 125..
30. A. Maldotti, C. Bartocci, R. Amadelli, G. Varani, E. Polo and V. Carassiti, in Chemistry and Properties of Biomolecular Systems, ed. E. Rizzarelli and T. Theophanides, Kluwer Academic Press, Dordrecht, The Netherlands, 1991, p. 103..
31. D. E. Van Sickle, F. R. Mayo and R. M. Arluck, J. Am. Chem. Soc., 1965, 87, 4824.
32. A. C. Cope and H. H. Lee, J. Am. Chem. Soc., 1957, 79, 964.

---