Synthesis and optical properties of trioxatriangulenium dyes with one and two peripheral amino substituents

Article abstract of DOI:10.1021/jo1009917

Figure presented. Several derivatives of two new dye systems, with one or two dialkylamino donor groups attached to resonant positions at the periphery of a trioxatriangulenium ion, were synthesized. The mono- and bis-dialkylamino trioxatriangulenium salts (A₁-TOTA⁺ and A ₂-TOTA⁺) were prepared from methoxy-substituted triphenylmethylium (TPM) compounds by aromatic nucleophilic substitution with secondary amines and subsequent intramolecular ring closure. The optical properties of the new triangulenium dyes and their TPM precursors were investigated and compared to those of known TPM and xanthenium dyes. The optical properties were found to be dependent on symmetry and charge localization in the conjugated framework. The trioxatriangulenium dye with two amino groups (A₂-TOTA⁺) was found to be a strong fluorophore with properties as a blue-shifted rhodamine B. The mono-substituted compound (A ₁-TOTA⁺) was found to be only weakly fluorescent. We assign the weak fluorescence of A₁-TOTA⁺ to an efficient but reversible formation of a nonfluorescent conformation in the excited state, favored by the large degree of charge localization in this dye with only one donor group.

Full text of DOI:10.1021/jo1009917

pubs.acs.org/joc
Synthesis and Optical Properties of Trioxatriangulenium Dyes with
One and Two Peripheral Amino Substituents
Thomas J. Sørensen and Bo W. Laursen*
Nano-Science Center and Department of Chemistry, University of Copenhagen, Universitetsparken 5,
DK-2100 København Ø, Denmark
bwl@nano.ku.dk
Received June 2, 2010
Several derivatives of two new dye systems, with one or two dialkylamino donor groups attached to
resonant positions at the periphery of a trioxatriangulenium ion, were synthesized. The mono- and
bis-dialkylamino trioxatriangulenium salts (A1-TOTAþ and A2-TOTAþ) were prepared from
methoxy-substituted triphenylmethylium (TPM) compounds by aromatic nucleophilic substitution
with secondary amines and subsequent intramolecular ring closure. The optical properties of the new
triangulenium dyes and their TPM precursors were investigated and compared to those of known
TPM and xanthenium dyes. The optical properties were found to be dependent on symmetry and
charge localization in the conjugated framework. The trioxatriangulenium dye with two amino
groups (A2-TOTAþ) was found to be a strong fluorophore with properties as a blue-shifted
rhodamine B. The mono-substituted compound (A1-TOTAþ) was found to be only weakly fluorescent.
We assign the weak fluorescence of A1-TOTAþ to an efficient but reversible formation of a
nonfluorescent conformation in the excited state, favored by the large degree of charge localization
in this dye with only one donor group.
Introduction
on the nature of the central atom. Figure 1 shows the charged
triangulenium5-10 compounds (A) where the central atom is
carbon, and the heterotriangulenes (B) with nitrogen11-16 or
phosphor17,18 inthe central position. Trianguleniumcompounds
The triangulene (4,8,12-trioxa-4,8,12,12c-tetrahydrodi-
benzo[cd,mn]pyrene) compounds share a structure consisting
of six aromatic rings arranged in a triangular fashion.1,2 The all-
carbon triangulene is not a stable molecule but can exist as
either biradical or dianion.3,4 The synthesized triangulenes con-
tain heteroatoms and can be divided into two groups depending
(8) Laursen, B. W.; Krebs, F. C.; Nielsen, M. F.; Bechgaard, K.;
Christensen, J. B.; Harrit, N. J. Am. Chem. Soc. 1998, 120, 12255.
(9) Laursen, B. W.; Krebs, F. C. Angew. Chem., Int. Ed. 2000, 39, 3432.
(10) Laursen, B. W.; Krebs, F. C. Chem.;Eur. J. 2001, 7, 1773.
(11) Hellwinkel, D.; Melan, M. Chem. Ber. 1971, 104, 1001.
(12) Hellwinkel, D.; Melan, M. Chem. Ber. 1974, 107, 616.
(13) Hellwinkel, D.; Aulmich, G.; Melan, M. Chem. Ber. 1981, 114, 86.
(14) Field, J. E.; Venkataraman, D. Chem. Mater. 2002, 14, 962.
(15) Kuratsu, M.; Kozaki, M.; Okada, K. Angew. Chem., Int. Ed. 2005,
44, 4056.
(1) Clar, E.; Stewart, D. G. J. Am. Chem. Soc. 1953, 75, 2667.
(2) Clar, E.; Stewart, D. G. J. Am. Chem. Soc. 1954, 76, 3504.
(3) Hara, O.; Tanaka, K.; Yamamoto, K.; Nakazawa, T.; Murata, I.
Tetrahedron Lett. 1977, 2435.
(4) Allinson, G.; Bushby, R. J.; Paillaud, J. L. J. Mater. Sci.: Mater.
Electron. 1994, 5, 67.
(5) Martin, J. C.; Smith, R. G. J. Am. Chem. Soc. 1964, 86, 2252.
(6) Hellwinkel, D.; Melan, M.; Aulmich, G. Tetrahedron Lett. 1976, 17,
4137.
(16) Zhang, H. Q.; Wang, S. M.; Li, Y. Q.; Zhang, B.; Du, C. X.; Wan,
X. J.; Chen, Y. S. Tetrahedron 2009, 65, 4455.
(17) Krebs, F. C.; Larsen, P. S.; Larsen, J.; Jacobsen, C. S.; Boutton, C.;
Thorup, N. J. Am. Chem. Soc. 1997, 119, 1208.
(7) Neugebauer, F. A.; Hellwinkel, D.; Aulmich, G. Tetrahedron Lett.
1978, 4871.
(18) Faldt, A.; Krebs, F. C.; Thorup, N. J. Chem. Soc., Perkin Trans. 2
1997, 2219.
6182 J. Org. Chem. 2010, 75, 6182–6190
Published on Web 08/26/2010
DOI: 10.1021/jo1009917
r
2010 American Chemical Society

Sørensen and Laursen
JOCArticle
excellent fluorescent dyes.8-10,27,40-44 The azaoxa-triangu-
lenium dyes, with no donor group in the position para to the
formal cation center (X, Y, Z = hydrogen), are medium
strength fluorophores, where triangulenium compounds with
strong donor groups on the periphery are strong absorbers and
excellent fluorescent dyes with high brigthness.8,27,43,45 Tris-
(dialkylamino)-trioxatriangulenium (X, Y, Z = dialkylamino)
and trihydroxy-trioxatriangulenium (X, Y, Z = hydroxyl) are
3-fold symmetric extensions of rhodamines46/rosamines47and
fluorescein,48 respectively. Fluorescien can be considered a
hetero bridged version of a triphenylmethylium (TPM) dye
(phenolphthalein). The same is true for rhodamines and for
tris(dialkylamino)-trioxatriangulenium. The latter is a hetero
bridged version of crystal violet, where the bridging of all ortho
positions provides rigidity and planarity, giving excellent fluo-
rescence properties. Tris(dialkylamino)-trioxatriangulenium has
3-fold symmetry. This introduces degenerate transitions that give
characteristic absorption properties.42,43 So far, all the donor
substituted triangulenium derivatives synthesized have been
3-fold symmetric, with either three oxygen or three nitrogen
donor groups at the periphery (X, Y, Z positions). Here, the
synthesis and fluorescence properties of two new asymmetric
amino-trioxatriangulenium salts, with only one or two nitrogen
donor groups, are reported.
FIGURE 1. Planar triangulenium ions (A) and bowl-shaped het-
erotriangulenes (B). A: K, L, M can be nitrogen or oxygen; X, Y, Z
can be hydrogen, alkyl, hydroxy, or dialkylamino. B: W can be
nitrogen or phosphorus; K, L, M can be oxo groups or oxygen.
have also been synthesized with saturated carbon and oxo
groups in the bridge positions.1,4,7,19-21
The triangulene compounds are highly symmetric, which
makes the electronic structure interesting.4,7,19-24 The phy-
sical chemical properties of the triangulene compounds have
been investigated, with focus on crystal packing8,10,18,25,26
and cation stability.8,10,27 Because of the planar and rigid
triangular structure, triangulene compounds has found use
in supramolecular assemblies,28-33 as chiral platforms,33-35
and as DNA intecalators.36-39
The charged triangulenium compounds (A in Figure 1)
have been shown to be extremely stable carbocations and
Recent results from the trihydroxy-trioxatrianguelnium45 sys-
tem showed interesting, symmetry dependent behavior, and it
was decided to synthesize the dialkylamino-trioxatriangulenium
(8, A1-TOTAþ, 2-dialkylamino-4,8,12-trioxa-4,8,12,12c-tetra-
hydro-dibenzo[cd,mn]-pyrenylium) and bis(dialkylamino)-tri-
oxatriangulenium (9, A2-TOTAþ, 2,6-bis(dialkylamino)-4,8,12-
trioxa-4,8,12,12c-tetrahydro-dibenzo[cd,mn]-pyrenylium) dyes,
the structures of which are shown in Figure 2. The synthesis
was adapted from the synthetic route used in the preparation of
tris(dialkylamino)-trioxatriangulenium (A3-TOTAþ, 2,6,10-tris-
(dialkylamino)-4,8,12-trioxa-4,8,12,12c-tetrahydro-dibenzo[cd,
mn]-pyrenylium).8,27,28 The intermediates in these syntheses
are triphenylmethylium dyes with one or two para amino groups
similar to malachite green and sunset orange,49,50 yet with all
ortho positions carrying methoxy groups. These new hexa-
methoxy-TPM dyes are isolated in the synthesis of A1-TOTAþ
and A2-TOTAþ andallowedforacomparisonofthespectroscopic
properties of TPM dyes with a varying number and type of donor
groups. While the A2-TOTAþ target molecules show good fluo-
rescent dye properties, comparable to those of rhodamine, the
A1-TOTAþ dyes display unexpected excited state properties.
(19) Allinson, G.; Bushby, R. J.; Paillaud, J. L.; Oduwole, D.; Sales, K.
J. Am. Chem. Soc. 1993, 115, 2062.
(20) Allinson, G.; Bushby, R. J.; Paillaud, J. L.; Thorntonpett, M.
J. Chem. Soc., Perkin Trans. 1 1995, 385.
(21) Allinson, G.; Bushby, R. J.; Jesudason, M. V.; Paillaud, J. L.; Taylor,
N. J. Chem. Soc., Perkin Trans. 2 1997, 147.
€
(22) Muller, E.; Moosmaye., A; Rieker, A.; Scheffle., K Tetrahedron Lett.
1967, 3877.
(23) Kessler, H.; Moosmaye., A; Rieker, A. Tetrahedron 1969, 25, 287.
(24) Sabacky, M. J.; Johnson, C. S.; Smith, R. G.; Gutowsky, H. S.;
Martin, J. C. J. Am. Chem. Soc. 1967, 89, 2054.
(25) Krebs, F. C.; Laursen, B. W.; Johannsen, I.; Faldt, A.; Bechgaard,
K.; Jacobsen, C. S.; Thorup, N.; Boubekeur, K. Acta Crystallogr., Sect. B:
Struct. Sci. 1999, 55, 410.
(26) Krebs, F. C. Tetrahedron Lett. 2003, 44, 17.
(27) Laursen, B. W.; Sørensen, T. J. J. Org. Chem. 2009, 74, 3183.
(28) Simonsen, J. B.; Kjaer, K.; Howes, P.; Nørgaard, K.; Bjørnholm, T.;
Harrit, N.; Laursen, B. W. Langmuir 2009, 25, 3584.
(29) Lofthagen, M.; Chadha, R.; Siegel, J. S. J. Am. Chem. Soc. 1991, 113,
8785.
(30) Lofthagen, M.; Vernonclark, R.; Baldridge, K. K.; Siegel, J. S. J.
Org. Chem. 1992, 57, 61.
(31) Lofthagen, M.; Siegel, J. S. J. Org. Chem. 1995, 60, 2885.
(32) Baisch, B.; Raffa, D.; Jung, U.; Magnussen, O. M.; Nicolas, C.;
Lacour, J.; Kubitschke, J.; Herges, R. J. Am. Chem. Soc. 2009, 131, 442.
(33) Narasimhan, S. K.; Kerwood, D. J.; Wu, L.; Li, J.; Lombardi, R.;
Freedman, T. B.; Luk, Y.-Y. J. Org. Chem. 2009, 74, 7023.
(34) Narasimhan, S. K.; Lu, X.; Luk, Y. Y. Chirality 2008, 20, 878.
(35) (a) Mobian, P.; Banerji, N.; Bernardinelli, G.; Lacour, J. Org.
Biomol. Chem. 2006, 4, 224. (b) Mobian, P.; Nicolas, C.; Francotte, E.;
Results and Discussion
Synthesis. The previously reported synthesis of A3-TOTAþ
salts is based on the unique chemistry of the tris(2,4,6-
trimethoxyphenyl)-methylium ion ((TMP)3Cþ), which due to
the steric shielding of the central carbenium atom by the six o-
methoxy groups, undergoes aromatic nucleophilic substitution
(SNAr) in the para positions (Scheme 1).8,27,51 In the final step,
€
Burgi, T.; Lacour, J. J. Am. Chem. Soc. 2008, 130 (20), 6507–6514.
(36) Pothukuchy, A.; Tuntiwechapikul, W.; Gopidas, K. R.; Salazar, M.
Abstr. Papers Am. Chem. Soc. 2002, 224, 194.
(37) Pothukuchy, A.; Ellapan, S.; Gopidas, K. R.; Salazar, M. Bioorg.
Med. Chem. Lett. 2003, 13, 1491.
(38) Pothukuchy, A.; Mazzitelli, C. L.; Rodriguez, M. L.; Tuesuwan, B.;
Salazar, M.; Brodbelt, J. S.; Kerwin, S. M. Biochemistry 2005, 44, 2163.
(39) Reynisson, J.; Schuster, G. B.; Howerton, S. B.; Williams, L. D.;
Barnett, R. N.; Cleveland, C. L.; Landman, U.; Harrit, N.; Chaires, J. B.
J. Am. Chem. Soc. 2003, 125, 2072.
(40) Dileesh, S.; Gopidas, K. R. Chem. Phys. Lett. 2000, 330, 397.
(41) Dileesh, S.; Gopidas, K. R. J. Photochem. Photobiol., A 2004, 162,
115.
(45) Westerlund, F.; Hildebrandt, Christoffer B.; Sørensen, Thomas J.;
Laursen, Bo W. Chem.;Eur. J. 2010, 16, 2992.
(46) Drexhage, K. H. Top. Appl. Phys. 1973, 1, 144.
(47) Wu, L. X.; Burgess, K. J. Org. Chem. 2008, 73, 8711.
(48) Baeyer, A. Ber. Deutch. Chem. Ges. 1871, 4, 555.
(49) Duxbury, D. F. Chem. Rev. 1993, 93, 381.
(42) Reynisson, J.; Wilbrandt, R.; Brinck, V.; Laursen, B. W.; Norgaard,
K.; Harrit, N.; Brouwer, A. M. Photochem. Photobiol. Sci. 2002, 1, 763.
(43) Laursen, B. W.; Reynisson, J.; Mikkelsen, K. V.; Bechgaard, K.;
Harrit, N. Photochem. Photobiol. Sci. 2005, 4, 568.
(50) Nair, V.; Thomas, S.; Mathew, S. C.; Abhilash, K. G. Tetrahedron
2006, 62, 6731.
(44) Sørensen, T. J.; Laursen, B. W.; Luchowski, R.; Shtoyko, T.;
Akopova, I.; Gryczynski, Z.; Gryczynski, I. Chem. Phys. Lett. 2009, 476, 46.
(51) Wada, M.; Konishi, H.; Kirishima, K.; Takeuchi, H.; Natsume, S.;
Erabi, T. Bull. Chem. Soc. Jpn. 1997, 70, 2737.
J. Org. Chem. Vol. 75, No. 18, 2010 6183

Sørensen and Laursen
JOCArticle
FIGURE 2. Amino-triangulenium dyes: trioxatriangulenium (TOTAþ), mono(dialkylamino)-trioxatriangulenium (A1-TOTAþ, 8), bis-
(dialkylamino)-trioxatriangulenium (A2-TOTAþ, 9), and tris(dialkylamino)-trioxatriangulenium (A3-TOTAþ).
SCHEME 1. General Synthetic Route to Amino-trioxatriangulenium Dyes
the same o-methoxy groups form the oxygen bridges in the
trioxatriangulenium system, by a ring-closure reaction facili-
tated by an ether cleaving reagent (e.g., LiI). In order to apply
this strategy in the synthesis of the two asymmetric amino-
trioxatriangulenium dyes A1-TOTAþ (8) and A2-TOTAþ (9),
the heptamethoxy- and octamethoxy-substituted TPM salts 3
and 4 were required.
SCHEME 2. Synthesis of 2,6-Dimethoxybenzoic Acid Ethyl
Ester (1) and 2,4,6-Trimethoxybenzoic Acid Ethyl Ester (2)
The synthesis of the starting materials, heptamethoxy-triphe-
nylmethylium tetrafluoroborate (3, bis(2,6-dimethoxyphenyl)-
(2,4,6-trimethoxyphenyl)-methylium BF4) and octamethoxy-
3
triphenylmethylium tetrafluoroborate (4, (2,6-dimethoxyphe-
nyl)-bis(2,4,6-trimethoxyphenyl)-methylium BF4), differs from
3
the synthesis of the symmetric nonamethoxy-triphenylmethy-
lium((TMP)3Cþ) starting material used to make A3-TOTAþ,as
they are of lower symmetry.5,8,51,52
The syntheses of the symmetric hexa- and nona-methoxy
TPM compounds were achieved via lithiation of the corre-
sponding di- or trimethoxy benzene and reaction with an
appropriate carbonyl compound, e.g., diethyl carbonate, to
yield a triphenylcarbinol, which in turn can be isolated5,51 or
directly reacted to the triphenylmethylium salts.8,10 To ob-
tain the asymmetric TPM compounds 3 and 4, the carbonate
was exchanged for di- or trimethoxy ethyl benzoate 1 or 2,
respectively. The simple syntheses of these compounds are
based on quenching of the lithiated di- and trimethoxyben-
zenes with excess diethylcarbonate (Scheme 2). From 2 and 1
respectively, 3 and 4 are prepared as shown in Scheme 3. The
crude products are obtained in good yields and used in the sub-
sequent reactions steps without further purification. If 99.9%
pure materials are needed the overall yields drop to 17% and
36% for 3 and 4, respectively, due to loss of material in the
consecutive recrystallizations.
SCHEME 3. Synthesis of Heptamethoxy-triphenylmethylium
Salt 3 and Octamethoxy-triphenylmethylium Salt 4
The products from aromatic nucleophilic substitution reac-
tions, where the p-methoxy groups in 3, 4, and (TMP)3Cþ is
exchanged with dialkyl amines, are analogues of well-known
TPM dyes, with the addition of six o-methoxy groups. The
symmetrically substituted tris(dialkylamino)-hexamethoxy-
TPMþ compound can be considered a crystal violet deri-
vate. Crystal violet is tris(4-dimethylaminophenyl)-me-
thylium, and tris(dialkyl)amino-hexamethoxy-TPMþ is
(52) Morita, M.; Ohmi, T.; Hasegawa, E.; Kawakami, M.; Ohwada, M.
J. Appl. Phys. 1990, 68, 1272.
6184 J. Org. Chem. Vol. 75, No. 18, 2010

Sørensen and Laursen
JOCArticle
SCHEME 4. Synthesis of Monoamino-hexamethoxy-triphe-
nylmethylium Salt 5, Diamino-hexamethoxy-triphenylmethylium
Salt 6, and Amino-heptamethoxy-triphenylmethylium Salt 7
SCHEME 5. Synthesis of Dialkylamino-trioxatriangulenium
PF6 (8) and Bis(dialkylamino)-trioxatriangulenium PF6 (9)
3
3
method failed to produce the target molecules. The target
molecules were formed according to MALDI-TOF MS, but
before the reaction had run to completion (with consumption of
all the starting material and intermediates), the target com-
pound was converted into unidentified decomposition products.
Using MALDI-TOF MS as monitoring technique, several
reaction conditions were tested, including LiI/NMP, LiCl/
NMP, LiI/NMP/Collidine, LiCl/Collidine, LiI/MeCN, and
HCl/AcOH. All combinations of nucleophile/solvent were
tested at various temperatures. No reaction conditions were
found that fully solved the issues. The best conditions found
were LiI in NMP at 140 °C with 10-50% collidine as cosolvent,
resulting in moderate to low yields as shown in Scheme 5. In all
cases the methyl derivatives 5c and 6c decomposed much faster
than the ethyl and decyl derivatives and no fully ring-closed
triangulenium compounds were obtained from these com-
pounds. This observation suggests that dealkylation is the first
step in the decomposition,8,28 as the methyl groups are cleaved
from anilines much more quickly by nucleophiles than, e.g.,
ethyl groups.57 The positive effect of collidine suggests that in situ
methylation by methyl iodide generated in the ring closure also
may contribute to the degradation.27 The main pathway of
degradation in the amino-trioxatriangulenium compounds is
through dealkylation of the aminogroups. The large degree of
decomposition in 8 and 9, compared to A3-TOTAþ, is believed
to be due to a larger partial positive charge on the amino groups
in these systems. The fewer donor groups increases the charge
on the amino groups in 8and 9, which leads to a higher reactivity
in nucleophilic dealkylation reactions. Thus, the chemical sta-
bility of the amino-trioxatriangulenium dyes increase as the
number of donor groups is increased. This also explains why 8a
is less stable than 9a. The apparent low yield of 9b is not due to
chemical degradation but is a result of difficult purification (see
Experimental Section).
tris(4-dialkylamino-2,6-dimethoxyphenyl)-methylium. The bis-
(dialkylamino)-hexamethoxy-TPMþ salt 6 (bis(4-dialkylamino-
2,6-trimethoxyphenyl)-(2,6-dimethoxyphenyl)-methylium) is
by analogy a malachite green (bis(4-dimethylaminophenyl)-
phenylmethylium) derivative, and the monoamino-hexa-
methoxy-TPMþ derivative 5 (4-dialkylamino-2,6-dimethoxy-
phenyl)-bis(2,6-dimethoxyphenyl)-methylium) is a sunset
orange (4-dimethylaminophenyl-diphenyl-methylium)-type
compound.53-55 Several derivatives of the new trioxatriangule-
nium dyes 5and 6were obtained by reaction of 3and 4with vari-
ous secondary amines in acetonitrile solutions as outlined in
Scheme 4. The SNAr reaction is practically quantitative for the
more nucleophile amines, but difficult purification results in
moderate yields for some derivatives (28-79%).
The driving force in the SNAr reactions is the cationic
nature of the TPM substrates, and reactivity decreases as
more and stronger stabilizing donor groups are introduced.8
The reactivity of 3 follows this trend in the sense that the first,
and only, substitution is fast even with low nucleophilic
didecylamine.28 Compound 4 is less reactive and the differ-
ence in reactivity between first and second substitution
allows for preparation of the intermediate product 7. In all
reactions the very strong MALDI-TOF mass spectrometric
signals,10,56 from the stable and permanently charged carbe-
nium ions, were use to monitor the progress of the reactions
and to determine when to work up the desired products.
The amino-trioxatriangulenium target molecules (8 and 9)
were obtained from 5 and 6 by a ring-closure reaction.
Initially the conditions reported for triamino-trioxatriangu-
lenium system were tried, using LiI in NMP.8 However, this
Optical Properties of Triphenylmethylium Derivatives. Tri-
phenylmethylium (TPM) dyes are nonfluorescent, potent dyes
and colorants.49,58 Figure 3 shows the absorption spectra of the
hexamethoxy-TPM dyes with one to three p-methoxy or nitro-
gen donor groups. The data from the spectra are compiled in
Table 1 that gives the symmetry, absorption maximum, molar
absorption coefficient, and para substituent pattern of the
(53) Holmes, E. O. J. Phys. Chem. 1966, 70, 1037.
(54) Holmes, E. O. J. Phys. Chem. 1969, 73, 273.
(55) Kubin, R. F.; Fletcher, A. N. J. Lumin. 1982, 27, 455.
(56) Denekamp, C.; Lacour, J.; Laleu, B.; Rabkin, E. J. Mass Spectrom.
2008, 43, 623.
(57) Chambers, R. A.; Pearson, D. E. J. Org. Chem. 1963, 28, 3144.
(58) Zollinger, H. Color Chemistry; 3rd ed.; Wiley VCH: New York, 2001.
J. Org. Chem. Vol. 75, No. 18, 2010 6185

Sørensen and Laursen
JOCArticle
individual dyes. For comparison, literature values for analo-
gous TPM dyes without the o-methoxy groups are compiled in
Table 2.
three phenyl substituents to be coplanar simultaneously. This
competition between steric and electronic factors may lead to
decoupling of phenyl groups with weak or no donor groups, in
favor of more coplanarity with the stronger donor groups. Such
conformations will results in an effective reduction of the length/
size of the chromophore unit and consequently a blue shift of the
first electronic transition. This is exactly what is observed for the
TPM dyes with only one donor-substituted phenyl group.
Thus, the first electronic transitions in compounds 3, 5, and
sunset orange display significantly lower absorption coeffi-
cients and higher transition energies compared to that of the
(DMP)3Cþ system with no para donor groups. In the latter,
symmetry favors a π-system conjugated over the entire mole-
cule. In the compounds with two and three donor groups, the
conjugation pathway is extended to at least two phenyl arms,
and a high absorption coefficient and low transition energy is
observed. As expected, a higher absorption coefficient and a
lower transition energy is observed when going from weaker
methoxy to the stronger nitrogen electron donor.59
In order to understand the spectral properties of these
dyes, it is in each case necessary to identify the active
chromophore unit. For these systems the chromophore units
consist of the cationic center and between one and three of
the donor-substituted phenyl rings. A coplanar orientation
of the sp2 hybridized central carbon atom and the phenyl
rings will provide optimal charge delocalization and reso-
nance stabilization. However, steric interactions prevent all
The effect of the o-methoxy groups can be estimated by
comparison of Tables 1 and 2. The direct effect is a small red
shift of the absorption maxima and a factor 2 decrease in the
absorption coefficient. The decrease can be explained by the
higher twist of the conjugation pathway, lowering the transi-
tion probability. Within each type of donor groups the trends
are similar and exemplified by the compounds in Table 2.
The absorption maximum does not change much between
two and three donor groups, as the length of the chromo-
phore unit is unchanged. However, the absorption coeffi-
cient still increases, as a new degenerate transition is added in
the dyes with 3-fold symmetry.
Optical Properties of Amino-trioxatriangulenium Deriva-
tives. The normalized absorption spectra of the amino-trioxa-
triangulenium dyes in dichloromethane are shown in Figure 4.
Trioxatriangulenium (TOTAþ, black curve) and tris(dialkyl-
amino)-trioxatriangulenium (A3-TOTAþ, red curve) have high
symmetry, and the two lowest transitions are, in theory, degene-
rate. The symmetry is lifted in TOTAþ, resulting in a broad
peak arising from two close-laying transitions.42 The spectrum
of A3-TOTAþ shows one degenerate transition.43 Going to the
less symmetric A2-TOTAþ the degenerate transition is split into
FIGURE 3. Absorption spectra of methoxy (top) and dimethyla-
mino (bottom) donor substituted hexamethoxy-triphenylmethy-
lium dyes in acetonitrile. The compounds have 0 para donor
groups (black), 1 donor (gray), 2 donors (blue), and 3 donors (red).
TABLE 1. Spectral Properties of o-Hexamethoxy-triphenylmethylium Dyes
compounda
(DMP)3Cþb
symmetry
solvent
A
B
C
λ
max (nm)
ε (M-1 cm-1
)
D3
C2
C2
D3
C2
C2
D3
C1
C2
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
H
H
H
OMe
H
H
NR2
H
H
H
H
524
491
580
577
457
637
634
512
534
18,400
14,100
18,400
23,400
16,900
40,200
49,400
19,700
27,400
3
4
OMe
OMe
OMe
NR2
NR2
NR2
NR2
NR2
OMe
OMe
H
NR2
NR2
OMe
OMe
(TMP)3Cþc
5
6
(R2N-DMP)3Cþc
7
(R2N-DMP)(TMP)2Cþc
OMe
aDMP = 2,6-dimethoxyphenyl, TMP = 2,4,6-trimethoxyphenyl. bSynthesized as described in ref 10. cSynthesized as described in ref 8.
6186 J. Org. Chem. Vol. 75, No. 18, 2010

Sørensen and Laursen
JOCArticle
TABLE 2. Spectral Properties of Triphenylmethylium Dyes
compound
TPMþ
sunset orange
malachite green
crystal violet
symmetry
solvent
A
B
C
λ
max (nm)
ε (M-1 cm-1
)
D3
C2
C2
D3
HFSO3/SbF5
H2O/AcOH
H2O/AcOH
H2O/AcOH
H
H
H
NR2
H
H
NR2
NR2
H
429a
463b
621a
589b
38,900a
NR2
NR2
NR2
13,300c
105,000a
112,000a
aFrom ref 58. bFrom ref 54. cFrom ref 55.
FIGURE 4. Normalized absorption spectra of the amino-trioxa-
triangulenium dyesindichloromethane:TOTAþ (black), A1-TOTAþ
(gray), A2-TOTAþ (blue), and A3-TOTAþ (red).
two transitions as shown by the blue curve in Figure 4, one blue-
shifted band at 400 nm and one red-shifted at 510 nm. The high
energy transitions borrows intensity from the low energy transi-
tions, resulting in a lower than expected intensity of the main ab-
sorption in A2-TOTAþ. The compound is essentially a rhoda-
mine type chromophore and an absorption coefficient of
80,000-100,000 M-1 cm-1 was expected. The observed absorp-
tion coefficient is 60,000 M-1 cm-1. This is possible in A2-
TOTAþ as the structure is planar. Compared to rhodamines A2-
TOTAþ has two additional hetero bridges, thus allowing for the
high energy band to have significant intensity. In rhodamines/
rosamines the low energy band is all dominating, and usually the
only band observed in the visible region of the spectrum.47,60
A1-TOTAþ similarly shows two bands at 425 and 500 nm,
in this system the high energy band is hidden in the pro-
nounced vibrational structure of the low energy band. The
absorption coefficient is 42,000 M-1 cm-1, comparable to
other small donor-acceptor systems.59
FIGURE 5. Absorption (line), fluorescence excitation (dash), and
emission (dot) spectra of A1-TOTAþ in dichloromethane (top, gray)
and A2-TOTAþ (bottom, blue).
these structures are the obvious chromophores in the two
compounds.
The emission and excitation spectra of A1-TOTAþ and A2-
TOTAþ are shown in Figure 5. The mirror image rule is obeyed
by both species. The excitation and absorption spectra are iden-
tical, albeit the excitation and absorption spectra of A1-TOTAþ
differs in the intensity of the vibrational bands. These observa-
tions show that the observed emission originates from the chro-
mophore responsible for the absorption spectrum and that only
this species is emitting. However, significant deviations from
ordinary fluorophore behavior become clear when the quantum
yields and fluorescence lifetimes of A1-TOTAþ are taken into
account.
A3-TOTAþ is an excellent fluorophore, with optical prop-
erties comparable to those of rhodamine.8,43 The symmetry of
the structure result in two degenerate transitions in this extended
xanthenium system. Consequently it is expected that the less sym-
metric A2-TOTAþ and A1-TOTAþ systems should be regarded
as diamino-xanthenium and monoamino-xanthenium dyes, as
The fluorescence properties of the amino-trioxatriangule-
nium dyes are a compiled in Table 3. To enable comparison,
values for TOTAþ and A3-TOTAþ in dicholoromethane are
included. The unsubstituted system (TOTAþ) is markedly
different, with lower absorption coefficient, longer excited
state lifetime, and overall lower oscillator strength of the
(59) Griffiths, J. Colour and Constitution of Organic Molecules; Academic
Press: New York, 1976.
(60) Brackmann, U. Lambdachrome Laser Dyes; 3rd ed.; Lambda Physik:
Goettingen, Germany, 2000.
J. Org. Chem. Vol. 75, No. 18, 2010 6187

Sørensen and Laursen
JOCArticle
TABLE 3. Photophysical Properties of the Amino-trioxatriangulenium Dyes
compound
TOTAþ
A
B
C
λ
max (nm)
ε (M-1 cm-1
)
λ
fl (nm)
φflc (%)
τ (ns)
τ0 (ns)
DCMa
DCM
MeCN
EtOH
DCM
MeCN
EtOH
DCMb
H
H
H
H
H
NR2
NR2
NR2
NR2
H
H
H
H
H
H
H
NR2
475
507
502
499
513
512
512
471
8,200
41,700
39,800
37,900
59,700
46,000
46,600
132,900
520
529
541
530
544
541
537
494
11
5.1
1.6
2.8
100
44
11.7
4.2
106
82d
A1-TOTAþ
NR2
NR2
NR2
NR2
NR2
NR2
NR2
c
4.03
3.73
4.16
1.34
1.8
250d
140d
4.2e
3.0e
3.3e
5.4
A2-TOTAþ
55
67
A3-TOTAþ
3.6
aFrom ref 42. From ref 43. The error is estimated to be 10% of the nominal value. dThe radiative lifetime assuming no excited state reactions
b
occurring. eThese values are within the error of the quantum yield determination identical.
main transition.42 The addition of one nitrogen donor on the
periphery (A1-TOTAþ) increases the oscillator strength signifi-
cantly, resulting in a molar absorption of 40,000 M-1 cm-1 at
509 nm. However, the emission quantum yield is surprisingly
low (1-5%). A low fluorescence quantum yield and high oscil-
lator strength is expected to result in a short fluorescence
lifetime.61 Yet we find a lifetime (τ) of∼4 ns, resulting in a radia-
tive lifetime (τ0) between 80 and 250 ns, depending on the
solvent (Table 3). The radiative lifetime is much too long for an
optical transition with this intensity. A radiative lifetime of ∼7 ns
is expected on the basis of the absorption intensity, according
to the Strickler-Berg equation.61 To find an explanation for
these unusual properties, we consider that the A1-TOTAþ fluo-
rophore is likely to have similar pathways and rates of non-
radiative energy dissipation as rhodamines.46,62-68 Thus the
difference in absorption and excitation spectrum, the anoma-
lous long radiative lifetime, and low quantum yield is explained
by a very efficient formation of a twisted internal charge transfer
or TICT state on the excited state potential surface.68 The
process is expected to be more favorable in A1-TOTAþ, due
to the presence of only one donor group, which results in a
cation with more charge localization compared to rhodamines
and A3-TOTAþ.27 The good mirror image relationship suggests
that the emission indeed originates from the xanthenium chro-
mophore. When the long radiative lifetime is taken into
account, this emission must be due to a repopulation of the ab-
sorbing and emitting state from the dark TICT state. As the
only degree of freedom present is rotation of the donor group,
the TICT state formation is the most plausible explanation for
the unusual observation of simultaneous long radiative lifetime,
high oscillator strength, and good mirror image relationship.
The observed solvatochromatic effects can be explained by the
stabilization of the TICT state in agreement with the proposed
model.
The addition of the second donor group gives the system
properties close to those of rhodamine B. A2-TOTAþ has a high
quantum yield and a fluorescent lifetime as would be expected.
Table 3 includes data for three different solvents. The change
in properties of A2-TOTAþ follows the observations for other
amino-xanthenium dyes such as rhodamine B.
Conclusions
New amino-trioxatriangulenium dyes with one and two
amino donor groups on the periphery were synthesized along
with their TPM precursors. In the final ring-closure reactions
pronounced N-dealkylation was encountered emphasizing
the increased charge density on the nitrogen donor groups in
these compound compared to the previous reported tris-
(dialkylamino)trioxatriangulenium systems (A3-TOTAþ).
The optical properties of monoamino-trioxatriangule-
nium and diamino-trioxatriangulenium were also found to
be strongly influenced by the degree of cation stabilization/
localization. Thus, A1-TOTAþ was found to be a surpris-
ingly poor fluorophore. The proposed explanation is a very
efficient formation of a TICT state in the system. A2-TOTAþ
was shown to a good fluorescent dye with strong emission at
540 nm. This dye may be considered as a blue-shifted version
of rhodamine B.
The optical properties of the o-methoxy-substituted TPM
dye intermediates were determined. They were found to
depend on the twist in the conjugated system and the type
and number of donor groups on the periphery.
(61) Turro, N. J. Modern Molecular Photochemistry: University Science
Books: Sausalito, 1991.
(62) Lippert, E.; Rettig, W.; Bonacickoutecky, V.; Heisel, F.; Miehe, J. A.
Adv. Chem. Phys. 1987, 68, 1.
(63) Vogel, M.; Rettig, W.; Fiedeldei, U.; Baumgartel, H. Chem. Phys.
Lett. 1988, 148, 347.
(64) Vogel, M.; Rettig, W.; Sens, R.; Drexhage, K. H. Chem. Phys. Lett.
1988, 147, 452.
(65) Vogel, M.; Rettig, W.; Sens, R.; Drexhage, K. H. Chem. Phys. Lett.
1988, 147, 461.
(66) Plaza, P.; Hung, N. D.; Martin, M. M.; Meyer, Y. H.; Vogel, M.;
Rettig, W. Chem. Phys. 1992, 168, 365.
(67) Rettig, W. In Electron Transfer I; Springer-Verlag: Berlin: Berlin,
1994; Vol. 169, p 253.
Experimental section
All materials were used as received. Solvents were HPLC
grade unless otherwise noted. Experiments were run under
ambient conditions (20-24 °C) except where noted. The fluor-
escence lifetimes were measured using a TC-SPC setup and the
decay data fitted with standard software.
Ethyl 2,6-Dimethoxybenzoate (1).69 To a solution of dimethyl-
resorcinol (13.8 g, 0.1 mol) in ether/benzene 1:1 (100 mL) was
(68) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Chem. Rev. 2003, 103,
3899.
(69) Arrieta, A.; Garcia, T.; Palomo, C. Synth. Commun. 1982, 12, 1139.
6188 J. Org. Chem. Vol. 75, No. 18, 2010

Sørensen and Laursen
JOCArticle
added a 1.6 M solution of n-BuLi in hexane (65 mL, 0.104 mol) at
room temperature. After the reaction mixture was stirred for 3 h,
the reaction was quenched by addition of diethylcarbonate (61 mL,
0.5 mol) and was left to stand overnight. The reaction mixture was
poured onto water (100 mL) and extracted with ether (3 ꢀ 100 mL),
and the combined organic layers were washed with water (3 ꢀ 100
mL) and dried over MgSO4. The solvent was removed to produce
a yellow oil. Recrystallization in heptanes/toluene 3:1 yielded 1
(6.3 g, 30%) as white crystals. 1H NMR (400 MHz, CD2Cl2) δ 7.30
(t, J = 8.4 Hz, 1H), 6.58 (d, J = 8.4 Hz, 2H), 4.31 (q, J = 7.1 Hz,
2H), 3.80 (s, J = 5.2 Hz, 9H), 1.33 (t, J = 7.1 Hz, 3H). GC-MS m/z
210 (Mþ, 20%), 165 (100).
organic layers were washed with water (3 ꢀ 100 mL) and dried
over MgSO4. The solvent was removed, the residue was taken up
in methanol (25 mL), and a 60% aqueous HPF6 (4 mL, 28
mmol) was added followed by 0.2 M KPF6(aq) (350 mL). The
product was extracted with dichloromethane (3 ꢀ 100 mL), the
organic phase was dried over MgSO4, and the solvent volume
was reduced in vacuum. Precipitation with ether allowed for
isolation of the crude 4 (8.5 g, 68%) as a deep blue powder. This
material was used for synthesis without further purification.
Analytical samples of 4 for spectroscopy and analysis were
obtained by recrystallization from methanol, yielding dark blue
powder (4.5 g, 36%). 1H NMR (400 MHz, DMSO) δ 7.35 (t, J =
8.4, 1H), 6.63 (d, J = 8.4, 2H), 6.24 (s, 4H), 4.01 (s, 6H), 3.52 (s,
12H), 3.49 (s, 6H). 13C NMR (101 MHz, CD3CN) δ 174.2,
171.0, 167.0, 160.4, 134.8, 124.4, 122.0, 106.2, 93.6, 58.2, 57.9,
57.4. 19F NMR (376 MHz, DMSO-d6) δ 9.10, 7.21. 31P NMR
(162 MHz, DMSO-d6) δ -132.20, -138.07, -143.93, -149.80,
-155.67. MALDI-TOF (Dithranol matrix) m/z 482.82 (Mþ).
Ethyl 2,4,6-Trimethoxybenzoate (2).70 To a solution of tri-
methylphloroglucinol (16.8 g, 0.1 mol) in ether/benzene 1:1 (100 mL)
was added a 1.6 M solution of n-BuLi in hexane (65 mL, 0.104
mol) at room temperature. After the reaction mixture was stirred
for 3 h, the reaction was quenched by addition of diethylcarbo-
nate (61 mL, 0.5 mol) and was left to stand overnight. The
reaction mixture was poured onto water (100 mL) and extracted
with ether (3 ꢀ 100 mL), and the combined organic layers were
washed with water (3 ꢀ 100 mL) and dried over MgSO4. The
solvent was removed to produce a yellow oil. Recrystallization in
heptanes/toluene 3:1 yielded 1 (7.8 g, 33%) as white crystals. 1H
NMR (400 MHz, CDCl3) δ 6.07 (s, 2H), 4.32 (q, J = 7.1 Hz, 2H),
3.78 (s, 3H), 3.77 (s, 6H), 1.32 (t, J = 7.1 Hz, 3H). 13C NMR (101
MHz, CDCl3) δ 166.6, 162.6, 158.7, 106.6, 90.9, 61.2, 56.2, 55.6,
14.4. GC-MS m/z 240 (Mþ, 25%), 195 (100). Anal. Found: C,
60.3; H, 6.7. Calcd for C12H16O5: C, 60.0; H, 6.7.
λ
max(CH3CN)/nm 576 (log(ε) 4.37), 472sh (4.07), 323 (3.92), 297
(3.87), 271 (4.00). Anal. Found: C, 51.7; H, 5.0. Calcd for
C27H31F6O8P: C, 51.6; H, 5.0. V2O5 added.
General Procedure to 5, 6, and 7. To a solution of the
appropriate starting material (2 mmol), 3 for 5 and 4 for 6/7, in
acetonitrile (5 mL) was added dialkylamine (0.1 mol). The reaction
was stirred at room temperature until MALDI-TOF MS showed
complete reaction of the starting material. The reaction mixture
was then poured onto 0.2 M KPF6(aq) (350 mL), and the crude
product was collected by filtration and washed with water. Re-
crystallization from methanol was used to obtain the pure product
unless another method for purification is mentioned.
Bis(2,6-dimethoxyphenyl)-(2,4,6-trimethoxyphenyl)-methylium
Hexafluorophosphate (3). To a solution of dimethoxybenzene
(6.9 g, 50 mmol) in dry ether (50 mL) was added a 1.6 M solution
of n-BuLi in hexane (31.5 mL, 50 mmol) at room temperature.
After 48 h of stirring a solution of 2 (4.8 g, 20 mmol) in ether
(100 mL) was added, and the reaction was left overnight. The
reaction mixture was poured onto 1 M KOH (100 mL), the
reaction mixture was extracted with ether (3 ꢀ 100 mL), and the
combined organic layers were washed with water (3 ꢀ 100 mL)
and dried over MgSO4. The solvent was removed, the residue
was taken up in methanol (25 mL), and a 60% aqueous HPF6 (4
mL, 28 mmol) was added followed by 0.2 M KPF6(aq) (300 mL).
The product was extracted with dichloromethane (3 ꢀ 100 mL),
the organic phase was dried over MgSO4, and the solvent
volume was reduced in vacuum. Precipitation with ether allowed
for isolation of crude 3 (4.8 g, 56%) as a dark blue powder with a
metallic sheen. This material was used for synthesis without
further purification. Analytical samples of 3 for spectroscopy
and analysis was obtained by repeated recrystallization from
methanol, yielding dark blue needle-like crystals (2.0 g, 17%).
1H NMR (400 MHz, DMSO) δ 7.36 (t, J = 8.4, 2H), 6.64 (d, J =
(4-Diethylamino-2,6-dimethoxyphenyl)-bis(2,6-dimethoxyphe-
nyl)-methylium Hexafluorophosphate (5a). Isolated as a dark
orange powder (840 mg, 78.7%) 1H NMR (400 MHz, CD3CN)
δ 7.24 (t, J = 8.4 Hz, 2H), 6.56 (d, J = 8.4 Hz, 4H), 5.82 (s, 2H),
3.75 (q, J = 7.2 Hz, 4H), 3.51 (s, 12H), 3.49 (s, 6H), 1.33 (t, J =
7.2 Hz, 6H). 13C NMR (101 MHz, DMSO) δ 166.2, 160.8, 157.3,
154.0, 130.1, 124.4, 122.0, 104.2, 90.6, 56.5, 55.9, 46.7, 13.1.
MALDI-TOF (Dithranol matrix) m/z 494.18 (Mþ). λmax
-
(CH3CN)/nm 457 (log(ε) 4.23), 365sh (4.01), 271 (4.10). Anal.
Found: C, 54.6; H, 5.6; N, 2.2. Calcd for C29H36F6NO6P: C,
54.5; H, 5.7; N, 2.2. V2O5 added.
(4-Didecylamino-2,6-dimethoxyphenyl)-bis(2,6-dimethoxyphe-
nyl)-methylium Tetrafluoroborate (5b). Recrystallized from ethyl
acetate to yield a yellow microcrystalline powder (120 mg, 28%).
1H NMR (400 MHz, DMSO) δ 7.19 (t, J = 8.3, 2H), 6.48 (d, J =
8.4, 4H), 5.80 (s, 2H), 3.69 (t, J = 7.6, 4H), 3.55 (s, 12H), 3.51 (s,
6H), 1.73 (m, 4H), 1.31 (m, 28H), 0.87 (t, J = 6.7, 6H). 13C NMR
(101 MHz, CDCl3) δ 167.0, 162.0, 158.0, 154.6, 130.8, 125.3,
122.7, 104.5, 90.6, 56.5, 53.3, 32. 1, 29.7, 29.7, 29.6, 29.5, 28.6,
27.0, 22.9, 14.3. 19F NMR (376 MHz, DMSO-d6) δ -3.55.
8.4, 4H), 6.30 (s, 2H), 4.21 (s, 3H), 3.58 (s, 6H), 3.48 (s, 12H). 13
C
MALDI-TOF (Dithranol matrix) m/z 718.71 (Mþ). λmax
-
NMR (101 MHz, DMSO-d6) δ 164.9, 157.2, 128.7, 128.7,
125.44, 124.7, 122.6, 104.1, 100.0, 55.8, 55.2, 48.6. 19F NMR
(376 MHz, DMSO) δ 9.12, 7.23. 31P NMR (162 MHz, DMSO) δ
-132.20, -138.07, -143.93, -149.80, -155.67. MALDI-TOF
(Dithranol matrix) m/z 452.96 (Mþ). λmax(CH3CN)/nm 492
(log(ε) 4.15), 320 (4.06), 311sh (4.03), 272 (3.92). Anal. Found:
C, 52.5; H, 4.9. Calcd for C26H29F6O7P: C, 52.2; H, 4.9. V2O5
added.
(2,6-Dimethoxyphenyl)-bis(2,4,6-trimethoxyphenyl)-methylium
Hexafluorophosphate (4). To a solution of trimethoxybenzene
(8.4 g, 50 mmol) in dry ether (50 mL) was added a 1.6 M solution
of n-BuLi in hexane (31.5 mL, 50 mmol) at room temperature.
After 48 h of stirring a solution of 1 (4.2 g, 20 mmol) in ether
(100 mL) was added, and the reaction was left overnight. The
reaction mixture was poured onto 1 M KOH (100 mL), the pro-
duct was extracted with ether (3 ꢀ 100 mL), and the combined
(CH3CN)/nm identical to 5a. Anal. Found (on PF6 salt): C,
62.6; H, 8.2; N, 1.4. Calcd for C45H68F6NO6P: C: 62.6, H: 7.9, N:
1.6. V2O5 added.
(4-Dimethylamino-2,6-dimethoxyphenyl)-bis(2,6-dimethoxyphe-
nyl)-methylium Hexafluorophosphate (5c). Isolated as a red pow-
der (95 mg, 42%). 1H NMR (400 MHz, CD2Cl2) δ 7.23 (t, J = 8.4,
2H), 6.51 (d, J = 8.4, 4H), 5.71 (s, 2H), 3.53 (s, 12H), 3.51 (s, 6H),
3.43 (s, 6H). 13C NMR (101 MHz, DMSO-d6) δ 165.7, 162.0,
157.2, 154.0, 130.1, 128.8, 128.2, 125.3, 124.4, 122.1, 104.3, 91.0,
56.4, 55.9, 41.9. 31P NMR (162 MHz, DMSO) δ -132.20,
-138.07, -143.93, -149.80, -155.67. MALDI-TOF (Dithranol
matrix) m/z 466.12 (Mþ). λmax(CH3CN)/nm identical to 5a. Anal.
Found: C: 53.1, H: 5.3, N: 2.5. Calcd for C27H32F6NO6P: C: 53.0,
H: 5.3, N: 2.3. V2O5 added.
Bis(4-diethylamino-2,6-dimethoxyphenyl)-(2,6-dimethoxyphe-
nyl)-methylium Hexafluorophosphate (6a). Isolated as a dark
1
(70) Hendrickson, J. B.; Schwartzman, S. M. Tetrahedron Lett. 1975, 277.
blue powder (0.862 g, 76%). H NMR (500 MHz, CD3CN) δ
J. Org. Chem. Vol. 75, No. 18, 2010 6189

Sørensen and Laursen
JOCArticle
7.22 (t, J = 8.3, 1H), 6.57 (d, J = 8.3, 2H), 5.78 (s, 4H), 3.57 (q,
J = 7.1, 9H), 3.51 (d, J = 11.0, 6H), 3.47 (s, 11H), 1.24 (t, J =
7.1, 12H). 13C NMR (101 MHz, DMSO-d6) 164.4, 158.5, 156.2,
155.1, 130.4, 124.2, 117.6, 104.9, 88.9, 77.5, 77.2, 77.0, 56.8, 56.2,
45.9, 13.2. MALDI-TOF (Dithranol matrix) m/z 565.26 (Mþ).
1H NMR (500 MHz, CDCl3) δ 8.01 (t, J = 8.5 Hz, 2H), 7.45 (d,
J = 8.4 Hz, 2H), 7.42 (d, J = 8.5 Hz, 2H), 6.76 (s, 2H), 3.59 (s,
4H), 1.76 (s, 4H), 1.42 (s, 8H), 1.29 (s, 20H), 0.89 (s, 6H). 13C
NMR (126 MHz, CDCl3) δ 154.5, 152. 9, 152.3, 139.0, 119.3,
112.2, 104.4, 96.2, 53.0, 31.8, 29.6, 29.5, 29.3, 29.3, 27.2, 26.9,
22.7, 14.1, (incomplete, the quaternary carbon missing). MAL-
DI-TOF (Dithranol matrix) m/z 580.22 (Mþ). Hi-Res ESI-TOF
λmax(CH3CN)/nm 634 (log(ε) 4.69), 547sh (4.15), 421 (3.78), 380
(3.78), 300sh (4.07), 255 (4.36). Anal. Found: C, 56.0; H, 6.1; N,
3.9. Calcd for C33H45F6N2O6P: C, 55.8; H, 6.4; N, 3.9. V2O5
added.
Found: 580.3791. Calcd for C39H50NO3þ: 580.3785. λmax
-
(CH3CN)/nm 502 (log(ε) 4.60), 475 (4.57), 441sh (4.29), 302.5
(4.45), 268.5 (4.57), 235 (4.83). Anal. Found: C, 64.8; H, 7.0; N,
1.9. Calcd for C39H50F6NO3P: C, 64.5; H, 6.9; N, 1.9. V2O5
added.
Bis(4-didecylamino-2,6-dimethoxyphenyl)-(2,6-dimethoxyphe-
nyl)-methylium Hexafluorophosphate (6b). The compound co-
crystallizes with a didecylamine as a dark purple wax (0.85 g,
39%). 1H NMR (500 MHz, CDCl3) δ 7.19 (t, J = 8.3 Hz, 1H),
6.51 (d, J = 8.4 Hz, 2H), 5.68 (s, 4H), 3.57 (s, 6H), 3.54 - 3.40
(m, J = 24.1 Hz, 20H), 2.65 (t, 4H), 1.82 - 1.60 (m, J = 33.8 Hz,
12H), 1.59 - 1.45 (m, 4H), 1.41 - 1.19 (m, 80H), 0.88 (t, J = 6.9
Hz, 18H). MALDI-TOF (Dithranol matrix) m/z 1013.72 (Mþ).
Bis(diethylamino)-trioxatriangulenium Hexafluorophosphate
(9a). A solution of 6a (180 mg, 0.25 mmol) in N-methylpyrro-
lidone (3 mL) was heated to 140 °C, and collidine (3 mL) and LiI
(0.8 g, 6.0 mmol) were added. The reaction was followed on
MALDI-TOF, and when only the product was observed, the
reaction mixture was cooled and poured onto 0.2 M aqueous
KPF6 (250 mL) with 50% aqueous KPF6 (1 mL). The product
was extracted with dichloromethane (3 ꢀ 100 mL), and the
organic phase was washed with 0.2 M aqueous KPF6 (2 ꢀ 200 mL)
and dried over MgSO4. Recrystallization from ethyl acetate/
dichloromethane afforded the product (50 mg, 35%) as orange
λmax(CH3CN)/nm identical to 6a.
Bis(4-dimethylamino-2,6-dimethoxyphenyl)-(2,6-dimethoxyphe-
nyl)-methylium Hexafluorophosphate (6c). Dark violet powder
(200 mg, 39%). 1H NMR (400 MHz, DMSO-d6) δ 7.19 (t, J =
8.3, 1H), 6.57 (d, J = 8.3, 2H), 5.85 (s, 4H), 3.47 (s, 6H), 3.44
(s, 12H), 3.20 (s, 12H). 13C NMR (126 MHz, DMSO-d6) δ
163.2, 157.7, 157.2, 155.5, 129.6, 123.9, 117.0, 104.8, 89.2, 56.3, 55.9,
40.4. 19F NMR (376 MHz, DMSO) δ 9.10, 7.21. 31P NMR (162
MHz, DMSO) δ=-132.20, -138.07, -143.93, -149.80, -155.67.
MALDI-TOF (Dithranol matrix) m/z509.30 (Mþ).λmax(CH3CN)/
nm identical to 6a. Anal. Found: C, 53.1; H, 5.7; N, 4.5. Calcd for
C29H37F6N2O6P: C, 53.2; H, 5.7; N, 4.3.
1
platelets. H NMR (500 MHz, CDCl3) δ 7.85 (t, J = 8.4 Hz,
1H), 7.28 (t, J = 6.0 Hz, 2H), 6.66 (d, J = 1.8 Hz, 2H), 6.58 (s,
2H), 3.60 (q, J = 7.1 Hz, 8H), 1.34 (t, J = 7.2 Hz, 12H). 13C
NMR (126 MHz, DMSO-d6) δ 157.5, 154.7, 154.2, 152.6, 138.4,
133.0, 112.2, 104.3, 96.5, 95.5, 95.3, 46.2, 12.9. MALDI-TOF
(Dithranol matrix) m/z 427.00 (Mþ). Hi-Res ESI-TOF Found:
427.1986. Calcd for C27H27N2O3þ: 427.2016. λmax(CH3CN)/nm
511 (log(ε) 4.66), 478.5sh (4.35), 412.5 (4.25), 388.5 (4.03),
367.5sh (3.73), 317 (3.96), 305sh (3.87), 244.5 (4.58), 235
(4.58). Anal. Found: C, 48.0; H, 4.7; N, 4.9. Calcd for C27H27-
F6N2O3P: C, 56.7; H, 4.8; N, 4.9. Incomplete combustion even
with V2O5 added. This is commonly observed as a too low
carbon value for this compound class.8
(4-Didecylamino-2,6-trimethoxyphenyl)- (2,6-dimethoxyphe-
nyl)-(2,4,6-trimethoxyphenyl)-methylium Tetrfaluoroborate (7).
Isolated as a dark violet powder (40 mg, 7.1%). 1H NMR
(300 MHz, CDCl3) δ = 7.19 (s, 1H), 6.49 (d, J = 8.4, 2H),
6.04 (s, 2H), 5.73 (d, J = 11.6, 2H), 3.83 (s, 3H), 3.64 (s, 4H), 3.57
(s, 12H), 3.53 (s, 6H), 1.26 (s, 32H), 0.86 (d, J = 6.5, 6H). 13
C
NMR (101 MHz, CDCl3) δ = 166.6, 163.0, 161.1, 159.3, 157.7,
130.4, 130.4, 123.9, 122.9, 104.2, 90.8, 89.9, 89.9, 56.3, 56.2, 56.1,
56.0, 55.3, 52.7, 31.8, 29.40, 29.35, 29.23, 29.16, 28.2, 28.1, 26.7,
Bis(didecylamino)-trioxatriangulenium Hexafluorophosphate
(9b). A solution of 6b (850 mg, 0.75 mmol) in N-methylpyrro-
lidone (3 mL) was heated to 140 °C, and collidine (3 mL) and LiI
(3 g, 22 mmol) were added. The reaction was followed on
MALDI-TOF, and when only the product was observed, the
reaction mixture was cooled and poured onto 0.2 M aqueous
KPF6 (300 mL) with 50% aqueous KPF6 (1 mL). The product
was extracted with dichloromethane (3 ꢀ 100 mL), and the or-
ganic phase was washed with 0.2 M aqueous KPF6 (2 ꢀ 200 mL)
and dried over MgSO4. The crude product (900 mg) was
collected by filtration and dried; the main impurities were free
amine and monodealkylated product. Repeated recrystalliza-
tions from methanol/dichloromethane afforded the pure pro-
22.6, 14.0. MALDI-TOF (Dithranol matrix) m/z 748.68 (Mþ).
þ
Hi-Res ESI-TOF Found: 748.5121. Calcd for C46H70NO7
:
748.5152. λmax(CH3CN)/nm 511 (log(ε) 4.29), 418sh (4.10),
371sh (4.04), 269 (4.12).
Diethylamino-trioxatriangulenium Hexafluorophosphate (8a).
A solution of 5a (100 mg, 0.16 mmol) in N-methylpyrrolidone
(50 mL) was heated to 140 °C, and collidine (5 mL) and LiI (1 g,
7.5 mmol) were added. The reaction was followed on MALDI-
TOF, and when the peak at m/z 402 disappeared, corresponding
to the two times ring-closed intermediate, the reaction mixture
was cooled and poured onto 0.2 M aqueous KPF6 (300 mL). The
crude product is collected by filtration and dried. Flash chro-
matography on silica with dichloromethane/methanol 1:20
afforded the pure compound in trace amounts as yellow hair
crystals (<3%). 1HNMR(500MHz,CD2Cl2) δ8.08 (t, J=8.5Hz,
2H), 7.56 - 7.37 (m, 4H), 6.81 (s, 2H), 3.70 (q, J = 7.3 Hz, 4H),
1.41 (t, J = 7.2 Hz, 6H). MALDI-TOF (Dithranol matrix) m/z
355.89 (Mþ). Hi-Res ESI-TOF Found: 356.1280. Calcd for
C23H18NO3þ: 356.1281. λmax(CH3CN)/nm identical to 8b.
Didecylamino-trioxatriangulenium Hexafluorophosphate (8b).
A solution of 5b (50 mg, 0.06 mmol) in N-methylpyrrolidone (3
mL) was heated to 140 °C, and collidine (3 mL) and LiI (0.4 g,
3.0 mmol) were added. The reaction was followed on MALDI-
TOF, and when only the product was observed, the reaction
mixture was cooled and poured onto 0.2 M aqueous KPF6 (250
mL) with 50% aqueous KPF6 (1 mL). The product was ex-
tracted with dichloromethane (3 ꢀ 100 mL), and the organic
phase was washed with 0.2 M aqueous KPF6 (2 ꢀ 200 mL) and
dried over MgSO4. Recrystallization from ethyl acetate/n-heptane
afforded the product (10 mg, 23%) as hair thin yellow crystals.
1
duct in trace amounts as a dark red powder. H NMR (500
MHz, CD2Cl2) δ 7.90 (t, J=8.4 Hz, 1H), 7.35 (d, J=22.0, 8.6 Hz,
2H), 6.60 (s, 4H), 3.50 (t, J=15.2, 7.1 Hz, 8H), 1.79 - 1.49 (m,
8H), 1.31 (dd, J=45.1, 31.4 Hz, 24H), 0.90 (t, 12H). MALDI-
TOF (Dithranol matrix) m/z 875.05 (Mþ). Hi-Res ESI-TOF
Found: 875.7026. Calcd for C59H91N2O3þ: 875.7024. λmax
(CH3CN)/nm identical to 9a.
-
Acknowledgment. We would like to thank Dr. Christian
Tortzen for his help with the characterization, Prof. Klaus
Bechgaard and Professor Mark van der Auweraer for
fruitful discussions and the Danish Council for Indepen-
dent Research | Natural Sciences for funding (Grant no.
272-06-0102).
Supporting Information Available: 1H and 13C NMR spec-
tra of all prepared compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
6190 J. Org. Chem. Vol. 75, No. 18, 2010