4,4-Difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) dyes modified for extended conjugation and restricted bond rotations

Article abstract of DOI:10.1021/jo991927o

Five new, constrained, aryl-substituted 4,4-difluoro-4-bora-3a,4a-diaza- s-indacene (BODIPY) dyes (3f,g and 4h-j) were prepared and investigated to see if they have more favorable fluorescence characteristics than the unconstrained systems 2 that were prepared in previous studies. Dye types 3 and 4 have relatively rigid conformations caused by the heteroatom (3f and 3g) or ethylene bridge (4h-j) linkers that preclude free rotation of the substituted-benzene molecular fragments. In the event, the new dye types 3 and 4 have longer λ(max abs) (620-660 nm) and λ(max fluor) (630-680 nm) values than compounds 2. They also exhibit higher extinction coefficients (> 100 000 M^-1 cm^-1, except for 3g). Their fluorescent quantum yields are high (up to 0.72 for 4j), with the exception of compound 3g, which has a quantum yield of only 0.05. The redox properties of dyes 3 and 4 have also been examined.

Full text of DOI:10.1021/jo991927o

2
900
J . Org. Chem. 2000, 65, 2900-2906
4
,4-Diflu or o-4-bor a -3a ,4a -d ia za -s-in d a cen e (BODIP Y) Dyes
Mod ified for Exten d ed Con ju ga tion a n d Restr icted Bon d
Rota tion s
J iong Chen, Armin Burghart, Agnes Derecskei-Kovacs, and Kevin Burgess*
Texas A & M University, Chemistry Department, P.O. Box 30012, College Station, Texas 77842
Received December 16, 1999
Five new, constrained, aryl-substituted 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) dyes
(
3f,g and 4h -j) were prepared and investigated to see if they have more favorable fluorescence
characteristics than the unconstrained systems 2 that were prepared in previous studies. Dye types
and 4 have relatively rigid conformations caused by the heteroatom (3f and 3g) or ethylene bridge
4h -j) linkers that preclude free rotation of the substituted-benzene molecular fragments. In the
event, the new dye types 3 and 4 have longer λmax abs (620-660 nm) and λmax fluor (630-680 nm)
3
(
-
1
-1
values than compounds 2. They also exhibit higher extinction coefficients (>100 000 M cm
,
except for 3g). Their fluorescent quantum yields are high (up to 0.72 for 4j), with the exception of
compound 3g, which has a quantum yield of only 0.05. The redox properties of dyes 3 and 4 have
also been examined.
In tr od u ction
A project to develop new fluorescent labels for DNA
are well-known to enhance fluorescence, hence one of the
objectives of this work was to prepare dyes that are
related to systems 2 but are more rigid. This paper
reports syntheses, spectral properties, and electrochemi-
cal data for the “planar aromatic” dyes 3 and 4.
sequencing and other biotechnological applications has
led us to modify the 4,4-difluoro-4-bora-3a,4a-diaza-s-
indacene1 (BODIPY) framework to obtain dyes that
fluoresce strongly over a broad range of different wave-
lengths. Alkyl-substituted BODIPY dyes such as 1 are
1
well-known. These emit strongly in the 510 nm region;
hence, alternative labels that complement these with
higher wavelength fluorescence maxima were sought. In
previous studies, we had reasoned that aryl substituents
would shift the fluorescence to longer wavelengths and
increase the molar extinction coefficients of these mol-
ecules. Consequently, systems 2 were prepared and
examined via a variety of spectroscopic methods.2 These
experiments were successful insofar as the fluorescence
emission maxima of these molecules were shifted to
longer wavelengths (588-626 nm). However, the aryl
substituents did not markedly increase the extinction
coefficients of these molecules and the fluorescence
quantum yields of 2 were significantly lower than that
of the alkyl-substituted dye 1.
,3
Resu lts a n d Discu ssion
Syn th eses of th e P yr r ole-Der ived Sta r tin g Ma te-
r ia ls. The first steps necessary for synthesis of com-
pounds 3 and 4 were preparation of the requisite pyrrole
derivatives. These syntheses are summarized in Scheme
1
. The benzofuro[3,2-b]pyrrole 5f is a known compound;
4
it was prepared according to the literature procedure but
with a significant modification in the final step. The
literature decarboxylation was previously performed
using copper(II) chromite in quinoline (43% yield). In the
5
current work, NaOH in ethylene glycol was shown to
give a better yield (80%) via an easier purification
(
3.
1) Treibs, A.; Kreuzer, F.-H. Liebigs Ann. Chem. 1968, 718, 208-
2
(2) Thoresen, L. H.; Kim, H.; Welch, M. B.; Burghart, A.; Burgess,
K. Synlett 1998, 1276-8.
3) Burghart, A.; Kim, H.; Welch, M. B.; Thoresen, L. H.; Reiben-
spies, J .; Burgess, K.; Bergstr o¨ m, F.; J ohansson, L. B.-A. J . Org. Chem.
999, 64, 7813-9.
4) Krutosikova, A.; Kovac, J .; Dandarova, M.; Bobalova, M. Collect.
Czech. Chem. Commun. 1982, 47, 3288-96.
(5) Lash, T. D. Tetrahedron 1998, 54, 359-74.
A possible rationale for the reduced quantum yields of
the aryl-substituted dyes 2 relative to the alkyl-substi-
tuted systems 1 is that energy is lost from the excited
state via thermal pathways involving rotation of the
aromatic substituents. In general, molecular constraints
(
1
(
1
0.1021/jo991927o CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/18/2000

Extended Conjugation by BODIPY Dyes
J . Org. Chem., Vol. 65, No. 10, 2000 2901
Sch em e 1. Syn th eses of th e P yr r ole-Ba sed
Sta r tin g Ma ter ia ls
Interestingly, and for reasons that are unclear, the
Trofimov methodology did not give the desired products
1
0 when the oximes 11 were reacted under the conditions
described in Scheme 1b. Moreover, our attempts to form
the desired product 10 (R ) Me) via condensation of the
3
keto-aldehyde 12 and NH (or its equivalents) were also
unsuccessful (compound 13 was made as an intermediate
in the synthesis of 12). No further attempts were made
to prepare compound 10.
The literature procedure for preparation of the 3-meth-
yl-4,5-dihydro-1H-benz[g]indole 14j calls for Na/NH
3
9
reduction in an autoclave. In this work, we attempted
to find a more convenient procedure that could be easily
scaled up. The route that was developed involved a Knorr
pyrrole synthesis10 of the carboxypyrrole derivative 16
followed by hydrolysis and decarboxylation as indicated
in Scheme 1c. Unfortunately, the yield of 16 obtained
from the Knorr synthesis was low, but the procedure
could be carried out on a 20 g scale to give a crystalline
product without chromatographic separation.
Syn th eses of Dyes 3 a n d 4. The pyrrole-based
starting materials 5f,g, 8h ,i, and 14j can be divided into
two classes with respect to their reactivity toward elec-
trophiles. Compounds 5f and 5g are relatively stable;
they will not react with 4-iodobenzoyl chloride unless a
suitable Lewis acid is added. Conversely, 8h ,i and 14j
degrade on silica gel and darken after several hours on
contact with the air. They react with 4-iodobenzoyl
chloride in the absence of Lewis acids. In view of these
reactivity differences, two synthetic methods were de-
vised for the preparation of the BODIPY systems 3 and
4 (Scheme 2).
As mentioned above, the heterocycles 5f and 5g did
not react with 4-iodobenzoyl chloride, even in refluxing
toluene. A reaction did occur when boron trifluoride
etherate was added, but a complex mixture of products
resulted. However, addition of triethylamine, presumably
procedure. Thianaphtheno[3,2-b]pyrrole 5g is also a
known compound, but it was previously made via a
6
relatively long and tedious route. In the current work,
the approach used for compound 5f was successfully
adopted for compound 5g, as indicated in Scheme 1a.
Scheme 1b shows the route used to obtain the 4,5-
dihydro-1H-benz[g]indoles 8h and 8i. These compounds
were obtained by procedures of Trofimov that involve
reaction of the appropriate ketoxime with ethyne in KOH/
7
DMSO. Those conditions afforded the product contami-
3
to attenuate the Lewis acidity of the BF , gave the desired
nated with between 30 and 40% of the corresponding
N-vinyl derivative 9h that presumably results from over-
reaction of 8h with ethyne. Other contributions by
Trofimov and co-workers reported that use of LiOH/
material more selectively (Scheme 2a). The yields ob-
tained in this transformation were low, especially for the
sulfur-containing compound 3g. However, the products
were easy to track by TLC and during flash chromatog-
raphy, since they display red fluorescence.
The more reactive nucleophiles 8h ,i and 14j reacted
smoothly in refluxing 1,2-dichloroethane to form the
dipyrromethene intermediates. These dipyrromethenes
are isolable, but it was more convenient in this work to
add boron trifluoride/triethylamine and resume the reflux
8
DMSO suppressed formation of N-vinyl byproducts. Our
studies demonstrated a similar effect in the synthesis of
compound 8h , as shown in Scheme 1b. Compound 8i had,
to the best of our knowledge, not been made previously,
but was produced in this study using the method il-
lustrated in Scheme 1b.
(
6) Shkurko, O. P.; Mamaev, V. P. Izv. Sib. Otd. Akad. Nauk SSSR,
Ser. Khim. Nauk 1967, 2, 112-7.
(9) Enders, D.; Maassen, R.; Han, S.-H. Liebigs Ann. 1996, 1565-
(
(
7) Trofimov, B. A. Adv. Heterocycl. Chem 1990, 51, 177-301.
74.
8) Korostova, S. E.; Trofimov, B. A.; Sobenina, L. N.; Mikhaleva,
(10) Quizon-Colquitt, D. M.; Lash, T. D. J . Het. Chem. 1993, 30,
477-82.
A. I.; Sigalov, M. V. Chem. Heterocycl. Comput. 1982, 14, 1043-5.

2
902 J . Org. Chem., Vol. 65, No. 10, 2000
Chen et al.
Sch em e 2. Syn th eses of Dyes 3 a n d 4
Ta ble 1. Selected Op tica l P r op er ties of Com p ou n d s 3f,g
a n d 4h -j
λmax (abs)
(nm)
ꢀ
λmax (em)
(nm)
Stokes shift
(nm)
(M- cm-1
)
1
Φa
dye
3
3
4
4
4
8
f
637
658
634
658
619
634
151 000
54 250
126 250
139 500
145 750
41 000
647
673
647
673
629
668
10
15
13
15
10
34
0.34
0.05
0.38
0.13
0.72
0.17
g
h
i
j
1
a
Measured in CHCl3 using cresyl violet perchlorate19 in MeOH
as reference (Φ ) 0.54).
Sch em e 3. Oxid a tion of Com p ou n d 4h
4
i was reported as part of that study along with semiem-
pirical calculations to simulate the conformations of other
compounds in this series. Calculations were also per-
formed for compound 4i and then compared with the
crystallographic data to test the simulation approach. In
the event, the correspondence was good, therefore vali-
dating the method used for the calculations. No further
discussion of those particular data will be presented here,
but the scope of the study was expanded as described
below.
to give the desired BODIPY products 4. Overall yields of
the products in this approach were satisfactory (Scheme
2
b).
F u n d a m en ta l UV a n d F lu or escen ce P r op er ties of
Dyes 3 a n d 4. Dyes 3 and 4 are strongly colored solids
that have a metallic luster (3f, green; 3g, dark blue; 4h ,
gray-blue; 4i, copper; 4j, copper). They form intensely
colored solutions (3f, green; 3g, dark blue; 4h , bright
blue; 4i, greenish blue; 4j, bright blue) that appear to
glow with a beautiful red fluorescence when irradiated.
These dyes have absorption maxima between 620 and
BODIPY 4h was oxidized using excess DDQ at room
temperature in CH Cl . This oxidation reaction rapidly
2 2
and cleanly gave the half-oxidized compound 18, not the
doubly oxidized compound 19. Further treatment of
compound 18 with DDQ (>10 equiv) at room temperature
for overnight gave no reaction. If this solution was heated
to reflux, then a complicated mixture resulted (Scheme
6
60 nm and their emission maxima between 630 and 680
nm. All five except 3g have very high extinction coef-
-1
-1
ficients (>100 000 M cm ). Their fluorescence quantum
yields were found to vary considerably (0.05 for 3g; 0.72
for 4j). Table 1 lists the salient spectral properties of
these compounds.
3
).
Steric effects could account for the fact that only one
ring of compound 4h was oxidized easily. To test this
hypothesis, semiempirical calculations were performed
on the series 4h , 18, and 19 (Cl for I analogues) using
the protocol described previously.11 The results are shown
in Figure 1.
The average interatomic distance between the fluorine
atoms and their closest hydrogen neighbors in compound
4h was calculated to be 2.35 Å. Oxidation to compound
Str u ctu r e a n d Con for m a tion s of Dyes 3 a n d 4.
Compounds 3g and 4h -i all give long-range, through-
space, 1 C- F coupling involving the carbon closest to
3
19
the boron in the crowded bay region. These data have
been communicated.11 A crystal structure of compound
(
11) Chen, J .; Reibenspies, J .; Derecskei-Kovacs, A.; Burgess, K.
Chem. Commun. 1999, 2501-2.

Extended Conjugation by BODIPY Dyes
J . Org. Chem., Vol. 65, No. 10, 2000 2903
F igu r e 1. Simulated structures of chlorine-for-iodine analogues of a , compound 4h ; b, compound 18; and, c, the hypothetical
compound 19.
Ta ble 2. Electr och em ica l Da ta for Com p ou n d s 3f a n d
1
8 caused a slight compression of this interatomic
4
h -ja
distance to an average of 2.29 Å. Compounds 4h and 18
had similar F‚‚‚C couplings (11.0 and 11.5/11.3 Hz,
respectively). Simulations for compound 19 implied that,
if it were formed, it would have a flat structure as
indicated in Figure 1c. The corresponding interatomic
distances for this hypothetical structure were predicted
to be 2.25 Å. This is considerably shorter than the sum
of the van der Waals radii for hydrogen and fluorine.12
Consequently, it seems probable that the activation
energy required to form compound 19 is excessive com-
pared with the half oxidized form 18 due to steric reasons.
The optical properties of compound 18 were not as we
had expected (Table 1). First, the maximum absorbance
of 18 was not red shifted relative to 4h ; in fact the two
values were coincident, though the fluorescence emission
was red-shifted by 21 nm relative to 4h . Second, despite
the more conjugated characteristics of 18, it has an
extinction coefficient that is approximately three times
smaller than compound 4h . Finally, the quantum yield
for 18 was less than half that of 4h . The consequence of
these characteristics is that compound 18 is not a
particularly fluorescent compound and has limited po-
tential as a molecular label.
E1/2 (V)
Epc (V)
first ox.
wave
second ox.
wave
first red.
wave
second red.
wave
3f
+0.81
+0.57
+0.33
+0.51
b
b
-1.05
-1.29
-1.39
-1.44
-1.92
-2.07
-2.11
b
4
4
4
h
i
j
+0.91
b
a
All data are referenced versus ferrocene. b Electrochemical
wave could not be detected under the conditions chosen.
Electr och em ica l Stu d ies of Dyes 3f a n d 4. The
BODIPY derivatives 3f and 4 were examined via cyclic
voltammetry (3g, the least soluble dye in the series, was
excluded). This technique gives a measure of the acces-
sibility and distribution of redox states; extent of π-con-
jugation and conformational parameters can be deduced
2 2
F igu r e 2. Cyclic voltammogram of dye 4i in CH Cl (refer-
enced versus Ag wire).
(Figure 2). Both oxidation waves for this compound were
shown to be reversible; this implies that even the dication
is sufficiently long-lived to be reduced in the back sweep
under these conditions.
2 2
from these data. Throughout, CH Cl solutions and a scan
-
1
rate of 100 mVs were used. All systems displayed
reversible waves for one-electron oxidation and for one-
electron reduction (Table 2). Some of the dyes exhibited
a second redox step for either oxidation and/or reduction.
Notably, the methoxy-substituted BODIPY 4i gave four
distinct redox steps: two oxidations and two reductions
Within the series of compounds 3f and 4, the methoxy-
BODIPY 4i was shown to have the lowest oxidation
potential (E1/2 ) 0.33 V vs ferrocene), as expected.
Overall, the rigid BODIPYs 3f and 4 exhibit much lower
oxidation potentials than systems 2 whose conjugation
3
is effectively less extended. Comparison of compounds
(12) Bondi, A. Phys. Chem 1964, 68, 441.
4h and 4j (R ) H, and Me, respectively) affords insights

2
904 J . Org. Chem., Vol. 65, No. 10, 2000
Chen et al.
into the effect of the two methyl groups flanking the
central 4-iodobenzene ring in 4j. Curiously, compound
are expressed in ppm relative to internal solvent peaks, and
coupling constants were measured in Hz. For 11B NMR, BF
‚
3
3
Et
2
O has been used as an external reference; similarly, CFCl
4
j is easier to oxidize but harder to reduce. The electronic
1
9
was used as an external standard for the F spectra.
effect imparted by the methyl R-groups seems to lower
the oxidation potential of 4j. The fact that 4j is more
difficult to reduce may be indicative of less effective
conjugation of the indacene system with the 4-iodophenyl
group. This would be caused by steric interactions
involving the methyl groups that prevent the aryl sub-
Geometry optimizations at the AM1 level using the MOPAC
module within the MSI Cerius2 program package were per-
11
formed as described previously.
4
Ben zofu r o[3,2-b]p yr r ole 5f. Acid 7f (0.25 g, 1.24 mmol)
and NaOH (0.20 g, 5 mmol) were mixed in ethylene glycol (10
mL) and heated at 140 °C for 10 min under nitrogen. The
reaction mixture was cooled to room temperature, diluted with
stituent from becoming coplanar with the rest of the dye
system.13 Compound 3f exhibits unique redox properties
water (100 mL), and extracted with Et
combined ether solution was washed with water and brine and
dried over Na SO . Evaporation gave a gray solid that was
then dissolved in approximately 10 mL of CH Cl and passed
2
O (3 × 50 mL). The
within the series. It is very easily reduced having the
least negative reduction potential in the series, but on
the other hand, it has the highest oxidation potential.
Separations between the first and the second oxidation
or reduction steps provide information regarding charge
distribution. Small separation implies that the charge of
the corresponding radical ion is confined to a particular
region of the molecule, and the second redox step occurs
predominantly at a remote site. The expected outcome
in such a situation is a dication or dianion with relatively
localized and distinct charges. Large separations, how-
ever, are indicative of a widely delocalized charge in the
radical ion. Compound types 3f and 4 display relatively
large separations both for oxidation and reduction,
indicative of delocalization. The smallest separation
observed was 580 mV (oxidation of 4i; E1/2 ) 0.33 and
2
4
2
2
through a short column of silica gel. This gave 5f as white
1
solid after removal of solvent (0.17 g, 79%). The H NMR data
obtained for this compound match that in the literature for
4
preparation of the same material via a different route.
BODIP Y 3f. Pyrrole 5f (0.15 g, 0.96 mmol) and 4-iodoben-
zoyl chloride (0.15 g, 0.56 mmol) were dissolved in toluene (20
mL) and heated to reflux under nitrogen, and then Et
3
N (1.0
mL, 7.2 mmol) and BF etherate (1.2 mL, 9.8 mmol) were
3
added. The heating was continued, and two more batches (0.05
g, 0.19 mmol each) of 4-iodobenzoyl chloride were added, one
at 40 h and one at 60 h. The solvent was removed after 3 d,
and the resulting dark brown residue was purified via a simple
chromatography on silica gel column, eluting with 3:1 hexanes/
CH Cl , to remove the polar residues, then a more careful flash
2 2
chromatography, using the same solvent combination. This
gave compound 3f as a dark green solid after the removal of
0
.91 V). This is in striking contrast to, for example,
solvent (60 mg, 22%): mp >300 °C; R
f
value 0.42 (4:1 hexanes/
) δ 8.21 (d, J ) 7.5 Hz, 2H), 7.91 (d,
J ) 8.4 Hz, 2H), 7.51-7.35 (m, 6H), 7.36 (d, J ) 8.4 Hz, 2H),
1
tetraanisylethylene for which the first and second oxida-
EtOAc); H NMR (CDCl
3
1
4
tion potentials are separated by only 120 mV.
6
1
1
.36 (s, 2H); 13C NMR (CDCl
38.7, 137.8, 133.4, 132.1, 130.5, 124.2, 123.9, 117.0, 112.8,
3
) δ 164.8, 154.6, 146.1, 143.8,
Con clu sion s
19
11
04.3, 96.8; F NMR (CDCl
) δ 0.41 (t, J ) 28 Hz); FABHRMS calcd for
574.0163, found 574.0167.
Eth yl Th ia n a p h th en o[3,2-b]p yr r ole-2-ca r boxyla te 6g.
3
) δ -150.6 (q, J ) 28 Hz);
B
NMR (CDCl
2 2
IN O
3
Relative to the unconstrained aromatic dyes 2, com-
pound types 3 and 4 have several desirable attributes
with respect to applications as fluorescent labels. They
absorb and fluoresce more intensely at longer wave-
lengths, and the quantum yields for this process are
generally higher, with the exception of compound 3g.
These observations confirm the hypothesis that restricted
rotation of the aryl fragments in these molecules is
advantageous.
The sulfur compound 3g has a lower extinction coef-
ficient and quantum yield than the other compounds in
this series; the spectroscopic disparities between 3g and
the rest of the series cannot be explained with confidence.
It is also less soluble than the others in common organic
solvents. However, the other BODIPYs reported here
may find a range of applications including our own
objective of developing new fluorescent labels for bio-
technological applications.
C
27
H
14BF
2
Benzothiophene-2-carbaldehyde (0.21 g, 1.3 mmol) and ethyl
azidoacetate (0.33 g. 2.6 mmol) were dissolved in EtOH (10
mL). A solution of EtONa in EtOH (21 wt %, 0.95 mL, 2 equiv)
was added dropwise at 0 °C. The mixture was then stirred for
another 1 h. A yellow precipitate formed. Excess saturated
4
NH Cl(aq) was added, and the mixture was filtered to give a
yellow solid that was washed with water and dried under
vacuum. It was then dissolved in 20 mL of toluene and refluxed
for 30 min. Evaporation of the solvent gave a yellow residue,
which was recrystallized from EtOH giving 6g as pale yellow
1
solid: mp 188-189 °C; R
f
value 0.29 (4:1 hexanes/EtOAc); H
NMR (CDCl
3
) δ 10.15 (bs, 1H), 7.94 (d, J ) 7.8 Hz, 1H), 7.84
(
1
d, J ) 8.4 Hz, 1H), 7.29-7.42 (m, 2H), 7.22 (d, J ) 1.5 Hz,
1
3
H), 4.49 (q, J ) 7.2 Hz, 2H), 1.47 (t, J ) 7.2 Hz, 3H);
) δ 161.6, 144.5, 137.3, 128.4, 127.6, 125.7,
25.4, 124.9, 123.7, 121.3, 108.5, 61.0, 14.9.
Th ia n a p h th en o[3,2-b]p yr r ole-2-ca r boxylic Acid 7g. Es-
ter 6g (0.31 g, 1.17 mmol) and NaOH (1 g, 25 mmol) were
mixed in 20 mL of EtOH and 10 mL of H O. The mixture was
C
NMR (acetone-d
6
1
2
refluxed for 1 h, acidified to approximately pH 3 using 6 N
HCl, and filtered. Acid 7g was obtained after washing and
drying as pale yellow solid (0.24 g, 36% from the ben-
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. All chemicals were
obtained from commercial suppliers and used without further
purification. Toluene and THF were distilled from Na/benzo-
phenone prior to use. Dichloromethane used for the electro-
zothiophene-2-carbaldehyde starting material): mp 235 °C dec;
1
R
(
f
value 0.70 (1:1 acetone/MeOH); H NMR (DMSO-d
6
) δ 12.75
s, 1H), 12.73 (s, 1H), 8.13 (d, J ) 7.2 Hz, 1H), 7.91 (d, J ) 7.8
Hz, 1H), 7.34-7.42 (m, 2H), 7.12 (d, J ) 2.1 Hz, 1H).
chemical studies was distilled from CaH
ethane was distilled from CaH . All NMR spectra were
recorded on a Varian instrument at 300 MHz ( H), 75 MHz
2
, and 1,2-dichloro-
Th ia n a p h th en o[3,2-b]p yr r ole 5g. Acid 7g (0.24 g, 1.1
mmol) and NaOH (0.4 g, 10 mmol) were mixed in 10 mL of
ethylene glycol and heated to 140 °C under nitrogen. After 30
min, 100 mL of ice/water was added, and the mixture was
filtered. The yellow solid was washed thoroughly with water
and dried. It was then dissolved in approximately 10 mL of
2
1
1
3
11
19
(
C), 64 MHz ( B), or 282 MHz ( F). NMR chemical shifts
(
13) Li, F.; Yang, S. I.; Ciringh, Y.; Seth, J .; Martin, C. H.; Singh,
D. L.; Kim, D.; Birge, R. R.; Bocian, D. F.; Holten, D.; Lindsey, J . S. J .
Am. Chem. Soc. 1998, 120, 10001-17.
CH
2 2
Cl and passed through a short column of silica gel to give
(
14) Rathore, R.; Lindeman, S. V.; Kumar, A. S.; Kochi, J . K. J . Am.
5g as white solid after removal of solvent (0.175 g, 91%): mp
1
Chem. Soc. 1998, 120, 6931-9.
f
113-114 °C; R value 0.26 (4:1 hexanes/EtOAc); H NMR

Extended Conjugation by BODIPY Dyes
J . Org. Chem., Vol. 65, No. 10, 2000 2905
(
acetone-d
6
) δ 11.05 (bs, 1H), 7.95 (d, J ) 8.1 Hz, 1H), 7.88 (d,
) δ
matographed using 5% EtOAc in hexanes eluant. This gave
J ) 7.8 Hz, 1H), 7.40 (t, J ) 7.8 Hz, 1H), 7.24-7.29 (m, 2H),
4i as a copper colored solid after removal of solvents (0.61 g,
.56 (d of d, J ) 2.5 Hz, 2.0 Hz, 1H); 13C NMR (acetone-d
42.5, 133.5, 128.2, 124.7, 124.6, 124.3, 123.2, 122.8, 119.2,
02.3.
57%): mp >300 °C; R
NMR (CDCl
2H), 7.27 (d, J ) 8.4 Hz, 2H), 6.98 (d of d, J ) 8.7 Hz, 2.7 Hz,
value 0.22 (4:1 hexanes/EtOAc);
1
H
6
1
1
6
f
3
) δ 8.75 (d, J ) 8.7 Hz, 2H), 7.83 (d, J ) 8.4 Hz,
2
(
1
1
H), 6.81 (d, J ) 2.7 Hz, 2H), 6.47, (s, 2H), 3.88 (s, 6H), 2.88
BODIP Y 3g. Pyrrole 5g (0.18 g, 1.04 mmol), 4-iodobenzoyl
bs, 4H), 2.67 (bs, 4H); 13C NMR (CDCl
) δ 160.7, 151.9, 142.8,
37.3, 136.0, 135.4, 134.9, 132.2, 132.1, 130.1 (t, J ) 11 Hz),
chloride (0.14 g, 0.52 mmol), BF
and Et N (1.0 mL, 7.2 mmol) were mixed in toluene (100 mL)
and refluxed for 5 d. Evaporation gave a dark brown residue,
which was passed through a silica gel column, eluting with
3
etherate (1.2 mL, 9.8 mmol),
3
3
1
9
24.4, 121.5, 114.7, 112.4, 95.5, 55.3, 30.9, 22.3; F NMR
(
CDCl
J ) 34 Hz); FABHRMS calcd for C33
found 658.1142. Anal. Calcd for C33
3
) δ -137.5 (q, J ) 34 Hz); 11B NMR (CDCl
IN
3
) δ 9.29 (t,
658.1093,
H
26BF
2
2
O
2
3
:1 hexanes/CH
via flash chromatography, eluting with 4:1 hexanes/CH
to give 3g as dark blue solid (10 mg, 3%): mp >300 °C; R
2 2
Cl , to give a crude product. This was purified
H
26BF
2
IN
2
O
2
: C, 60.21; H,
2
Cl
2
,
3
.98; N, 4.26. Found: C, 60.49; H, 4.01; N, 4.35.
Keto-a ld eh yd e 12. 3,3-Dimethyl-1-indanone (0.31 g, 1.94
f
1
16
value 0.38 (4:1 hexanes/EtOAc); H NMR (CDCl
6.9 Hz, 2H), 7.89 (d, J ) 8.4 Hz, 2H), 7.66 (d of d, J ) 6.9
Hz, 2.0 Hz, 2H), 7.45-7.52 (m, 4H), 7.35 (d, J ) 8.4 Hz, 2H),
3
) δ 8.75 (d, J
)
mmol) was dissolved in 50 mL of dry THF and cooled to 0 °C.
LDA (2 M in THF/n-heptane, 1.2 mL, 2.4 mmol) was added
dropwise, and the resulting yellow solution was stirred at 0
1
3
6
1
1
.84 (s, 2H); C NMR (CDCl
33.4, 132.3, 129.6, 126.7 (t, J ) 5.2 Hz), 125.7, 125.6, 124.1,
19.8, 96.9; F NMR (CDCl ) δ -147.0 (q, J ) 30 Hz).
3
) δ 148.3, 140.4, 137.7, 137.5,
1
7
°C for 30 min. Glyoxal mono(dimethylhydrazone) (0.3 g, 3
mmol) was then added, and the resulting yellow solution was
1
9
3
8
stirred at room temperature for 3 h. Excess NH Cl
(aq)
was
BODIP Y 4h . Pyrrole 8h (0.26 g, 1.54 mmol) and 4-iodo-
4
added, and the mixture was extracted with Et
The combined ether solution was washed with water and brine
and dried over Na SO . Removal of solvent gave a yellow
viscous oil, which was purified on silica gel column, eluting
with 3% Et O in CH Cl . This gave 13 as a yellow solid
mixture of two isomers, 0.40 g, 85%).
Compound 13 (0.40 g, 1.65 mmol) was dissolved in 10 mL
of THF and was added dropwise to a solution of 2.0 g SnCl
2
O (3 × 150 mL).
benzoyl chloride (0.19 g, 0.71 mmol) were dissolved in 1,2-
dichloroethane (50 mL), and the solution was refluxed for 2 d
under nitrogen. A deep blue solution was formed. The solution
was cooled to 25 °C, triethylamine (0.5 mL, 3.6 mmol) was
2
4
added, and the mixture was stirred for 5 min. Then BF
3
2
2
2
(
etherate (0.75 mL, 6.1 mmol) was added, and the mixture was
refluxed for 30 min under nitrogen. The solvent was then
removed, and the residue was passed through a silica gel
2
‚
1
8
column using 3:1 hexanes/CH
2
Cl
2
. The crude compound was
2H
2
O
2
in 5 mL of concentrated HCl and 5 mL of H O with
then further purified via flash chromatography, slowly eluting
with 5% EtOAc in hexanes to give 4h as blue solid (0.26 g,
vigorous stirring at 0 °C. After the addition, the yellow solution
was further stirred for 10 min at room temperature. Water
5
7%): mp >300 °C; R
NMR (CDCl ) δ 8.80 (d, J ) 7.8 Hz, 2H), 7.85 (d, J ) 8.7 Hz,
H), 7.45 (d of t, J ) 7.8 Hz, 1.7 Hz, 2H), 7.35-7.25 (m, 6H),
.53 (s, 2H), 2.90 (t, J ) 6 Hz, 4H), 2.68 (t, J ) 6 Hz, 4H); 13
) δ 152.6, 140.5, 137.9, 137.3, 135.8, 134.5, 133.2,
32.1, 129.8, 128.4 (t, J ) 11 Hz), 128.3, 128.2, 127.5, 125.0,
f
value 0.41 (4:1 hexanes/EtOAc); 1
H
(100 mL) was added, and the mixture was extracted with CH
Cl
(3 × 50 mL). The combined organic layers were washed
with water and brine and dried over Na SO . Evaporation gave
a brown oil. This was then dissolved in small amount of CH
Cl and passed through a short column of silica gel. Removal
of solvent gave 12 as light yellow oil (0.26 g, 78%): H NMR
(CDCl ) δ (d, J ) 1.2 Hz, 1H), 7.67 (d, J ) 7.6 Hz, 1H), 7.61 (t,
2
-
3
2
2
6
2
4
C
2
-
NMR (CDCl
1
9
3
2
1
1
9
11
5.8, 30.5, 22.3; F NMR (CDCl
3
) δ -137.3 (q, J ) 34 Hz);
) δ 3.98 (t, J ) 34 Hz); FABHRMS calcd for C31 22BF
2 2
598.0891, found 598.0889. Anal. Calcd for C31 22BF IN :
B
3
(CDCl
3
H
2
-
J ) 7.4 Hz, 1H), 7.50 (d, J ) 7.8 Hz, 1H), 7.34 (t, J ) 7.5 Hz,
IN
2
H
1H), 2.97-3.09 (m, 2H), 2.52-2.65 (m, 1H), 1.50 (s, 3H), 1.09
(s, 3H); 13C NMR (CDCl
) δ 205.3, 200.6, 162.4, 135.0, 133.6,
C, 62.24; H, 3.71; N, 4.88. Found: C, 62.35; H, 3.89; N, 4.73.
-Meth oxy-4,5-d ih yd r o-1H-ben z[g]in d ole 8i. Lithium
hydroxide monohydrate (1.0 g, 23.8 mmol) and 6-methoxyte-
3
1
27.4, 123.3, 123.2, 53.9, 41.3, 40.3.
7
E t h yl 3-Met h yl-4,5-d ih yd r o-1H -b en z[g]in d ole-2-ca r -
1
5
tralone oxime (1.0 g, 5.2 mmol) were mixed in 25 mL of
DMSO. Acetylene was bubbled into the mixture. This was then
heated to 140 °C with stirring under nitrogen for 5 h. Ice/water
boxyla te 16. A solution of 1-tetralone (17.5 g, 0.12 mole) and
NaOAc (75 g, 0.9 mole) in propionic acid (300 mL) was heated
to reflux. Oxime 1510 (16 g, 0.1 mole), dissolved in 200 mL of
propionic acid, was added dropwise, while simultaneously
adding small portions of activated Zn dust (40 g, 0.6 mol). After
the additions, the mixture was refluxed for 1 h and cooled to
about 70 °C. It was then poured to 4 L of ice/water and was
allowed to stand overnight. A yellow solid precipitated and was
isolated by filtration. The solid was thoroughly washed with
water until the filtrates were neutral. The dried solid was then
washed several times with hexanes to remove unreacted
1-tetralone and other nonpolar impurities. The solid was then
recrystallized from EtOH several times to give 16 as white
(
(
250 mL) was added, and the product was extracted into Et
3 × 100 mL). The combined ether solution was washed with
SO . Evaporation gave a
light brown residue. This was flash chromatographed 2:1
hexanes/CH Cl to give the vinyl pyrrole side product 9i (0.10
g, 8%). Further eluting with 1:1 hexanes/CH Cl gave the
desired pyrrole 8i as a sticky white solid (0.67 g, 65%):
NMR (CDCl
2
O
water and brine and dried over Na
2
4
2
2
2
2
1
H
3
) δ 8.23 (bs, 1H), 7.05 (d, J ) 8.1 Hz, 1H), 6.90
(
(
(
1
1
d, J ) 2.7 Hz, 1H), 6.75 (d of d, J ) 8.1 Hz, 2.7 Hz, 1H), 6.72
t, J ) 2.4 Hz, 1H), 6.20 (t, J ) 2.4 Hz, 1H), 3.88 (s, 3H), 3.00
t, J ) 7 Hz, 2H), 2.82 (t, J ) 7 Hz, 2H); 13C NMR (CDCl
) δ
solid (2.3 g, 11.5% based on recovered 1-tetralone): mp 136-
3
1
57.2, 136.7, 127.5, 122.7, 119.2, 118.2, 117.3, 114.6, 111.1,
07.7, 55.2, 30.4, 21.7. This compound was not particularly
138 °C; R
f
value 0.40 (4:1 hexanes/EtOAc); H NMR (CDCl
3
) δ
9.28 (bs, 1H), 7.41 (d, J ) 7.8 Hz, 1H), 7.15-7.30 (m, 3H),
stable with respect to long-term storage at room temperature.
4.40 (q, J ) 7.2 Hz, 2H), 2.98 (t, J ) 7.5 Hz, 2H), 2.69 (t, J )
1
3
7
(
1
.5 Hz, 2H), 2.34 (s, 3H), 1.43 (t, J ) 7.2 Hz, 3H); C NMR
BODIP Y 4i. Pyrrole 8i (0.65 g, 3.27 mmol) and 4-iodoben-
zoyl chloride (0.44 g, 1.72 mmol) were dissolved in 1,2-
dichloroethane (50 mL), and the solution was refluxed for 2 d
under nitrogen. A deep greenish blue solution resulted. The
solution was cooled to room temperature, triethylamine (1.0
mL, 7.2 mmol) was added, the solution was stirred for 5 min
3
CDCl ) δ 161.5, 136.3, 131.2, 128.5, 128.0, 126.8, 126.6, 125.5,
21.6, 120.0, 119.4, 60.0, 29.6, 19.5, 14.6, 10.5.
3-Met h yl-4,5-d ih yd r o-1H -b en z[g]in d ole-2-ca r b oxylic
Acid 17. Ester 16 (500 mg, 1.96 mmol) and NaOH (0.5 g, 12.5
mmol) were mixed in 100 mL of EtOH and 50 mL of H O and
refluxed for 2 h. The solution was cooled and acidified by 6 N
2
3
at room temperature, BF etherate (1.2 mL, 9.8 mmol) was
added, and the mixture was refluxed for 30 min under
nitrogen. The solvent was then removed, and the residue was
(
16) Pearson, B. D.; Ayer, R. P.; Cromwell, N. H. J . Org. Chem. 1962,
passed down a silica gel column (1:1 hexanes/CH
2 2
Cl ) to
2
7, 3038-44.
remove polar impurities. The residue was then flash chro-
(17) Severin, T.; Supp, W.; Manninger, G. Chem. Ber. 1979, 112,
013-22.
3
(
18) Severin, T.; Adam, R.; Lerche, H. Chem. Ber. 1975, 108, 1756-
(15) Sawa, Y.; Kato, T.; Masuda, T.; Hori, M.; Fujimura, H. Chem.
67.
(19) Magde, D.; Brannon, J . H. J . Phys. Chem. 1979, 83, 696-9.
Pharm. Bull. 1975, 23, 1917-27.

2
906 J . Org. Chem., Vol. 65, No. 10, 2000
Chen et al.
HCl. The resulting precipitate was filtered and dried to give
added, and the solution was stirred at room temperature and
monitored by TLC. After 5 min, TLC showed that the reaction
was completed. Solvent was removed, and the residue was
purified by passing a short column of silica gel using 2:1
1
1
(
1
7
1
1
7 as gray solid (440 mg, 99%): mp 129-134 °C; H NMR
acetone-d ) δ 11.0 (bs, 1H), 9.75 (bs, 1H), 7.91 (d, J ) 7.8 Hz,
H), 7.15-7.30 (m, 3H), 2.93 (t, J ) 7.5 Hz, 2H), 2.66 (t, J )
.5 Hz, 2H), 2.38 (s, 3H); 13C NMR (acetone-d
) δ 165.0, 137.1,
33.0, 129.1, 129.0, 127.6, 126.9, 122.2, 122.1, 119.8, 30.1, 20.1,
0.6.
6
6
hexanes/CH
2
Cl
2
as eluent. This gave 18 as a dark blue solid
value 0.35 (4:1 hexanes/
) δ 9.41 (d, J ) 8.1 Hz, 1H), 9.13 (d,
(18.5 mg, 93%): mp >300 °C; R
f
1
EtOAc); H NMR (CDCl
3
3
-Meth yl-4,5-d ih yd r o-1H-ben z[g]in d ole 14j. Acid 17 (440
mg, 1.94 mmol) and NaOH (0.5 g, 12.5 mmol) were mixed in
0 mL of ethylene glycol and heated to 140 °C under nitrogen.
After 20 min, ice/water (100 mL) was added, and the precipi-
J ) 8.1 Hz, 1H), 7.93 (d, J ) 8.1 Hz, 2H), 7.83 (s, J ) 7.8 Hz,
1H), 7.70 (t, J ) 7.8 Hz, 1H), 7.39-7.61 (m, 6H), 7.38 (d, J )
8.1 Hz, 2H), 6.99 (s, 1H), 6.76 (s, 1H), 3.00 (t, J ) 6.7 Hz, 2H),
2
2.79 (t, J ) 6.7 Hz, 2H); 13C NMR (CDCl
) δ 158.4, 145.9, 142.0,
3
tate was filtered and washed with H
dissolved in small amount of CH Cl
2
O. This solid was then
and passed through a
140.5, 138.7, 137.5, 137.0, 136.5, 135.1, 134.3, 132.3, 131.7,
2
2
130.2 (t, J ) 11.5 Hz), 129.7, 129.0, 128.6, 128.0, 127.5, 127.4,
short column of silica gel, eluting with 1:1 CH
give 14j as white solid (270 mg, 76%). The H NMR for this
2
Cl
2
/hexanes to
127.3, 127.1(t, J ) 11.3 Hz), 126.7, 124.7, 123.9, 121.5, 120.6,
1
19
96.4, 30.4, 22.6; F NMR (CDCl
3
) δ -138.8 (q, J ) 33.5 Hz);
IN 596.0742, found
material correlates well with that published for the same
HRFABMASS: calcd for
596.0732.
C
31
H
20BF
2
2
9
material prepared via a different route.
BODIP Y 4j. Pyrrole 14j (260 mg, 1.42 mmol) and 4-iodo-
benzoyl chloride (190 mg, 0.71 mmol) were dissolved in 250
mL of 1,2-dichloroethane and refluxed under nitrogen for 3 d.
The mixture was cooled to room temperature, triethylamine
Op tica l Sp ectr oscop y. All experiments were conducted in
chloroform solutions (4 µM for absorption spectra, 1 µM for
emission spectra). Quantum yields were determined as de-
scribed earlier.3
(0.5 mL, 3.6 mmol) was added, the mixture was stirred at room
Cyclic Volta m m etr y. These experiments were conducted
temperature for 5 min, BF etherate (1 mL, 8.1 mmol) was
3
in anhydrous CH
2 2
Cl at room temperature. The potentials are
added, and the mixture was refluxed for 30 min. Evaporation
gave a dark brown residue that was passed though a short
reported vs ferrocene as internal standard using a scan rate
-
1
of 100 mV s , glassy carbon working electrode, Ag/AgNO
3
silica gel column (30% CH
flash chromatographed eluting with 3% EtOAc in hexanes.
2
Cl
2
in hexanes, eluant) and then
reference electrode, platinum counter-electrode, and a sup-
porting electrolyte 0.1 M tetrabutylammonium hexafluoro-
phosphate.
This gave 4j as copper-colored solid (210 mg, 47%): mp >300
1
°C; R
f
value 0.39 (4:1 hexanes/EtOAc); H NMR (CDCl
3
) δ 8.83
(
d, J ) 8.1 Hz, 2H), 7.92 (d, J ) 8.1 Hz, 2H), 7.45 (t, J ) 8.1
Ack n ow led gm en t. Financial support was provided
via NIH grant HG01745 and The Robert A. Welch
Foundation. A.B. thanks the Deutsche Forschungsge-
meinschaft for a fellowship. Ben Lane and Long Thai
Hz, 2H), 7.33 (t, J ) 8.1 Hz, 2H), 7.28 (d, J ) 8.1 Hz, 2H),
7
7
1
.17 (d, J ) 8.1 Hz, 2H), 2.91 (t, J ) 7.1 Hz, 4H), 2.56 (t, J )
13
.1 Hz, 4H), 1.42 (s, 6H); C NMR (CDCl
3
) δ 151.0, 140.7,
38.3, 138.1, 135.8, 135.7, 133.6, 132.2, 130.7, 129.5, 128.6 (t,
11
are thanked for assistance in obtaining B NMR.
1
9
J ) 12 Hz), 128.3, 128.1, 127.3, 94.7, 30.5, 20.4, 12.4; F NMR
1
1
(
CDCl
3
) δ -135.0 (q, J ) 34 Hz); B NMR (CDCl
I 626.1202, found
2 2
26.1314. Anal. Calcd for C31H26BF N I: C, 63.29; H, 4.18; N,
3
) δ 1.42 (t,
Su p p or tin g In for m a tion Ava ila ble: Copies of H and 13C
1
J ) 35 Hz); FABHRMS calcd for C31
6
4
H
26BF
2
N
2
19
11
NMR spectra for new compounds; F and B NMR spectra
for new dyes. This material is available free of charge via the
Internet at http://pubs.acs.org.
.47. Found: C, 63.53; H, 4.25; N, 4.51.
Com p ou n d 18. BODIPY 4h (20 mg, 0.033 mmol) was
dissolved in 10 mL of CH
2
Cl
2
. DDQ (23 mg, 0.1 mmol) was
J O991927O