Figure 5. (a). Comparison of the experimental CD spectrum of
oligomer 3a (black) and predicted CD spectra of 3a (RI-BP86/
TZVP in orange, B3LYP/SV(P) in blue, and BHLYP/SV(P) in red),
calculated for geometries optimized at the respective levels of
theory. A comparison with oligomer 2a is in the Supporting
Information (Figure S1).
Figure 4. Molar CD spectra of oligomers 1a-3b and UV-vis
spectra of 3a in MeCN at -10 °C.
contrast, two-turn oligomer 2a ((Cbz-L-Ala)2(NdN)4) dis-
played negative CD bands at 320 and 477 nm, and an intense
positive band at 407 nm. The presence of the CD transitions
in this region of the spectrum indicate the induction of a
preferred sense of helicity in the azobenzene oligomer. The
four-turn oligomer 3a ((Cbz-L-Ala)2(NdN)8) also exhibited
these CD bands with intensities greater than 2-fold that of
2a, as might be expected based on the difference in the
number of azobenzene chromophores. This nonlinear en-
hancement of the CD intensity for 3a, compared with 2a,
indicates a more efficient amplification of the terminal chiral
influences.
The UV-vis and CD spectra of the optimized geometry
of oligomer 2a were calculated using time-dependent density-
functional theory (TD-DFT) at the RI-BP86/TZVP, B3LYP/
SV(P), and BH&HLYP/SV(P)13 levels of theory (Figure 5).14
The resulting spectra were uniformly blue-shifted by 0.5,
0.1, and -0.55 eV, respectively, and a 0.3 eV Gaussian line-
broadening was applied to each excitation.
a fully unfolded state. Hence, the calculated CD spectra of
a static M or P helix would qualitatively reproduce the
features of an experimental spectrum. The agreement be-
tween the simulated spectra and experiment was fairly good,
with the B3LYP spectrum most closely matching experiment.
The qualitative properties of the spectrum and, in particular,
the trough at ∼475 nm and the peak at ∼400 nm were
reproduced; however, the trough at 340 nm shows signifi-
cantly more negative polarization in the calculated spectra.
The BH&HLYP and RI-BP86 spectra show larger deviations
than the B3LYP spectrum, suggesting that the latter has the
proper amount of inclusion of Fock-exchange.15 It should
be noted that the geometries optimized using hybrid DFT
most closely resemble the crystal structure,7 with RI-BP86
overestimating the spacing between the helical turns. Inter-
estingly, the molecular mechanics geometry was also similar
to experiment (see Supporting Information).
The calculated spectra most closely resemble the experi-
mental spectrum of the 4-turn oligomer 3a, particularly in
the region around 320-370 nm, compared with the 2-turn
oligomer 2a. The B3LYP spectrum, in particular, matches
exceptionally well. As the CD spectra were calculated using
static, helical structures, it is reasonable that they might
resemble the experimental spectrum of the very strongly
biased oligomer 3a more closely than that of 2a. Although
oligomers 2a and 3a differ in length, the chromophores are
identical and the excited-state properties are likely quite
similar. With the multiple levels of theory employed,
including the generally more accurate hybrid DFT methods,
and a clear similarity with the spectrum of oligomer 3a, the
calculated CD spectra strongly support the conclusion that
the P helical sense is present in the experimental oligomers.
The prediction of CD spectra has only recently been
applied to the analysis of conformations of large molecules,
and our protocol mirrors that utilized successfully else-
where.15 Oligomers 2a and 3a are ideal candidates for CD
calculations because, as suggested by the REMD studies,
the M and P helical forms interconvert without going through
(13) (a) Becke, A. D. Phys. ReV. A. 1988, 38, 3098. Perdew, J. P. Phys.
ReV. B. 1986, 33, 8822. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(c) Becke, A. D. J. Chem. Phys. 1993, 98, 1372. (d) Lee, C.; Yang, W.;
Parr, R. G. Phys. ReV. B 1988, 37, 785.
(14) (a) The lowest energy conformation obtained from the Monte-Carlo
conformational search was subsequently optimized using the respective DFT
methods prior to UV-vis and CD spectrum calculations. These calculations
were performed using Turbomole 5.91. Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.;
Horn, H.; Ko¨lmel, C. Chem. Phys. Lett. 1989, 62, 165. (b) Ha¨ser, M.;
Ahlrichs, R. J. Comput. Chem. 1989, 10, 104. (c) Treutler, O.; Ahlrichs, R.
J. Chem. Phys. 1995, 102, 346–354. Arnim, M. v.; Ahlrichs, R. J. Comput.
Chem. 1998, 19, 1746. (d) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995,
102, 346. (e) Arnim, M. V.; Ahlrichs, R. J. Comput. Chem. 1998, 19, 1746.
(15) (a) Diedrich, C.; Grimme, S. J. Phys. Chem. A 2003, 107, 2524–
2539. (b) Mori, T.; Inoue, Y.; Grimme, S. J. Org. Chem. 2006, 71, 9797–
9806. (c) Crawford, T. D.; Tam, M. C.; Abrams, M. L. J. Phys. Chem. A
2007, 111, 12057–12068. (d) Stephens, P. J.; Devlin, F. J.; Gasparrini, F.;
Ciogli, A.; Spinelli, D.; Cosimelli, B. J. Org. Chem. 2007, 72, 4707–4715.
¨
(f) Eichkorn, K.; Treutler, O.; Ohm, H.; Ha¨ser, M.; Ahlrichs, R. Chem.
Phys. Lett. 1995, 242, 652. (g) Eichkorn, K.; Weigend, F.; Treutler, O.;
Ahlrichs, R. Theor. Chem. Acc. 1997, 97, 119. (h) Ahlrichs, R. Phys. Chem.
Chem. Phys. 2004, 6, 5119.
Org. Lett., Vol. 10, No. 9, 2008
1673