A. C. Benniston et al. / Tetrahedron Letters 46 (2005) 7291–7293
7293
concerns the values of the reduction potentials. For C3,
the first (E1 = ꢀ0.64 V (57 mV) vs Fc/Fc+) and second
(E2 = ꢀ0.87 V (60 mV) vs Fc/Fc+) reduction potentials
are located at more negative potentials than those
observed for MV2+ (E1 = ꢀ0.35 V (70 mV); E2 =
ꢀ0.76 V (70 mV) vs Fc/Fc+) under identical experi-
mental conditions. This observation can be attributed
to the alkoxy substituents at the 3,30-positions donating
electron density into the pyridinium rings. However, it is
worth noting that the values of E1 and E2 are very sim-
ilar for C1 (E1 = ꢀ0.31 V (51 mV) E2 = ꢀ0.73 V
(61 mV) vs Fc/Fc+) and MV2+. The peak difference
(E2 ꢀ E1, DE) for C3 (DEC3 = 0.23 V) is considerably
smaller than those found for either C1 (DEC1 = 0.42 V)
MV+Å, suggesting that the alkoxy chain severely curtails
C–C bond rotation.
Tentatively, there seems to be a correlation between DE
and the dihedral angle for the one-electron reduced
form,13 but the true relationship will only become more
clear when a sizable series of strapped derivatives
are available. These are currently under preparation
and will be reported at a later date when a full theoret-
ical model to explain the angle effect will also be
presented.
Acknowledgements
or MV2+ ðDE2þ ¼ 0:41 VÞ. In comparing the two
MV
strapped derivatives it is clear that DEC1 is considerably
larger than DEC3. That the length of the constraining
strap can significantly alter the reduction potential is
consistent with reports by Thummel et al. on analogues.
They showed that the first reduction potential of diquat
(2,20-dipyridinium) derivatives, where the two nitrogen
atoms were linked via alkyl spacers, gradually shifted
to more negative potentials as the length of the spacer
increased.11 No detailed theoretical description to
account for the observations was given.
We thank the EPSRC (GR/R23305) for financial
support and the EPSRC-funded Mass Spectrometry
Service at Swansea for obtaining the electrospray mass
spectra.
References and notes
1. Bird, C. L.; Kuhn, A. T. Chem. Soc. Rev. 1981, 10, 49–82.
2. Bockman, T. M.; Kochi, J. K. J. Org. Chem. 1990, 55,
4127–4135.
Molecular modelling calculations12 were carried out on
C1/C3 and MV2+ to calculate dihedral angles for the
dicationic (fully oxidised) and mono-cationic (one-elec-
tron reduced) species (Table 1). Evidently, the short
methylene constraining strap of C1 severely restricts
the connector C–C bond rotation and hence dihedral
angle, whereas for both C3 and MV2+ there is more flex-
ibility (Fig. 3). Mono-reduction of both C1 and C3 does
not lead to a fully planar structure, as in the case for
3. Hunig, S.; Gross, J. Tetrahedron Lett. 1968, 9, 2599–2604.
4. Sleegers, A.; Dehmlow, E. V. Leibigs Ann. Chem. 1992,
953–959.
5. Suzuki, A. J. Organomet. Chem. 1999, 576, 147–168.
6. Miah, M. A. J.; Snieckus, V. J. Org. Chem. 1985, 50,
5436–5438.
7. Wuest, H. M.; Sakal, E. H. J. Am. Chem. Soc. 1951, 73,
1210–1216.
8. Analytical data: 1H NMR (d, 300 MHz, CD3CN): 4.39 (s,
6H, CH3), 5.84 (s, 2H, H of O–CH2), 8.59 (s, br, 4H, H of
Py), 8.78 (s, br, 2H, H of Py). 13C NMR (d, 75 MHz,
CD3CN): 48.0, 96.3, 127.3, 134.4, 139.2, 140.2, 156.1. ES-
MS (m/z): 231.0 (calcd Mr = 230.3 for [Mꢀ2PF6]+), 114.9
(calcd Mr = 115.1 for [Mꢀ2PF6]2+). Anal. Calcd for
C13H14N2O2ÆP2F12: C, 30.01; H, 2.71; N, 5.39. Found:
C, 30.13; H, 2.75; N, 5.41.
Table 1. Calculated torsion angles for constrained derivatives and
methylviologen
Compound
Dihedral
angle +2 (°)a
Dihedral
angle +1 (°)b
Angle change
(°)
9. Analytical data: 1H NMR (d, 300 MHz, CD3CN): 2.25 (q,
J = 5.1Hz, 2H, –CH2–), 4.39 (s, 6H, CH3), 4.66 (t,
J = 5.1Hz, 4H, H of –OCH2–), 8.04 (d, J = 6.1Hz, 2H,
Py–H6), 8.57 (d, J = 6.1Hz, 2H, Py–H5), 8.77 (s, 2H, Py–
H3). 13C NMR (d, 75 MHz, CD3CN): 28.9, 48.5, 74.3,
127.6, 137.4, 140.5, 141.6, 155.6. ES-MS (m/z): 403.2
(calcd Mr = 403.1 for [MꢀPF6]+), 257.1 (calcd Mr = 257.1
for [Mꢀ2PF6ꢀH]+), 128.9 (calcd Mr = 129.1 for
[Mꢀ2PF6]2+). Anal. Calcd for C15H18N2O2ÆP2F12: C,
32.86; H, 3.31; N, 5.11. Found: C, 33.15; H, 3.49; N, 5.10.
10. One-electron reduction process confirmed by comparison
with an equimolar solution of ferrocene.
C1
C3
MV2+
35.0
57.6
89.7
22.3
42.7
0.0
12.7
14.9
89.7
a Fully oxidised dicationic compound.
b One-electron reduced mono-cationic compound.
11. Thummel, R. P.; Lefoulon, F.; Chirayil, S.; Goulle, V.
J. Org. Chem. 1988, 53, 4745–4747.
12. Energy minimised structures were calculated in the gas
phase using the AM1 semiempirical method found in
Quantum CAChe.
13. A relationship between dihedral angle and the reduction
potential has been reported for diquat derivatives, see:
Brienne, S. H. R.; Boyd, P. D. W.; Schwerdtfeger, P.;
Bowmaker, G. A.; Clooney, R. P. J. Mol. Struct. 1995,
356, 81–94.
Figure 3. Ball and stick molecular modelling representations of C1
(left) and C3 (right).