Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) Detection of EPR spectra in S = 2 states of MnIII(salen) complexes Konstantin P. Bryliakov,a Dmitrii E. Babushkinb and Evgenii P. Talsi*b a Department of Natural Sciences, Novosibirsk State University, 630090 Novosibirsk, Russian Federation. Fax: +7 3832 35 5756 b G. K. Boreskov Institute of Catalysis, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Russian Federation.Fax: +7 3832 34 3766; e-mail: talsi@catalysis.nsk.su The EPR signals of MnIII(salen) complexes (R,R)-(–)-N,N'-bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediaminomanganese(III) chloride 1 and N,N'-bis(salicylidene)ethylenediaminomanganese(III) chloride 2 were detected and used for the characterisation of intermediates in catalytic epoxidation.A recent major achievement in catalytic enantioselective oxidation is the epoxidation of prochiral unfunctionalised olefins catalysed by Mn(salen) complexes.1–4 Two practicable catalytic systems for the enantioselective epoxidation of unfunctionalised olefins were developed. One of them involves a two-phase system with commercial aqueous buffered bleach phase and an organic phase that is a solution of a substrate and a catalyst in a suitable solvent.1 The other system is a solution of m-chloroperbenzoic acid (m-CPBA), N-methylmorpholine N-oxide (NMO) and a catalyst in dichloromethane at a low temperature (–78 °C).5,6 The latter system is effective in the enantioselective epoxidation of styrene.5 To elucidate the mechanism of MnIII(salen)-catalysed oxidation, it is important to monitor transformations of the catalyst in the course of the catalytic reaction.In this work, we report the first EPR data on MnIII(salen) complexes (R,R)-(–)-N,N'-bis(3,5-ditert- butylsalicylidene)-1,2-cyclohexanediaminomanganese(III) chloride 1 and N,N'-bis(salicylidene)ethylenediaminomanganese(III) chloride 2 in various solvent systems.The preliminary data on the interaction of complex 1 with iodosylbenzene (PhIO) and m-CPBA are also presented. EPR spectroscopy has rarely been applied to study the electronic structure of trivalent manganese complexes with an even number of unpaired electrons. This is a result of large zero-field splittings or fast spin relaxation processes. Although a few of EPR studies of trivalent manganese impurity ions and complexes have been reported, these have relied largely on indirect detection methods or very high observation frequencies.7–10 There is the only report where weak EPR transition at g ª 8 was observed for manganese(III) acetylacetonate at 12 K by conventional X-band EPR spectroscopy.11 Dexheimer et al.interpreted the Mn3+ spectrum using the following spin Hamiltonian: The zero-field interaction splits the levels of an S = 2 spin system into two doublets, one of them is a linear combination of the ms = ô±2ñ states, and the other, of the ms = ô±1ñ states, and a singlet corresponding to the ms = ô0ñ state.The forbidden EPR transitions may be observed between the levels of the ô±2ñ non-Kramers doublet.The X-band EPR spectrum of a frozen 0.1 M solution of complex 1 in CH2Cl2 at 77 K is shown in Figure 1(a).† The field position and shape of the observed weak signal at g = 8.0 ± 0.3 are close to those for the signal observed for manganese(III) acetylacetonate and attributed to forbidden transitions within ô±2ñ non-Kramers doublet.11 The relatively sharp resonance at g = 4.3 is characteristic of rhombic FeIII complexes and belongs to very small impurities (less than 1 mol%) of FeIII species in complex 1.The addition of FeCl3·6H2O to a solution of complex 1 gives rise to a sharp increase in the signal at g = 4.3. The FeIII impurities were detected not only in our particular sample. The EPR spectrum of the optical isomer of † General experimental details.Complex 1 [(R,R)-(–)-N,N'-bis(3,5-ditert- butylsalicylidene)-1,2-cyclohexanediaminomanganese(III) chloride] and N-methylmorpholine N-oxide from Aldrich were used as received. Iodosylbenzene was prepared by hydrolysis of the corresponding diacetate (Aldrich) with aqueous sodium hydroxide and stored at 253 K. Complex 2 [N,N'-bis(salicylidene)ethylenediaminomanganese(III) chloride] and its MnII precursor were prepared as described in ref. 12. All other chemicals and solvents were of reagent grade, and they were used without further purification. EPR spectra were recorded in quartz tubes (d = 5 mm) at 77 K using a Bruker ER-200D X-band spectrometer. O H N Mn O H N H H Cl O H N Mn O H N Cl 1 2 H=b(gz Hz Sz +gy HySy+gzHzSz) + D(Sz 2 – 2) + E(Sz 2 – Sy 2) (1) (a) (b) (c) (d) (e) * 0.03 0.13 0.23 0.33 H(T) 0.067 0.093 g = 8.0 g = 4.3 g = 2.0 Mn A = 43 G H(T) Figure 1 X-band EPR spectra (77 K) of 0.05 M solutions of complex 1 (a) in CH2Cl2 and (b)–(c) in CH2Cl2 containing N-methylmorpholine N-oxide ([NMO] = 1 M); (d) EPR spectrum (77 K) of 0.05 M solution of a MnII(salen) precursor of complex 2 in DMSO; (e) EPR spectrum (77 K) of MnIV(salen) complex recorded 1 min after the onset of reaction of complex 1 ([1] = 0.05 M) with one equivalent of m-chloroperbenzoic acid at 273 K [spectrometer frequency, 9.3 GHz; microwave power, 40 mW; modulation frequency, 100 kHz; modulation amplitude, 20 G; gain, 2.5×105 (a)–(c), 2.5×103 (d), 2.5×104 (e)].Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) complex 1 (S,S)-(+)-N,N'-bis(3,5-di-tert-butylsalicylidene)-1,2- cyclohexanediaminomanganese(III) chloride 1' (Aldrich) also exhibited a resonance signal at g = 4.3 but its intensity was lower than that in complex 1 by a factor of two.The signals at g = 8.0 for complexes 1 and 1' coincided. The nature of the additional low-field line marked in Figure 1(a) with an asterisk is still unclear. Coordination of N-methylmorpholine N-oxide to complex 1 changes the shape of the EPR signal, and the six-line hyperfine structure from one manganese ion (I = 5/2) can be clearly seen [Figures 1(b)–(c)].The hyperfine splitting (44±3 G) that appears at the g = 8 signal is rather close to that determined for MnIII impurity ions in TiO2 (Az = 53 G)10 and for manganese(III) acetylacetonate (Az = 55 G).11 We have compared the EPR signals of 0.1 M solutions of complexes 1 and 2 in dimethylsulfoxide (DMSO) at 77 K.DMSO was used as a solvent owing to proper solubility of both complexes. The positions, shapes and intensities of the signals observed at g = 8.0 for complexes 1 and 2 coincided. This result supports the assignment of a resonance at g = 8.0 to complex 1 rather than to any manganese impurities.It is improbable that the concentrations of such impurities were equal in complexes 1 and 2. Low-symmetry S = 5/2 MnII species, which may be present as impurities in MnIII compounds, can also give rise to downfield EPR signals. The two species can be clearly distinguished, however, because the S = 5/2 MnII(salen) system produces very intense resonance at g = 2.0 in addition to any other downfield signals.Figure 1(d) shows the EPR spectrum of a MnII(salen) precursor of complex 2 in DMSO, which was prepared according to the procedure described in ref. 12. This spectrum was recorded at an amplification lower than that in Figure 1(a) by two orders of magnitude. It is seen that MnII(salen) exhibits an intense signal at g = 2.0 with the partially resolved hyperfine splitting (87 G) from the manganese nucleus.The MnIV(salen) complex obtained via a reaction of complex 1 with one equivalent of m-CPBA in CH2Cl2 at 273 K exhibits a resonance at g = 5.7±0.3, with the hfs splitting (73 G) from manganese nucleus typical of MnIV species with D > hn13–16 [Figure 1(e)]. The amplification in Figure 1(e) is lower than that in Figure 1(a) by one order of magnitude, while concentrations of MnIII and MnIV species are equal.Thus, the signal of MnIII(salen) is much weaker than that of MnIV(salen) at equal manganese concentrations. This result agrees with the literature data. It was found for manganese impurity ions in TiO2 that the resonances of MnIII are about an order of magnitude weaker than those of the same quantity of corresponding MnIV species.10 Based on the aforesaid, we can conclude that it is mononuclear MnIII(salen) complex 1 that exhibits the EPR signal at g = 8.0±0.3.Dimers or higher aggregates of 1 can be ruled out. Based on the data for MnIII/MnIV and MnII/MnIII mixed-valence binuclear complexes, more than six line hyperfine splitting is expected for dimeric or oligonuclear species.17,18 The EPR signal of complex 1 was found to be very sensitive to the nature of axial ligands [compare Figures 1(a) and 1(b)].Another illustration of this fact is presented in Figure 2. It can be seen, that the intensity and shape of the resonance at g = 8.0 dramatically changed with an increase in the concentration of (a) (b) (c) (d) (e) (f) 0.0 0.08 0.16 H(T) g = 8.0 g = 4.3 Figure 2 X-band EPR spectra (77 K) of 0.05 M solutions of complex 1 (a) in CH2Cl2 and (b)–(f) in CH2Cl2 containing various amounts of pyridine: (b) [Py] = 0.0125 M, (c) 0.025 M, (d) 0.0375 M, (e) 0.05 M, (f) 0.1 M.The spectrometer settings are given in Figure 1. (a) (b) g = 8.0 g = 4.3 0.03 0.13 H(T) Figure 3 (a) X-band EPR spectrum (77 K) of a 0.05 M solution of complex 1 in CH2Cl2; (b) EPR spectrum of sample (a) 3 min after stirring with one equivalent of PhIO at 273 K.Mendeleev Communications Electronic Version, Issue 1, 1999 (pp. 1–44) pyridine in the solution of complex 1. This change was stopped when the [Py]/[1] ratio reaches unity, in agreement with the coordination of one pyridine molecule per molecule of complex 1. The reason for the decrease of the EPR signal of complex 1 via pyridine coordination is still unclear.It is known that axial ligands dramatically affect the D value in MnIII porfirazine complexes.11 Unfortunately, we have not found the relations between the D value and the probability of the ô–2ñ ® ô+2ñ transition in the literature. The interaction of complex 1 with one equivalent of m-CPBA at 183 K in CH2Cl2 gives rise to an immediate five-fold decrease in the intensity of the resonance at g = 8.0 similarly to that observed in Figure 2 via pyridine coordination.The valent state of manganese remains unchanged during this reaction, because a very weak signal of MnIV can be detected. Thus, the observed drop of the intensity of the EPR signal at g = 8.0 is caused by the conversion of complex 1 into another MnIII complex, which is characterised by a lower intensity of the MnIII resonance.This complex will be referred to as complex 3. Complex 3 is extremely unstable. It exists only at 183–213 K and rapidly and quantitatively decomposes at higher temperatures to form metastable MnIV species, which were detected by EPR. By analogy with the well-known formation of acylperoxo complexes via the interaction of MnIII porphyrins with m-CPBA at low temperatures,19 it is reasonable to suggest that complex 3 is the acylperoxo complex MnIII(salen)(OOCOAr).The reactivity of this complex toward alkenes will be further investigated. An interesting behaviour was observed in the interaction of complex 1 with PhIO. Immediately after 3 min stirring of a 0.05 M solution of complex 1 with a suspension of PhIO in CH2Cl2 at 273 K, the EPR signal of complex 1 was transformed into another signal of MnIII.The field position, shape and intensity of this signal markedly differ from those of 1 [compare Figures 3(a) and 3(b)]. Thus, a new complex of MnIII, which is further denoted as complex 4, is formed. Probably, complex 4 is the adduct MnIII(salen)(OIPh).Recently, this adduct was detected by electrospray tandem mass spectrometry in the catalytic system 1 + PhIO in CH2Cl2.20 Further studies are needed to support our assumption. In conclusion, we have observed for the first time X-band EPR spectra of MnIII(salen) complexes and demonstrated the applicability of EPR to studies of these practically important systems. 1H NMR and EPR spectroscopic studies of the transformations of the MnIII(salen) catalyst in the course of enantioselective epoxidation are in progress.This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-32495a). References 1 E. N. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker and L. Deng, J. Am. Chem. Soc., 1991, 113, 7063. 2 T.Katsuki, Coord. Chem. Rev., 1995, 140, 189. 3 M. Palucki, N. S. Finney, P. J. Pospisil, M. L. Guler, T. Ishida and E. N. Jacobsen, J. Am. Chem. Soc., 1998, 120, 948. 4 W. Adam, R. T. Fell, V. R. Stegmann and Ch. R. Saha-Moller, J. Am. Chem. Soc., 1998, 120, 708. 5 M. Palucki, P. J. Pospisil, W. Zhang and E. N. Jacobsen, J. Am. Chem. Soc., 1994, 116, 9333. 6 M. Palucki, G. J. McCormick and E.N. Jacobsen, Tetrahedron Lett., 1995, 36, 5457. 7 D. P. Goldberg, J. Telser, J. Krzystek, A. Garrido Montalban, L. S. Brunel, A. G. M. Barrett and B. M. Hoffman, J. Am. Chem. Soc., 1997, 119, 8722. 8 A.-L. Barra, D. Catteschi, R. Sessoli, G. L. Abbati, A. Cornia, A. C. Fabretti and M. G. Uytterhoeven, Angew. Chem., Int. Ed. Engl., 1997, 36, 2329. 9 R. L. Aurbach and P. L.Richards, Phys. Rev. B., 1975, 12, 2588. 10 H. J. Gerritsen and E. S. Sabisky, Phys. Rev., 1963, 132, 1507. 11 S. L. Dexheimer, J. W. Gohdes, M. K. Chan, K. S. Hagen, W. H. Armstrong and M. P. Klein, J. Am. Chem. Soc., 1989, 111, 8923. 12 K. Srinivasan, P. Michaud and J. K. Kochi, J. Am. Chem. Soc., 1986, 108, 2309. 13 S. Pal, Ph. Ghosh and A. Chakravorty, Inorg. Chem., 1985, 24, 3704. 14 D. P. Kessissglou, W. M. Butler and V. L. Pecoraro, J. Chem. Soc., Chem. Commun., 1986, 1253. 15 A. Smegal and C. L. Hill, J. Am. Chem. Soc., 1983, 105, 3515. 16 J. T. Groves and M. K. Stern, J. Am. Chem. Soc., 1988, 110, 8628. 17 G. C. Dismukes, J. E. Sheats and J. A. Smegal, J. Am. Chem. Soc., 1987, 109, 7202. 18 B. Mabad, J.-P. Tuchagues, Y. T. Hwang and D. N. Hendrickson, J. Am. Chem. Soc., 1985, 107, 2801. 19 J. T. Groves and Y. Watanabe, Inorg. Chem., 1986, 25, 4808. 20 D. Feichtinger and D. A. Plattner, Angew. Chem., Int. Ed. Engl., 1997, 36, 1718. Received: Moscow, 31th July 1998 Cambridge, 23rd October 1998; Com. 8/06228E