1,2,3-Triazoles as conjugative π-linkers in push-pull chromophores: Importance of substituent positioning on intramolecular charge-transfer

Article abstract of DOI:10.1021/ol801253z

(Chemical Equation Presented) Isomeric charge-transfer chromophores using 1,2,3-triazol-diyl as linker have been studied experimentally and computationally. The instability of the polarized reactants precluded the use of the Huisgen reaction and alternative synthetic methodologies were employed. Charge-transfer absorptions between an N,N-dimethylanilino and a dicyanovinyl group are modest to strong, with maxima from λ_max = 400 to 453 nm depending on substituent positioning. TD-B3LYP/6-31G(d) calculations are within 0.6 eV of experiment and assign these bands as HOMO-LUMO transitions.

Full text of DOI:10.1021/ol801253z

ORGANIC
LETTERS
2008
Vol. 10, No. 15
3347-3350
1,2,3-Triazoles as Conjugative π-Linkers
in Push-Pull Chromophores:
Importance of Substituent Positioning
on Intramolecular Charge-Transfer
Peter D. Jarowski, Yi-Lin Wu, W. Bernd Schweizer, and Franc¸ois Diederich*
Laboratorium fu¨r Organische Chemie, Eidgeno¨ssische Technische Hochschule,
Ho¨nggerberg, HCI, CH-8093 Zu¨rich, Switzerland
diederich@org.chem.ethz.ch
Received June 4, 2008
ABSTRACT
Isomeric charge-transfer chromophores using 1,2,3-triazol-diyl as linker have been studied experimentally and computationally. The instability
of the polarized reactants precluded the use of the Huisgen reaction and alternative synthetic methodologies were employed. Charge-transfer
absorptions between an N,N-dimethylanilino and a dicyanovinyl group are modest to strong, with maxima from λmax ) 400 to 453 nm depending
on substituent positioning. TD-B3LYP/6-31G(d) calculations are within 0.6 eV of experiment and assign these bands as HOMO-LUMO transitions.
The independent discoveries of the Cu(I)-catalyzed Huisgen’s
1,3-dipolar cycloaddition1 by Sharpless2a and Meldal2b in
2002 have led to thousands of new examples of this reaction
subclass. The excellent regioselectivity for the 1,4-isomer
of the 1,2,3-triazole and the enhanced reactivity of the Cu(I)-
catalyzed process, along with the versatility of reaction
conditions (and other factors) have made it the flagship
reaction of the Sharpless “click” chemistry.3 As such, it has
been widely applied of late, particularly in bioconjugation
and drug discovery, and also in advanced material sciences
toward dendrimers and surface modification.4
While 1,2,3-triazole chemistry is historically related to the
manufacture of dyes and to photographic imaging,5 advance-
ments based on the “click” methodology have resulted in
only a few examples of photophysical interest. In general,
the recent examples favor the use of the 1,2,3-triazol-1,4-
diyl moiety as a structural linkage between two components
and are not concerned with mediating their electronic
communication. This is related to the stability and ease of
synthesis of alkyl azides, which are by far the most often
used; the resultant 1,2,3-triazole adducts are photophysically
(4) (a) Moses, J. E.; Moorhouse, A. D. Chem. Soc. ReV. 2007, 36, 1249.
(b) Nandivada, H.; Jiang, X.; Lahann, J. AdV. Mater. 2007, 19, 2197. (c)
Ou, T.-M.; Lu, Y.-J.; Tan, J.-H.; Huang, Z.-S.; Wong, K.-Y.; Gu, L.-Q.
Chem. Med. Chem 2008, 3, 690.
(1) Huisgen, R.; Szeimies, G.; Moebius, L. Chem. Ber. 1967, 100, 2494.
(2) (a) Rostovtsev, V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int Ed. 2002, 41, 2596. (b) Tornoe, C. W.; Christensen, C.;
Meldal, M. J. Org. Chem. 2002, 67, 3057.
(5) (a) Wamhoff, H. In ComprehensiVe Heterocyclic Chemistry; Katritz-
ky, A. R., Rees, C. W., Eds.; Pergamon: Oxford, 1984; Vol. 5, Part 4A, p
669. (b) Fan, W.-Q.; Katritzky, A. R. In ComprehensiVe Heterocyclic
Chemistry II; Katritzky, A. R., Rees, C. W., Scriven, E. F. V., Eds.; Elsevier:
Oxford, 1996; Vol. 4, p 1.
(3) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004.
10.1021/ol801253z CCC: $40.75
Published on Web 06/27/2008
2008 American Chemical Society

dark to through-bond interaction of the unified components.
Examples are only now appearing that suggest movement
in the direction of extended π-chromophores that incorporate
the 1,2,3-triazole moiety into the conjugation path.6
One active direction in organic materials chemistry
involves the development of charge-transfer (CT) chro-
mophores as nonlinear optical (NLO) components in pho-
tonics applications.7 A CT chromophore is composed of a
strongly electron donating moiety conjugated through a
π-linker to a strongly electron-accepting one. To our
knowledge, this basic motif has not been attempted with
1,2,3-triazol-diyls as the active π-linker. For a first systematic
study of such systems, the strong electron donor N,N-
dimethylanilino (DMA) and the strong electron acceptor
dicyanovinyl (DCV) were chosen as substituents.8 The 1,2,3-
triazole ring has four unique points of attachment, two on
the heteroatoms N-1 and N-2 and two on carbons C-4 and
C-5. Seven constitutional isomers are possible, compounds
1-7 (Figure 1).
π-linker leaving the synthetic challenge of producing such
materials by direct “click” chemistry for future work. Charge-
transfer between these groups is documented herein both
experimentally, when possible, and computationally9 for all
seven isomers.
Isomers 1-4, with substituents at N-1, C-4, and C-5,
formally result from the Huisgen cycloaddition under thermal
or Cu(I)-catalyzed “click” conditions. The reaction of 4-azido-
N,N-dimethylaniline (8)10 with dicyanoenyne (9) formed the
basis of the synthetic approach toward isomers 1 and 2
(Scheme 1), but with limited success.
Scheme 1. Synthesis of Isomeric 1,2,3-Triazoles 1, 2, 5 and 6.
Figure 1. Isomeric donor-acceptor-substituted 1,2,3-triazoles 1-7.
Intermediate 9 is not isolable and must be generated in situ
by desilylation of the corresponding trimethylsilyl-protected
enyne (10)11 with MeOH.12a Unfortunately, this synthesis of 1
was hampered by the apparent instability of 9, which went on
to subsequently react with 8 through a low-yielding process
(ca. 5%) after 2 or more days of reaction time. Alternatively,
isomer 1 could be prepared efficiently by reaction of 8 and
propynal to give triazole 12 (67%), followed by Knoevenagel
condensation (99%). Isomer 2, which is also not available
through the Cu(I)-catalyzed process, was prepared instead by
the thermal 1,3-dipolar cycloaddition of 8 and 10 to give 11
Of the isomers presented in Figure 1, only 1, 2, 5 and 6
could so far be prepared. Moreover, the use of “click”
chemistry as the key and final step of their synthesis was
severely limited by the instability of the requisite polarized
reactants. Thus, these isomers were prepared by alternative
synthetic methods in order to establish the feasibility of
constructing CT chromophores with 1,2,3-triazol-diyl as the
(6) (a) Schweinfurth, D.; Hardcastle, K. I.; Bunz, U. H. F. Chem.
Commun. 2008, 2171. (b) Meudtner, R. M.; Ostermeier, M.; Goddard, R.;
Limberg, C.; Hecht, S. Chem. Eur. J. 2007, 13, 9834. (c) Parent, M.;
Mongin, O.; Kamada, K.; Katan, C.; Blanchard-Desce, M. Chem. Commun.
2005, 2029.
(9) Frisch, M. J. et al. Gaussian, Revision C.02, Gaussian, Inc.,
Wallingford CT, 2004. A full reference may be found in the Supporting
Information.
(7) (a) May, J. C.; Biaggio, I.; Bures, F.; Diederich, F. Appl. Phys. Lett.
2007, 90, 251106/1. (b) Forest, S. R., ; Thomson, M. E., Eds. Organic
Electrooptics and Optoelectronics, Special Issue of Chem. ReV. 2007, 107,
923.
(10) Ugi, I.; Perlinger, H.; Behringer, L. Chem. Ber. 1958, 91, 2330.
(11) (a) Sorg, A.; Siegel, K.; Bru¨ckner, R. Chem. Eur. J. 2005, 11, 1610.
(b) Hopf, H.; Kreutzer, M. Angew. Chem., Int. Ed. 1990, 29, 393.
(8) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
3348
Org. Lett., Vol. 10, No. 15, 2008

with the TMS group at C-4 (19%), followed by desilylation.
Conveniently, the bulky TMS group controls the regiochemistry
of the reaction, which is highly selective (although low yielding)
for the 1,5-donor-acceptor-substituted 1,2,3-triazole product.13
Protodesilylation was accomplished under acidic conditions to
give 2 in high yield (78%). The regiochemistries of 1 and 2
are confirmed by the characteristic 1H NMR shifts for positions
C-4 and C-514 as well as by the X-ray crystal structures of the
precursors 11 (Figure 2) and 12 (see Supporting Information).
by the addition/elimination reaction of DMA-substituted
1,2,3-triazole (15) with (chloromethylidene)propanedinitrile
(16) (63%). The predominance of the 2H tautomer of the
1,2,3-triazole reactant18 and the corresponding stability of 6
over 3 and 4 according to calculations may explain this high
regioselectivity. The synthesis of isomer 7 was not attempted.
The UV/Vis spectra in CH2Cl2 of 1, 2, 5, 6 and 11 are
given in Figure 3.
Figure 3. UV/Vis spectra of 1 (black), 2 (red), 5 (blue), 6 (orange),
and 11 (purple) in CH2Cl2 at c ) 5 × 10-5 M.
Figure 2. ORTEP plot of crystal structure of 11 at 220 K, arbitrary
numbering. Atomic displacement parameters are drawn at the 50%
probability level.
As solids, isomers 1, 2 and 6 are red-orange in color, while
5 is brown. The spectra of all four isomers feature broad
CT absorptions of modest to strong intensity with maxima
between λmax ) 400 and 453 nm. Isomer 1 has its longest
wavelength CT absorption at λmax ) 400 nm (3.10 eV, ꢀ )
8.1 × 103 M-1 cm-1), while 2 absorbs at λmax ) 433 nm
(2.86 eV, ꢀ ) 2.5 × 103 M-1 cm-1). Silylated 11 has a
longest wavelength absorption at λmax ) 421 nm (2.94 eV,
ꢀ 1.7 × 103 M-1 cm-1), about 10 nm hypsochromatically
shifted compared to the desilylated analogue 2. Isomer 2 has
a B3LYP/6-31G(d)-optimized geometry where the DCV
moiety is nearly coplanar with the 1,2,3-triazole ring.19 The
gas-phase calculated and X-ray crystal structures of 11
compare well and are both nonplanar due to strong repelling
close contacts between the DCV and TMS groups. The X-ray
structure of Figure 2 shows the large out-of-plane twisting
of the DCV moiety with a torsion angle C4-C5-C6-C7
of 128°. (DMA torsion angle N2-N1-C16-C17 ) -57°).
Despite these geometric effects, the intramolecular CT is not
much affected. The conjugation pathway for isomer 5, which
avoids conjugation through the N-1 position, shows the
strongest CT band at λmax ) 423 nm (2.93 eV, ꢀ ) 11.3 ×
103 M-1 cm-1). Isomer 6, with the acceptor attached to N-2,
presents the longest-wavelength absorption of this series at
λmax) 453 nm (2.74 eV, ꢀ ) 7.3 × 103 M-1 cm-1). In all
The two isomers 3 and 4 were ultimately unobtainable
through the “click” or the thermal Huisgen cycloaddition
methodology. The most obvious approach involving reaction
of 4-ethynyl-N,N-dimethylaniline15 (13) with (azidometh-
ylidene)propanedinitrile16 in the presence (or absence) of
Cu(I) did not afford the expected target compounds 3 or 4,
even after the conditions were varied extensively.
Isomer 5 was obtained from alkyne 1417 by reaction with
NaN3 in DMF (61%) (Scheme 1). The tautomeric form with
hydrogen at N-2 is most stable according to calculations at
the B3LYP/6-31G(d) level of theory. Isomer 6 was obtained
(12) (a) Moonen, N. N. P.; Pomerantz, W. C.; Gist, R.; Boudon, C.;
Gisselbrecht, J.-P.; Kawai, T.; Kishioka, A.; Gross, M.; Irie, M.; Diederich,
F. Chem. Eur. J. 2005, 11, 3325. (b) Bures, F.; Schweizer, W. B.; May,
J. C.; Boudon, C.; Gisselbrecht, J.-P.; Gross, M.; Biaggio, I.; Diederich, F.
Chem. Eur. J. 2007, 13, 5378. (c) Meier, H.; Ju¨rgen, G.; Kolshorn, H.;
Mu¨hling, B. Chem. Eur. J. 2004, 10, 360. (d) Spitler, E. L.; Monson, J. M.;
Haley, M. M. J. Org. Chem. 2008, 73, 2211.
(13) Zanirato, P. J. Chem. Soc., Perkin Trans. 1 1991, 2789.
(14) (a) Alonso, G.; Garcia-Lopez, M. T.; Garcia-Munor, G.; Madronera,
R.; Rico, M. J. Heterocycl. Chem. 1970, 7, 1270. (b) Tsypin, G. I.;
Timofecva, G. N.; Mel’nikov, V. V.; Gidaspov, B. V. Zh. Org. Khim. 1975,
11, 1395.
(15) Leonard, K. A.; Nelen, M. I.; Anderson, L. T.; Gibson, S. L.; Hilf,
R.; Detty, M. R. J. Med. Chem. 1999, 42, 3942.
(16) Josey, A. D.; Dickinson, C. L.; Dewhirst, K. C.; McKusick, B. C.
J. Org. Chem. 1967, 32, 1941.
(18) Oziminski, W. P.; Dobrowolski, J. C.; Mazurek, A. P. J. Mol. Struct.
2003, 651-653, 697.
(17) (a) May, J. C.; Lim, J. H.; Biaggio, I.; Moonen, N. N. P.; Michinobu,
T.; Diederich, F. Opt. Lett. 2005, 30, 3057. (b) Michinobu, T.; May, J. C.;
Lim, J. H.; Boudon, C.; Gisselbrecht, J.-P.; Seiler, P.; Gross, M.; Biaggio,
I.; Diederich, F. Chem. Commun. 2005, 737.
(19) For an X-ray crystal structure of a planar 1,5-substituted triazole,
see: Brunner, M.; Maas, G.; Klarner, F.-G. HelV. Chim. Acta 2005, 88,
1813.
Org. Lett., Vol. 10, No. 15, 2008
3349

cases, the CT nature of these absorptions is confirmed by
quenching with trifluoroacetic acid and reconstitution of the
absorption upon addition of triethylamine.
analogous molecules with ethenediyl or ethynediyl as π-link-
ers have CT absorptions at λmax ) 490 nm (2.53 eV, ꢀ )
4.6 × 104 M-1 cm-1)20 and λmax ) 477 nm (2.60 eV, ꢀ )
4.4 × 104 M-1 cm-1), respectively.17b That 1,2,3-triazole
linkers only induce moderate shifts of the CT absorption
relative to these conventional olefinic and acetylenic refer-
ences is promising for their future applications.
The calculated transition energies of 1-7 and 11 at the
TD B3LYP/6-31G(d) level compare well with the set of
experimental values (see Supporting Information for full
details). There is a standard deviation (SD) of only 0.6 eV
between experimental and calculated values. The calculated
oscillator strengths (f) correlate well with the measured molar
extinction coefficients except for 6, which is predicted to be
more intense than as measured. All longest wavelength
absorption are composed of only HOMO to LUMO transi-
tions with a clear shift of electron density from the donor to
the acceptor according to molecular orbital analysis. Both
the HOMO and LUMO have considerable density on the
1,2,3-triazole ring as well (see images in abstract and
Supporting Information). Within the series of seven possible
isomers, 3 is predicted to have the lowest energy absorption
at 560 nm (2.21 eV, f ) 0.16), while substantially batho-
chromically shifted CT bands are also expected for 4 and 7,
with predicted absorptions at 468 nm (2.65 eV, f ) 0.11)
and 481 nm (2.58 eV, f ) 0.39), respectively.
For all isomers, except 5, the 1,2,3-triazole makes a
particularly intriguing linker where the CT path is “inter-
rupted” by a tertiary amine center at N-1, which in the CT
resonance structure will take on cationic character as a
quaternary ammonium center. Despite this “interruption,” it
is clear that CT can occur and may even be quite efficient
depending on the substitution pattern. It has recently been
noticed in our group and by others,12 that for a series of
donor-substituted cyanoethynylethenes, the longest-wave-
length CT absorptions are found for systems with inefficient
ground-state donor-acceptor conjugation. At weak conjuga-
tion, HOMO and LUMO energies remain close to those of
the individual, undisturbed donor and acceptor components,
respectively, as clearly revealed in electrochemical studies.
Comparison of the energies of the calculated HOMO and
LUMO levels at the B3LYP/6-31G(d) level of isomers 1-7
is not conclusive, as both the HOMO (SD ) 0.2 eV) and
LUMO (SD ) 0.1 eV) energies vary significantly making it
difficult to attribute the HOMO energies directly to the
predicted HOMO-LUMO gap. However, 3, which has the
longest wavelength absorption according to calculation, also
has the least negative HOMO energy at -5.4 eV (LUMO:
-2.9 eV) while the HOMO energy of isomer 1, with the
shortest predicted wavelength absorption, is considerably
more negative at -5.8 eV (LUMO: -2.6 eV).
In summary, except for 2, donor-acceptor-substituted
1,2,3-triazoles were prepared by rather indirect routes instead
of by using azide cycloaddition to an appropriate alkyne
under thermal or “click” conditions. The production of CT
chromophores, which must employ the requisite polarized
reactants, pushes beyond the current limit of the “click”
methodology. While, 4-aminophenylacetylenes and 4-ami-
nophenylazides are fairly common reactants for “click”
chemistry, strongly electron poor vinylic azides are rare and
electron deficient enynes have been attempted only in this
work. The complications that arise from the numerous side
reactions available to these systems may explain their lack
of use in “click” chemistry. The positioning of the donor
and acceptor on the 1,2,3-triazole ring in 1-7 is ultimately
responsible for about a 50 nm range (experimentally;
computationally: 100 nm) of transition energies and also
affects strongly the absorption intensity. To further motivate
research toward formally “clicked” CT chromophores,
electrochemical and NLO studies are currently underway on
the isomers studied here and related systems. Both the
synthetic approach and materials properties of 1,2,3-triazol-
diyl linked donor-acceptor chromophores are being opti-
mized. Exploration of their potency to form high-quality
optical films by vapor-phase deposition is of particular
interest.
Acknowledgment. We are grateful for financial support
provided by the ETH Research Council and by the Novartis
M.Sc. Fellowship (to Y.-L.W.). We are grateful for access
to the Competence Center for Computational Chemistry (C4)
Obelix cluster (ETH) where computational work was per-
formed.
Supporting Information Available: Optimized geom-
etries and TD parameters from B3LYP/6-31G(d) calculations,
complete Gaussian 03 reference, depiction of HOMO and
LUMO orbitals and energies, ORTEP plots of X-ray
structures and CIF files, and experimental procedures and
analytical characterization of all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Isomer 5 has a CT path that can avoid entirely the
N-(1,2,3) positions, but nonetheless does not show superior
CT in terms of the position of the absorption maximum
compared to 2, 6 and 11 and the calculated 3, 4 and 7,
although it has by far the most intense absorption. The
OL801253Z
(20) Staub, K.; Levina, G. A.; Barlow, S.; Kowalczyk, T. C.; Lackritz,
H.; Barzoukas, M.; Fort, A.; Marder, S. R. J. Mater. Chem. 2003, 13, 825.
3350
Org. Lett., Vol. 10, No. 15, 2008